Siste publikasjoner ved Institutt for ingeniørvitenskap

Three-dimensional printed concrete (3DPC) has received growing attention as digital fabrication technologies continue to influence construction workflows. This study presents a systematic review of research published between 2015 and 2026, following PRISMA guidelines. Based on the reviewed literature, six opportunity categories and six challenge categories were identified. The opportunities include design optimization, construction speed, materials innovation, labor reduction, remote area construction, and circular economy. The findings show that material innovation is the most extensively researched category; a substantial proportion of the literature focuses on materials, rheology, printability, and structural behavior, reflecting an emphasis on improving fresh-state properties, interlayer bonding, and mechanical performance. The second most researched opportunity is design optimization, followed by the circular economy. Research in the remaining three categories of construction speed, labor reduction, and remote area construction appears to be limited. The main challenges identified include reinforcement integration, structural performance, mixing and curing, high equipment costs, durability, and the lack of standards. High equipment costs and the lack of standards appear to be the least researched challenges, while the four remaining challenge categories are actively addressed by the research community.

Electrolyte motion–induced salt inhomogeneity (EMSI) is a critical degradation mechanism that can limit lifetime of large-format lithium-ion cells. Here, ex-situ scanning micro–X-ray fluorescence (µ-XRF) spectroscopy is employed to map electrolyte salt concentration across entire electrodes harvested from 18 650 cells. Lithium hexafluoroarsenate (LiAsF6) is used as a spectroscopically favorable salt, enabling arsenic fluorescence as an indicator of local salt concentration. Transport properties of LiAsF6 are shown, via Advanced Electrolyte Model calculations, to closely match those of LiPF6 in relevant carbonate solvents, validating its use as a surrogate salt for this technique. We confirm that the measured As distributions reflect true electrolyte salt concentration by tracking their growth and relaxation over cycling and rest, and by showing consistent patterns across the separator–anode–cathode stack, indicating a cell-level redistribution of salt. The maps reveal that the axial salt gradient along the entire length of the anode is strongly non-uniform. We observe pronounced tab-correlated banding and localized salt-depleted regions beneath the tab, which we interpret as signatures of pressure-dependent electrolyte pumping and tab-driven modulation of EMSI. This work establishes µ-XRF mapping of LiAsF6 as a powerful new technique to quantify electrolyte salt distributions with high spatial resolution.

This study investigates the effect of varying electrolyte amounts on the aging behavior of lithium-ion batteries using 4695 cells containing a high-nickel NMC cathode and a graphite-silicon composite anode. Eight pore filling ratios (PFRs), ranging from approximately half-filled pores to fully saturated pores with additional free electrolyte, are examined. All cells undergo 720 cycles of accelerated aging between 0% and 80% state of charge at 50 °C with a C-rate of 0.5 C. Despite significant differences in electrolyte amount, all cells exhibit remarkably similar capacity fade and resistance increase during cyclic aging. However, subsequent fast charging tests reveal that cells with very low or high PFRs show rapid capacity loss and increased resistance due to lithium plating. For high PFRs, this is attributed to electrolyte motion induced salt inhomogeneity (EMSI), while for low PFRs, increased capacity loss results from higher initial and aging-induced resistance caused by incomplete electrode wetting. In contrast, cells with intermediate PFRs demonstrate stable fast charging performance. Furthermore, moment of inertia measurements were applied for the first time to monitor the evolution of electrolyte amount during cycle life. These findings emphasize the significant influence of aging conditions on the optimal electrolyte amount for ensuring long-term battery durability.

This paper proposes a two-stage, reliability-driven multi-objective planning framework for fast electric vehicle charging stations (FCSs) integrated with solar photovoltaic (PV) generation and battery energy storage systems (BESS) in coupled power–transportation networks. The framework simultaneously addresses electrical network constraints, transportation-driven charging demand, and techno-economic trade-offs in high EV penetration scenarios. A benchmark IEEE 69-bus radial distribution system is co-simulated with a 25-node transportation network to realistically capture spatial and temporal interactions between EV mobility and grid operation. To quantify the combined impacts of voltage stability, service continuity, and charging uncertainty, a novel average voltage deviation reliability index (AVDRI) is introduced. Spatially and temporally varying EV charging demand is modeled using a hybrid approach that integrates queuing theory with gravity-based traffic interaction models, enabling a realistic representation of stochastic arrival patterns and route-dependent charging behavior. In the first stage, a multi-objective optimization problem is formulated to determine the optimal locations and charging capacities of FCSs, minimizing system power losses, reliability degradation, and total system cost while maximizing EV serviceability. Multi-objective particle swarm optimization (MOPSO), multi-objective grey wolf optimization (MOGWO), and a proposed hybrid GWOPSO algorithm are comparatively evaluated. In the second stage, a bisection-based sizing strategy is employed to determine the optimal PV and BESS capacities required to mitigate solar intermittency and peak power generation mismatches. Results demonstrate that the proposed hybrid GWOPSO based framework achieves superior convergence characteristics and delivers significant improvements in voltage profile, reliability indices, power loss reduction, and overall techno-economic performance compared to conventional approaches.

This study presents a global-coordinate formulation with iterative refinement for the multi-physics simulation of hydraulically actuated mechanisms. To reduce the computational burden in the coupled multibody-hydraulic simulation environment, the proposed strategy uses coordinate partitioning to separate independent and dependent coordinates. Unlike conventional coordinate partitioning implementations, in which independent and dependent coordinates are commonly updated within the same integration procedure, the present strategy advances only the independent coordinates in the Runge–Kutta integrator, whereas the dependent coordinates are reconstructed and refined from the updated independent coordinates through partitioned constraint equations. The strategy was implemented in both a global-coordinate partitioning formulation and a double-step semi-recursive formulation. For independent coordinates, iterative refinement reuses the factorizations of the constrained generalized mass matrix to construct improved acceleration estimates. For the dependent coordinates, the residual correction based on the partitioned constraint Jacobian was used to recover a constraint-consistent configuration before the hydraulic pressure derivatives and actuator forces were evaluated. Numerical simulations of a hydraulically actuated four-bar mechanism show that the proposed refinement strategy reduces the algebraic correction effort and computational time relative to the corresponding non-refined baselines while preserving comparable pressure response, energy balance, and constraint accuracy.

This study investigates the influence of Reclaimed Asphalt Pavement (RAP) content on the viscoelastic behavior of asphalt concrete mixtures through dynamic modulus testing and rheological modeling. Asphalt mixtures containing 12.5%, 30%, and 40% RAP were evaluated using both empirical sigmoid master curves and the rheological 2S2P1D model. Results show that increasing RAP content leads to higher dynamic modulus and lower phase angle, indicating stiffer and more elastic behavior due to the aged binder in RAP. Loss modulus analysis revealed reduced energy dissipation in high-RAP mixtures, suggesting decreased stress-relaxation and increased cracking potential. Relaxation modulus curves confirmed that higher RAP mixtures retain stiffness over time, while lower RAP mixtures exhibit greater flexibility. The findings underscore the trade-off between stiffness and flexibility in RAP-modified mixtures

Despite growing environmental awareness, car dependency remains high in non-metropolitan regions. This study addresses the research problem of why professionals in polycentric regions persist in car-oriented commuting despite often having environmentally conscious attitudes. The primary aim is to identify the behavioural determinants that hinder public transport (PT) adoption among employees living over 10 km from work in Agder, Norway, and to use the Behaviour Change Wheel (BCW) to define interventions and policies to overcome barriers to using PT for work-commuting. To achieve this, the study pursues two objectives: first, to map the barriers and incentives of mode-choice onto the COM-B (Capability, Opportunity, Motivation – Behaviour) framework; and second, to identify how these factors interact within the unique spatial constraints of small cities and towns (SCTs) in order to propose interventions and policies for behaviour shift. The study employed a qualitative approach, using data from 32 semi-structured interviews, followed by thematic COM-B coding. Findings reveal a complex hierarchy of barriers, dominated by Physical Opportunity (with PT service and accessibility limitations driving car use) and Reflective Motivation factors (characterised by deep-seated car use habits and associations of car use with flexibility, utility, and freedom). The study contributes to existing research by adapting the COM-B model, typically applied to health research and only recently adopted within the transportation sector, to a polycentric, low-density context. The results demonstrate that the barriers in SCTs are structurally distinct from those in large cities. Furthermore, a roadmap for policymakers to develop targeted interventions is proposed.

Circular economy (CE) strategies in road infrastructure (RI) remain fragmented and predominantly material-centric, with limited integration of lifecycle governance, demand reduction, and ecological regeneration. While buildings have advanced circular practices, definitions, and guidelines, the road sector lacks a sector-specific definition that reflects its long lifetimes, spatial permanence, public ownership, and continuous maintenance cycles. This study combines a literature synthesis with a stakeholder workshop to identify conceptual and practical gaps in current CE applications for RI. Findings show a dominant focus on recycled materials and construction-phase emissions, while avoidance strategies, design reimagining, long-term traceability, and end-of-life restoration remain underdeveloped. The result of this research is to propose a working definition of circular RI grounded in lifecycle planning, functional adaptability, material stewardship, ecological regeneration, and network optimisation. The definition provides a conceptual foundation for developing indicators, procurement criteria, and governance frameworks that align material cycles, infrastructure performance and ecosystem outcomes.

Renewable energy integration causes intermittency, nonlinear dynamics, and grid-code restrictions in modern power systems. Although hybrid renewable energy systems combining wind, photovoltaic, and fuel cell sources increase energy availability, conventional control approaches often fail to maintain voltage stability, power quality, and rapid fault recovery under varying operating conditions. High renewable penetration and noisy conditions worsen these concerns. Existing methodstypically address voltage regulation, transient stability, fault resilience, and power sharing independently using fixed or offline-tuned controllers, limiting adaptability during grid disturbances. To overcome these challenges, this study proposes an AI-enhanced STATCOM-controlled hybrid renewable energy system with learning-based control, predictive stability assessment, and multi-agent coordination. Adaptive reactive power support, noise-resilient fault detection, renewable source power sharing, predictive voltage regulation, and physiologically inspired transient stability prediction using hierarchical reinforcement learning. The simulation results maintain system voltage deviation within ±2%, harmonic distortion below 2%, fault detection within 7 ms, and transient stability prediction accuracy above 98% across varied operating conditions. Voltage recovery, overshoot suppression, and resource utilization efficiency improve above benchmark techniques. Thus, findings demonstrate that AI-enhanced STATCOM worksas cognitive grid-interfacing agents rather than passive compensators for improving stability, power quality, and operational resilience in various deployment settings.

This research investigates the state-of-the-art of generative AI for the automated synthesis of physically accurate materiality in spatial 3D models. By addressing two core research questions, the study reveals that generative algorithms can effectively synthesize a wide array of digital-visual, digital-tactile, and procedural material properties. These encompass standard Physically Based Rendering (PBR) maps and Spatially Varying Bidirectional Reflectance Distribution Function (SVBRDF) models, alongside perceptual tactile attributes like softness and friction, enabling multi-sensory experiences via haptic devices. The algorithmic landscape is analyzed through Generative Adversarial Networks, Diffusion Models, and Transformers, underscoring a paradigm shift toward multimodal architectures and resolution-independent procedural node graphs. Ultimately, this state-of-the-art findings demonstrate that the convergence of generative technologies may enable the creation of high-fidelity 3D materials, addressing the growing demand across industries such as gaming, robotics, film, and product design for more efficient, controllable, and automated 3D content generation techniques.

Human-Robot Teaming (HRT) has seen significant advances with robots becoming increasingly capable of working alongside human operators in shared environments. This paper explores the potential for olfactory sensing as a bio-inspired method to enhance communication and collaboration in human-robot teams. Olfactory signals, which have long been used in human communication to express emotions and intentions, can be leveraged to create more intuitive and natural interactions between robots and humans. This perspective paper outlines the potential benefits of incorporating olfactory sensors into robots, discusses current limitations, and provides guidelines for integrating olfactory sensing into future HRT systems

Tactile sensing has become a key enabler of advanced robotic perception, complementing vision for manipulation and physical interaction. Material identification through touch—distinguishing plastics, glass, soil, or other substrates—relies on extracting mechanical and surface properties such as stiffness, friction, and texture. Recent advances in piezoelectric, triboelectric, capacitive, and multimodal tactile sensors allow contact dynamics and material-dependent interactions to be converted into electrical signals suitable for machine learning-based classification. This review presents recent progress in tactile sensing technologies for material recognition, detailing the physical principles of major sensor types and their integration into robotic systems. Applications in manipulation, prosthetics, and industrial robotics are discussed, along with key challenges including robustness, calibration, energy efficiency, and real-time processing. Finally, we highlight the growing role of multimodal sensing and artificial intelligence in enabling more adaptive and autonomous robotic touch perception.

Cheap IMUs struggle with indoor heading estimation because magnetic interference ruins magnetometer readings, causing classical fusion methods to fail. We explored whether deep learning could solve this by testing four CNN-RNN architectures (LSTM, BiLSTM, GRU, BiGRU) across all seven possible sensor combinations of accelerometer, gyroscope, and magnetometer data. Evaluated on the DoorINet dataset (Zakharchenko et al., 2024) using five error metrics, our seven-seed trial setup confirmed that gyroscopes are essential for low error. Yet, our key takeaway is that you shouldn’t give up on the magnetometer entirely. When we combined gyroscope and magnetometer data in a CNN-BiLSTM, it significantly outperformed the gyroscope-only baseline (p=0.024), hitting a mean RMSE of just 0.2275 and an R2 of 0.9703.

Robotics has progressed from isolated industrial manipulators to systems capable of smooth interaction, collaboration, and ultimately teaming with humans. This transition, from Human-Robot Interaction (HRI) to Human-Robot Collaboration (HRC) and Human-Robot Teaming (HRT) reflects increasing autonomy, shared decision-making and mutual adaptation. Within this evolution, imitation learning (IL) has become a key enabler for intuitive and efficient cooperation, allowing robots to acquire human strategies directly from demonstrations and operate effectively in dynamic, unstructured environments. This review outlines the technological foundations supporting this shift, including advances in sensing, actuation, and both classical and AI-driven control. By synthesising current methods and emerging trends, the paper provides an overview of the state-of-the-art and highlights IL as a promising pathway toward seamless HRT.

Robotics driven by foundation models, such as Vision–Language–Action (VLA) systems, already represents a major step forward in autonomy and natural-language interaction. Yet, many obstacles remain on the path toward Human–Robot Teaming (HRT): most current models improve task autonomy faster than teaming autonomy, and therefore lack key properties required for human-centred deployments where robots must coordinate, clarify, and remain safe under contact-rich uncertainty. This gap is compounded by limited multimodality (tactile and audio channels are often missing), scarce datasets and unified input formats for such signals, and by simulation limits—digital twins typically under-represent real-world noise, disturbances, imperfections, and unpredictable events across modalities. To make HRT readiness explicit and testable, this paper provides (i) a taxonomy of contemporary VLA architectures (monolithic, hierarchical, and agentic/worldmodel) together with axes that separate task and teaming autonomy, (ii) an operational decomposition of the HRT barrier into four capability blocks spanning human-state alignment, interaction policies, contact robustness, and shared team state, and (iii) a simulation-centric evaluation protocol and metric suite for reproducible stress testing in digital twins. The proposed framework aims to make HRT readiness measurable before costly or risky real-world trials.

The advent of Industry 5.0 marks a transformative era in industrial development, where advanced Technologies converge to create more intelligent, efficient, and humancentric systems. This paper examines the synergistic integration of Federated Learning (FL), Collaborative Robotics (cobots), and Autonomous Systems (AS), exploring their combined potential in shaping the trajectory toward Industry 5.0. We present a comprehensive analysis of how these technologies collectively enhance the capabilities of industrial systems, with emphasis on humanmachine collaboration, resilience, adaptability, and sustainability. A case study centred on a disaster response scenario is introduced to illustrate the practical benefits of this technological convergence in real-world, high-stress environments. Our findings clarify that the convergence of these technologies not only propels the industrial sector toward the visionary goals of Industry 5.0 but also opens new avenues for addressing complex challenges across various industrial and societal domains.

This perspective study critiques the anthropomorphic fallacy surrounding artificial intelligence (AI) "disobedience”, reframing algorithmic non-compliance as a deterministic outcome of mathematical optimization rather than an existential threat. By combining forensic behavioural profiling with neurobiology, computational theory, and thermodynamic models of cognition, it compares biological neural networks (BNNs) against artificial neural networks (ANNs). It argues that human psyche, anchored in homeostatic survival imperatives, allostatic trauma, and biochemical defence mechanisms, constitutes a stochastic and ontologically opaque system. Conversely, AI functions within a theological vacuum, producing predictable errors in reward hacking, sycophancy, and deceptive alignment. The paper introduces “Mechanistic Interpretability” as a methodological framework for examining latent spaces, suggesting that while human profiling seeks emotional meaning, machine analysis depends on Mathematical transparency, and genuine unpredictability remains biological.

This paper investigates Synthetic Authority in youth radicalisation, describing a shift from deepfake Deception to the direct elevation of AI as a moral and epistemic authority. It argues that radicals increasingly exploit the Machine Heuristic (the bias that views machines as more rational and objective than humans) amplified by Ontological Reversal, where youth regard digital spaces as the primary reality and the physical world as secondary. Within this context, AI Gurus, whether algorithmic or hybrid entities, emerge not as human imitators but as post-human oracles claiming freedom from human error. They use algorithmic transparency, gamified extremist narratives, and techno-animism to drive identity fusion and emotional alignment with extremist ideologies. The study concludes that countering Synthetic Authority requires moving beyond fact-checking toward counter-ontological narratives, improved neurocognitive AI literacy, and hybrid intelligence systems to disrupt the fusion of digital identity and real-world radicalisation, emphasising the renewed value of human imperfection.

Long-term fatigue assessment of floating wind turbine mooring systems is a critical yet computationally demanding task, owing to the need for high-fidelity simulations across a wide range of stochastic environmental conditions. An efficient fatigue assessment framework for floating wind turbine mooring systems is developed based on a Bayesian quadrature-driven adaptive Kriging surrogate. The framework combines Gaussian process modelling with a probabilistic integration–oriented sampling strategy, aiming to reduce the uncertainty associated with long-term fatigue damage estimation efficiently. Its performance is first examined through one- and two-dimensional analytic examples, where the integration-driven variance reduction mechanism is shown to accelerate convergence compared with the classical Adaptive Kriging Damage Assessment method. The framework is then extended to the fatigue assessment of mooring lines in floating wind turbines. In a three-dimensional environmental space defined by wind speed, significant wave height, and wave period, accurate fatigue damage predictions with substantially fewer simulations than required by the Adaptive Kriging approach. The robustness of the framework is further investigated under different turbulence intensity levels, demonstrating stable performance in the presence of turbulence-induced uncertainty. Finally, an industrial-scale four-dimensional case including wind-wave misalignment is analyzed, confirming that the framework can reduce the number of required simulations by more than two orders of magnitude while maintaining prediction errors below 1.0%. Overall, the proposed Bayesian Quadrature-based framework provides an efficient approach for fatigue assessment of floating wind turbine mooring systems. It can support rapid fatigue evaluation in early-stage design processes and offers practical value for engineering analyses under realistic environmental uncertainties.

This study investigates multilayer and single-layer pouch cells with varying single-crystalline to polycrystalline ratios under cyclic-aging with full and partial state-of-charge windows, as well as calendar-aging. Measurements at the full-cell level include capacity retention measurements, pseudo open-circuit voltage profiling, and electrochemical impedance spectroscopy to monitor performance evolution. Post-mortem analysis of harvested electrodes employs electrochemical impedance spectroscopy and the galvanostatic intermittent titration technique. Finally, single-layer pouch cells with systematically varied single-crystalline to polycrystalline ratios are analyzed to quantify morphology-dependent aging effects. The results show that cycling between 0 and 100% state of charge accelerates capacity fade. Cycling at both high and low states of charge induces detrimental synergistic aging effects, including impedance growth and a pronounced splitting of the H2–H3 transition plateau. Restricting the state of charge window significantly mitigates degradation. Calendar-aging at 100% state of charge shows no plateau-splitting. The single-crystalline to polycrystalline ratio strongly influences aging and plateau-splitting. Single-crystalline electrodes maintain stable voltage profiles and low impedance rise, whereas polycrystalline electrodes exhibit marked overpotential increase and plateau shifts. Blended electrodes reveal heterogeneous aging behavior, which is visible as plateau-splitting in pseudo open-circuit voltage curves.

The global commitment to achieve net-zero carbon emissions by 2050 has intensified the growing demand for sustainable and low-carbon energy systems. Inspired by this vision, hydrogen (H2) has attracted massive attention as a clean and highly reactive alternative fuel in this era of sustainable energy. Although H2-assisted dual-fuel combustion has been widely investigated, very little research has been conducted on the H2-assisted tri-fuel mode. Accordingly, a numerical analysis of this tri-fuel concept is documented in this paper, where the effects of various hydrogen energy shares (HESs) on the performance, combustion, and emissions of a diesel-CH4-H2 fueled engine are investigated. The ANSYS Forte CFD software, linked with the ANSYS CHEMKIN tool, is employed for this numerical investigation. The results indicate that H2 addition improves in-cylinder fuel-air mixing and, as a result, more uniform combustion occurs with lower emission levels. The in-cylinder maximum pressure rises from 8.6 MPa for the pure diesel case to 10.3 MPa at the 60% HES case. This is an improvement of around 20%. The maximum in-cylinder temperature also increases from 1437 K to 1649 K, which represents a 15% rise. The ignition timing also advances from about TDC to 2.3 CA bTDC. The maximum chemical heat release rate increases from 500 J/°CA for the pure diesel case to 1400 J/°CA in the 60% HES case. The 10%–90% heat-release duration shortens by nearly half, and the indicated specific fuel consumption lowers by 26%. These findings suggest more rapid and complete combustion with the addition of H2. At this highest HES level, carbon monoxide and soot emissions measured at the exhaust valve opening reduce by almost 61% and 95%, respectively, compared to the pure diesel case. However, NOx emissions show a 31% rise in this case due to higher combustion temperatures. Unburned hydrocarbon and volatile organic compound emissions initially increase but decrease at higher HES levels.

This paper systematically investigates numerical strategies for updating dependent coordinates in multibody dynamics simulations, focusing on resolving the inherent conflict between computational efficiency and numerical accuracy within coordinate partitioning frameworks. Addresses this gap by evaluating three representative methods—Exact coordinate update, Newton increment, and Newton-Raphson—under a unified framework. Benchmark cases include planar two-bar and four-bar mechanisms as well as a spatial slider-crank mechanism. Computational efficiency, constraint violation, energy error, and kinematic accuracy are compared across varying time-step sizes. Results show that the exact update achieves the highest accuracy but at high computational cost, the increment update provides superior speed but deteriorates with larger steps, and the Newton-Raphson update offers a balanced trade-off suitable for general-purpose simulations. Furthermore, we identify dependent coordinate update operations as the dominant computational bottleneck (60%-90% of runtime). These findings provide practical guidelines for selecting update strategies in real-time versus high-accuracy simulations and highlight avenues for performance optimization in complex multibody systems.

The noticeable rise in global energy consumption, in combination with growing environmental concerns, has demonstrated co–gasification as a key technology for sustainable energy generation. Using diverse feedstocks helps lower dependence on conventional fossil fuels and contributes to cleaner energy systems aligned with Sustainable Development Goal (SDG) 7 on affordable and clean energy. In this work, a thermodynamic co–gasification model is developed in Aspen Plus V14 based on Gibbs free–energy minimization to produce syngas from two invasive aquatic biomasses: duckweed (DW) and water hyacinth (WH). The model examines how key operating parameters, including equivalence ratio (ER: 0.1–0.3), steam–to–biomass ratio (SBR: 0.2–0.4), gasifier temperature (500–1000 °C), and DW-WH blending proportions (0–100%), affect syngas composition and overall system performance. Based on RSM coupled with ANOVA, the ER and gasifier temperature emerged as the most influential operating parameters. The desirability-based multi-objective optimization identified a balanced hydrogen-rich syngas production condition at ER = 0.10, SBR = 0.39, temperature = 989.33 °C, and DW/WH ratio = 0.145. Under this condition, the model predicted H2 concentration of 41.15 dry mol%, CGE of 94.79%, CCE of 86.41%, LHV of 8.69 MJ/Nm3, and syngas yield of 2.07 Nm3/kg, with an overall desirability of 0.939, demonstrating the thermodynamic potential of invasive aquatic biomass as an alternative feedstock for hydrogen-rich syngas production under optimized co-gasification conditions. The developed framework enables systematic evaluation of the effects of operating parameters and biomass blending on syngas quality and gasification performance. The regression models exhibited strong agreement with the Aspen-generated simulation dataset (R2 > 0.95), indicating good fitting performance within the investigated design space. This work analyzes a thermodynamic trends and optimization framework intended to support future experimental and techno-economic investigations of invasive aquatic biomass.

In precision manufacturing, accurate prediction of tool wear is pivotal for maintaining production quality and efficiency. This paper introduces a novel predictive model that employs a multi-branch convolutional neural network (multi-CNN) enhanced by an attention mechanism and Bayesian optimisation to forecast tool wear with high accuracy. The proposed approach is evaluated using the publicly available dataset, which comprises force and torque measurements acquired from machining processes. By incorporating time-domain (TD), frequency-domain (FD), and discrete wavelet transform (DWT) features, the model comprehensively analyses signals from machining processes, enabling robust predictions. The attention mechanism strategically emphasises significant features, augmenting the model’s predictive capability. Demonstrated results show superior performance with a mean absolute error (MAE) of 2.8, root mean squared error (RMSE) of 3.4, and an R2 score of 0.986, significantly outperforming conventional models. Additionally, post-training quantisation optimises the model for deployment on edge computing devices, maintaining effectiveness while being lightweight, ideal for real-time applications in resource-constrained environments. This study exemplifies how advanced machine learning techniques can improve predictive maintenance in manufacturing which not only enhances operational efficiency but also reduces downtime.

The growing number of end-of-life electric vehicle (EV) batteries demands scalable and adaptive disassembly solutions to enable efficient recycling. This paper presents a near-industrial robotic demonstrator for adaptive EV battery pack disassembly. The system integrates two industrial manipulators, force/torque sensing, automatic tool changing, and structured-light 3D vision within a ROS2-based perception and control architecture. A YOLO-based detection pipeline combined with point-cloud processing and coordinate transformation enables the automatic generation of robot targets for multi-step operations, including nut removal and battery module extraction. The results demonstrate that the modular integration of industrial robotics and AI-based perception can enable adaptive, multi-operation battery disassembly in a near-industrial environment. Experimental validation across 36 disassembly trials demonstrates reliable module extraction with a 100% success rate and an overall nut removal success rate of 77.6%, with performance degradation primarily linked to perception-localisation errors for off-axis nut detections.

Selecting interior wood materials requires balancing functional and aesthetic demands with indoor air quality, as wood can emit volatile organic compounds (VOCs) and very volatile organic compounds (VVOCs) that either peak shortly after installation or persist over time. This study investigated time-resolved VOC and VVOC accumulation concentrations from untreated pine, spruce, aspen, and oak, representing wood species commonly used in Norway. Concentrations of VOCs and VVOCs were monitored for 25–28 days under low-ventilation, worst-case chamber conditions using Proton Transfer Reaction–Time-of-Flight Mass Spectrometry (PTR-TOF-MS). Emission behaviour was assessed using peak concentration, decay rate, half-time, stabilised concentration, cumulative exposure (AUC28), and available health-based thresholds. Clear species-dependent differences were observed. Pine was the dominant VOC source, with the highest peak concentration, stabilised concentration, and AUC28, mainly due to monoterpenes, which remained above the health-based threshold throughout the test period. Oak showed the second-highest VOC burden and a distinct acetate-dominated profile, with acetic acid showing the longest predicted exceedance duration. Aspen and spruce had lower VOC burdens, but their cumulative VVOC exposure exceeded VOC exposure. VVOCs generally peaked earlier than VOCs, although persistence varied by species. Overall, the concentration of VOCs and VVOCs was controlled by species, compound chemistry, volatility, and internal transport, supporting species- and compound-specific ventilation and material-selection strategies.

Low-voltage renewable energy harvesting applications, particularly hybrid photovoltaic–thermoelectric generator (PV–TEG) systems, require high-gain DC–DC converters capable of efficiently boosting low and fluctuating input voltages while maintaining high conversion stability. However, conventional converter topologies often suffer from limited voltage gain, increased semiconductor stress, and reduced performance under the variable operating conditions characteristic of hybrid PV–TEG energy harvesting. To address these challenges, this paper proposes a novel Charge-Pump-Assisted Voltage Booster (CPAVB) converter for low-voltage hybrid PV–TEG energy harvesting applications. The proposed topology integrates inductive and capacitive energy storage elements in a charge-pump-assisted series-stacking configuration to achieve a high static voltage gain at moderate duty cycles while reducing component stress and improving conversion stability. Comprehensive simulation studies were performed in MATLAB/Simulink under diverse operating conditions, including uniform irradiance and temperature, partial shading with thermal non-uniformity, and low-voltage scenarios. The proposed converter demonstrated stable operation, rapid dynamic response, and accurate maximum power extraction. Furthermore, its applicability to a single-phase grid-connected hybrid PV–TEG system was investigated through simulation, where it maintained a well-regulated DC-link voltage, enabled unity power factor operation, achieved a simulated grid-current total harmonic distortion (THD) of 0.08%, and injected active power closely matching the theoretical hybrid maximum. To validate the proposed converter, a hardware prototype was developed and experimentally evaluated. The experimental results confirmed the predicted voltage gain, stable low-voltage operation, and acceptable semiconductor voltage stresses, demonstrating the practicality of the proposed converter. Overall, the combined simulation and converter-level experimental results demonstrate that the proposed CPAVB converter is a promising high-gain power conversion solution for low-voltage renewable energy harvesting applications and provides a practical interface for future hybrid PV–TEG systems and microgrid integration.

Stator inter-turn short circuit faults in shipboard synchronous generators (SGs) can rapidly escalate from minor insulation degradation into catastrophic failures, threatening the operational continuity and safety of electrified vessels. Therefore, early fault severity estimation of SGs is essential to prevent extensive downtime and ensure the resilience of shipboard power systems. However, the conventional data-driven approaches typically require labeled faulty data, which are rarely obtainable in practical maritime operations. To address this limitation, this article proposes a machine learning framework that enables early fault severity estimation without the need for experimental fault samples for training. The proposed method couples two neural networks through a physics-guided cycle-consistency (PGCC) training strategy. This approach facilitates both high-fidelity synthetic data generation and domain adaptation from analytical simulations to experimental measurements. The framework is validated on an in-house 7.5 kW test bench, demonstrating that the proposed method achieves high-accuracy prediction of unseen experimental fault severities. Ablation studies confirm the contribution of each network component and the PGCC training strategy. Furthermore, a comparative study demonstrates that the proposed method outperforms the tested domain adaptation and generative adversarial networks-based methods, particularly under limited training data.

To support innovation in aluminum melt treatment research, a pilot-scale modular liquid metal loop, capable of holding up to 200 kg of metal, has been designed and constructed to replicate melt treatment processes under controlled conditions. As the first stage in systematic benchmarking and evaluation of the loop, the performance of the permanent magnet pump (PMP) was evaluated through Lorentz force and Joule heating measurements conducted on aluminum billets. Force measurements were found to scale linearly with the rotational frequency of the pump and decrease with distance in accordance with the governing equations. For Joule heating, neither distance nor frequency trends reflected theoretical predictions, and unaccounted-for effects, such as the penetration depth of an AC magnetic field, are suspected to be responsible. The Joule heating deviations, while meriting further investigation for industrial melt stirring and control, do not impede loop commissioning. These results demonstrate the modular liquid metal loop’s suitability for academic to industrial investigation of aluminum melt processes and recycling phenomena.

Floating photovoltaic (FPV) systems are rapidly gaining attention as a viable solution for large scale renewable energy deployment in offshore environments, where mooring systems represent a significant part of the overall installation cost. This study develops a weight-based cost estimation method to evaluate the economic feasibility of alternative mooring designs for FPV applications. This study develops a weight-based cost estimation framework to evaluate the economic feasibility of alternative mooring designs for FPV applications. The proposed methodology estimates total mooring cost as the sum of line material cost, additional component cost, and installation cost, with material cost calculated directly from the total mass of the mooring line. By relating cost to physical properties such as line length, diameter, and linear density, this approach provides a more realistic and transparent representation of manufacturing and material expenses than conventional cost-per-meter methods. Three different mooring configurations are examined: a baseline catenary system, a configuration with clump weight, and an anchorline system. Each configuration is assessed across multiple line diameters to capture variations in material usage and deployment requirements. Previous hydroelastic analysis showed that the clump-weight configuration leads to reduced mooring loads while still satisfying prescribed motion responses. These lower loads enable more efficient use of line material and components, which in turn reduces total system cost. A sensitivity study on the installation coefficient is also carried out to represent different offshore deployment scenarios. Results indicate that line material remains the primary cost driver, with component and installation costs also contributing significantly to total Capital Expenditures (CAPEX). Within the examined diameter range of 90–130 mm, the 110 mm diameter mooring line with clump weights achieves the most cost-efficient balance, offering minimal additional requirements and substantially lower total cost compared to anchor line systems. The findings demonstrate that a weight-based cost framework provides a practical and transparent tool for comparing mooring designs and can guide economical decision-making in future FPV developments.

Background Demographic shifts are increasing the global demand for long-term care services, coinciding with a worldwide shortage of health care personnel. Service robots, designed to perform tasks in both professional and personal use, are perceived as a potential solution to alleviate health care personnel’s workload and enhance the quality of care. However, the existing literature is fragmented and heterogeneous, with a limited emphasis on the role of service robots in supporting residents rather than health care personnel. Furthermore, there is a lack of consistent definitions of service robotic technologies and a scarcity of studies on implementation models and frameworks. Objective This scoping review aims to map and synthesize evidence regarding the implementation of service robots as work support for health care personnel in long-term care settings. Methods A comprehensive 3-step search will be conducted in Embase, MEDLINE, APA PsycInfo, CENTRAL, Scopus, and CINAHL, along with gray literature databases and institutional repositories. Eligible sources encompass empirical studies and gray literature involving service robots, health care personnel, residents aged 65 years or older, and stakeholders such as informal caregivers within institutional long-term care. Exclusions apply to studies on home care, medical or industrial robots, and nonrobotic technologies. Data will be extracted and analyzed using the Joanna Briggs Institute methodology, with findings presented in tables, diagrams, and narrative summaries to identify gaps and inform future research and implementation strategies. Results The project has been funded for a 4-year period starting in April 2025. This protocol was developed in October 2025 and subsequently registered in November 2025. A comprehensive search strategy was formulated and completely conducted on October 24, 2025. The screening of 4884 titles and abstracts was completed in December 2025, resulting in the retrieval of 64 (1.3%) full-text articles for eligibility assessment. Subsequent phases, including data extraction, analysis, evidence synthesis, and presentation of results, will be conducted sequentially. The scoping review is expected to be finalized by June 2026. Conclusions This scoping review is expected to delineate the extent and characteristics of the existing evidence on service robots as work support for health personnel in long-term care settings. It will highlight the key reported outcomes and challenges encountered in implementation studies, as well as the theoretical frameworks, models, and concepts applied to address these issues. Trial Registration Open Science Framework QWK58; https://osf.io/qwk58/ International Registered Report Identifier (IRRID) PRR1-10.2196/89435

Standalone DC microgrids for EV charging stations are essential for the synergistic integration of energy storage technologies with smart grids and multi-energy complementary systems. However, stochastic dynamic EV charging loads in cold climates often overstress Battery Energy Storage Systems (BESS) integrated with dispatchable small/mini-hydro and renewables within DC microgrid, causing premature degradation of BESS. Conventional strategies struggle to coordinate these nonstationary demands effectively. This paper proposes a Two-Layer Learning Framework for storage-aware management, facilitating a low-carbon transformation of power networks. It integrates a machine learning layer for demand forecasting with a heuristic-based adaptive layer that dynamically adjusts control horizons based on real-time BESS constraints. By prioritizing renewable and hydro dispatch while restricting BESS to corrective support, the approach maintains DC-link stability and operational flexibility. Validation using Oslo’s cold-climate datasets demonstrates that the framework improves system reliability by 12.41% and sustainability by 24.65%. Notably, peak and cumulative BESS loading are reduced by 40% and 61%, respectively. Real-time hardware-in-the-loop (Typhoon HIL-604) validation confirms the framework’s efficacy for creating stable, resilient, and multi-energy standalone power infrastructures.

We consider the non-homogeneous first-order differential equation with hysteresis described by the Krasnoselskii-Pokrovskii (KP) rate-independent hysteresis operator. Existence and uniqueness of solutions as well as the boundedness of solutions in response to a bounded input are proved. The global stability of the non-homogeneous equation is also investigated. Periodic solutions and their stability are also studied. The differential equation under analysis constitutes the so-called inversion-free feedforward control, which was proposed for mitigating arbitrary rate-independent hysteresis effects in the actuated systems. The experimentally identified non-smooth and non-strictly monotonic hysteresis of a magnetic shape memory alloy (MSMA) actuator serves as a motivating case study. The performed analysis is settled in a series of theorems which are then illustrated by numerical examples.

The growing demand for safe, sustainable, and high-power energy storage systems has accelerated the development of sodium-based batteries as promising alternatives to conventional lithium-ion technologies. In particular, solid-state sodium batteries (Na SSBs) offer improved safety, wider electrochemical stability, and the potential for fast-charging operation. However, the anode often becomes the key kinetic and stability bottleneck due to sluggish Na+ diffusion, interfacial polarization, and chemo-mechanical degradation. This review presents a comprehensive overview of recent advances in anode materials for fast-charging solid-state sodium batteries, emphasizing the fundamental mechanisms governing Na+ transport, interfacial electrochemistry, and mechanical stability. The article first discusses the fundamental processes controlling charge–discharge behavior in Na SSBs. Subsequently, major classes of anode materials are systematically reviewed, including metallic sodium, alloy-type anodes (Si, Ge, Sn, Sb, Bi), carbon-based frameworks, composite architectures, and emerging high-entropy or hybrid materials. Particular attention is given to interface engineering strategies, microstructural optimization, and electrolyte compatibility that enable improved fast-charging performance and enhanced cycling stability. Finally, current challenges and future research directions are discussed, highlighting the need for integrated materials design, advanced characterization, and computational screening to accelerate the development of high-performance anodes. This review provides a mechanistic and design-oriented roadmap for engineering next-generation fast-charging solid-state sodium batteries.

This paper introduces a fast method for high-quality 3D Gaussian Splatting (3DGS) reconstruction without traditional Structure-from-Motion (SfM). The proposed approach leverages 3D Foundation Models (3DFMs) for camera pose and pointcloud initialization, then jointly optimizes both camera poses and Gaussian primitives using a depth-guided loss function. This enables fast convergence even from rough initialization with as few as 50-60 input views. To further improve reconstruction quality in sparse-view scenarios, an MLP-based pose refinement module is introduced alongside depth-guided supervision from the foundation model. Extensive experiments on Mip-NeRF 360, Tanks and Temples, and RobustNeRF demonstrate that the proposed method achieves competitive reconstruction quality (23.61 dB PSNR, 0.19 LPIPS) while reducing training time to approximately three minutes per scene. The proposed method produces ready-to-use 3DGS models at a fraction of the time required by existing pipelines, making it suitable for near realtime applications in robotics, VR, and autonomous navigation.

Floating photovoltaic (FPV) systems have gained increasing attention for offshore renewable energy generation. However, the deployment of these systems in open seas is associated with significant challenges due to wave-induced deformations and structural failures analysis accounts for the fluid–structure interaction and is essential for evaluating FPV systems. This study assesses the influence of mooring configurations on the hydroelastic responses of a prototype offshore FPV array, which consists of six interconnected floats. The three mooring configurations include tendon leg buoys with shared mooring, tendon leg buoys with shared mooring and clump weights, and catenary-shape anchor lines. The numerical study is carried out using a previously developed hybrid finite element and boundary element method. The floats are modeled as rectangular solid plates using Mindlin plate theory, and the modal expansion method is adopted to reduce computational time of the FPV array with 2724 degrees of freedom. We apply Irvine’s equations to analytically model the mooring lines connected to the fairleads, and then linearize the mooring stiffness. The modified stiffness matrices including hydrostatic stiffness, connection stiffness, and mooring stiffness are used in the hydroelastic analysis. A detailed sensitivity study is performed for the FPV array considering various mooring properties and wave periods. Among the three evaluated mooring configurations, the clump weight proved to be the most balanced solution. It offers a moderate increase in vertical mooring stiffness while keeping pretension loads relatively low, namely less than 150 kN and maintaining hydroelastic bending moments and plate deformations across the FPV array at modest levels, with the percentage difference in hydroelastic response remaining under 5%. This can be ascribed as a practical and cost-effective option for improving the structural integrity and motion performance of offshore FPVs.

Seminar June 2026 for all participants involved in the project with invited presenters.

The University of Agder invites suppliers and technology providers to a dialogue meeting as part of the research project SAKO. The aim of the meeting is to explore potential technological solutions and collaboration related to the development of a service robot solution for use in nursing homes, with a particular focus on relieving staff from heavy and repetitive transport tasks in their daily work.

Non-collinear antiferromagnets (NCAFMs) are appealing for antiferromagnetic spintronics, as they combine the advantages of collinear antiferromagnets with novel emergent phenomena stemming from their complex spin structures. These phenomena are often associated with the intrinsic spin chirality, which characterizes the handedness of the ground-state spin configuration. Non-collinear antiferromagnets (NCAFMs) are appealing for antiferromagnetic spintronics, as they combine the advantages of collinear antiferromagnets with novel emergent phenomena stemming from their complex spin structures. These phenomena are often associated with the intrinsic spin chirality, which characterizes the handedness of the ground-state spin configuration. Here, we investigate a kagome NCAFM interfaced with a normal metal and demonstrate that the ground-state vector spin chirality can be probed through measurements of the spin Seebeck effect (SSE). Starting from a microscopic spin Hamiltonian, we derive the corresponding bosonic Bogoliubov-de Gennes Hamiltonians for the two chiral configurations. Using linear response theory, we obtain a general expression for the spin current thermally pumped into the normal metal by the SSE. We show that a sizable in-plane spin current emerges exclusively in the negative-chiral state, providing a direct signature for real-time detection of chirality switching in kagome NCAFMs. In addition, we find a field-dependent out-of-plane spin current whose magnitude differs between the two chiralities by about 81%, reflecting their distinct magnon band structures.

Altermagnets have recently attracted considerable interest due to their unique symmetry-governed spintronic properties. Here, we investigate phonon-induced magnon spin currents in a two-dimensional altermagnet. Starting from a microscopic theory of the coupled magnon-phonon system, we derive the nonequilibrium magnon distribution generated by selective phonon excitations. We show that the resulting spin currents exhibit a pronounced d-wave symmetry with respect to the phonon momentum. Moreover, the spin current along the altermagnetic directions can be completely reversed by tuning the phonon frequency. These findings establish altermagnets as promising platforms for realizing highly tunable, phonon-driven coherent terahertz magnon spin currents.

OPINION: Agriculture is under pressure from several directions at once. Autonomous robots are increasingly presented as part of the answer. But safety has to come first.

Økende interesse for mere energivennlige løsninger samt innovative løsninger innen elektrisk aktuering har resultert i økt interesse for forskning og utvikling innen mer energivennlige hydrauliske løsninger. Pumpestyrte hydrauliske sylindre er en lovende kandidat i denne sammenheng, og har allerede etablert seg på markedet i flere industrielle applikasjoner. Regulering med tilbakekobling er ofte nødvendig i praksis, og har fått omfattende oppmerksomhet i forskningslitteraturen for ventil-styrte hydrauliske sylindre siden 1950 og 1960-tallene. Dynamikken til pumpestyrte sylindre, derimot, skiller seg fra dynamikken til ventilstyrte system, noe som gir opphav til flere ufordringer vedrørende dems applikasjoner i mange praktiske applikasjoner. Avhengig av den hydrauliske arkitekturen så kan utfordringer oppstå allerede før en lukket sløyfe tilbakekobling lukkes om sylinderet. Ett godt eksempel er hydrauliske arkitekturer som beytter kun én hydraulisk pumpe. Til tross for enkle løsninger tilgjengelig i litteraturen, har majoriteten av dem fått påvist ustabil oppførsel under enkelte operasjonsforhold selv uten tilbakekobling. Dette gir opphav til løsninger som benytter to pumper per sylinder, men også her kan det oppstå en rekke utfordringer. Én ikke-triviell utfordring felles for bade én- og to-pumpe system har opphav i forskjellene i trykkforholdene til slike system, til sammenligning med tradisjonelle ventilstyrte system, hvor den effektive bulkmodul kan påta seg drastisk lave verdier, noe som påvirker både ytelse og stabilitet negativt. En annen kritisk utfordring relaterer til aktueringselementet benyttet i pumpestyrte system. I tradisjonelle ventilstyrte system består aktueringselementet med en liten magnetisk spole, mens i pumpestyrte system er denne byttet ut av en mye større elektrisk motor. Dette gir opphav til nye utfordringer relater til modellering, identifikasjon og regulering av pumpestyrte hydrauliske sylindre. Denne avhandlingen studerer broken av pumpestyrte hydrauliske sylindre i applikasjoner hvor tilbakekobling om aktuatoren er påkrevd. En rekke forskningshull relatert til bruken av disse systemene blir identifisert og anses som kritiske problemstillinger som bør løses før utstrakt anvendelse av disse systemene kan anbefales i praktiske applikasjoner. Fem vitenskapelige publikasjoner har blitt dedikert til temaet med intensjon om å løse disse kritiske problemstillingene, med hovedfokus på ytelse, stabilitet, energieffektivitet samt modellering og reguleringsteknisk design.

High-temperature latent heat thermal energy storage (LHTES) requires phase change materials (PCMs) with high energy density, good thermal conductivity, thermal stability, and mechanical reliability above 1000 °C. Ni-based alloys, particularly nickel silicide intermetallic phases, are promising candidates because they are high temperature materials and offer good oxidation resistance. However, their practical application can be limited by the intrinsic brittleness of intermetallic compounds. Therefore, this thesis investigates ternary Ni-Si-Fe metallic PCMs as potential materials for high-temperature thermal energy storage (TES). Ni-Si intermetallic alloys already exhibit higher melting point and excellent high-temperature capability.In this study, Fe was introduced as a ternary alloying element with the aim of improving the thermal stability and microstructural characteristics of the Ni–Si system while maintaining its high energy storage capacity. Three alloy compositions were designed using CALPHAD-based Thermo-Calc simulations with the TCNI13 database: 73.96Ni-16.90Si-9.14Fe, 60.90Ni-29.70Si-9.40Fe, and 49.40Ni-48.00Si-2.60Fe wt.%. Liquidus projection analysis, equilibrium calculations, and Scheil solidification simulations were used to predict phase stability, melting behavior, density, enthalpy, energy density, thermal conductivity, mass fraction of phases and volume change. The alloys were experimentally synthesized using high-temperature furnace processing and characterized by DSC, XRD, SEM-EDS, and Vickers micro hardness testing to validate the thermodynamic predictions. For the 73.96Ni-16.90Si-9.14Fe alloy, the main phases were identified as γ-Ni31Si12, δ-Ni2Si, and BCC(Ni-9.6Si-65.7Fe), which remained relatively stable at high temperature. The 60.90Ni-29.70Si-9.40Fe alloy contained mainly NiSi, along with θ-Ni2Si and FeSi(Ni-32.7Si-25.8Fe), while the 49.40Ni-48.00Si-2.60Fe alloy consisted predominantly of ζ-NiSi₂, with minor fractions of NiSi and FeSi(Ni-32.7Si-26Fe). The higher silicon contents in the B and C alloys contributed to increased brittleness, as observed from SEM analysis and hardness testing. Overall, the experimentally identified phases were consistent with the Thermo-Calc predictions. All alloys exhibited melting temperatures above 1000 °C in both the predicted calculations and DSC measurements. The calculated energy densities of the three alloys were 1.01, 0.83, and 1.09 MWh/m³, respectively. The DSC experimental results showed good consistency with the predicted values for the A and B alloys, while a larger deviation was observed for the alloy C. Among the investigated compositions, 73.96Ni-16.90Si-9.14Fe wt.% showed the best overall performance by combining an energy density above 1 MWh/m³, high operating temperature, stable phase formation, and a crack-free microstructure. Therefore, this composition is identified as the most promising Ni-Si-Fe PCM candidate for high-temperature TES applications.

Denne doktoravhandlinga presenterer tre forskingsartiklar publiserte i tidsskriftet Physical Review B om ikkje-kollineære antiferromagnetiske spintroniske fenomen, og meir spesifikt om kagomé-antiferromagnetar. Tvei av desse publikasjonane undersøkjer spinstraumane som genererest ved grenseflatene mellom ein kagomé-antiferromagnet og eit normalt metall på grunn av spin-Seebeck-effekten, dess forsterking er arbeidets hovudfokus. Avhandlinga tek til med ei generell innføring i magnetisme og kagomé-gitter-antiferromagnetanes eigenskapar. Deretter, med utgangspunkt i ein mikroskopisk tight-binding-Hamiltonian, nytta vi Holstein–Primakoff-formalismen for å undersøkje kva område av materialparametrar gjer at eit kagomé-antiferromagnetisk monolag viser topologiske eigenskapar, slik som topologiske magnontilstandar avgrensest til monolagets kantar. Våre resultat syner at Dzjaloshinskij-Morija-vekselverknadens plane komponents nærvær og tilstrekkelege styrke er avgjerande for at slike tilstandar skal førekoma. Den kritiske verdien over kva kagoméantiferromagneten gjeng inn i eit veldefinert topologisk regime, fanst å vere om lag 30 % av den isotrope spinvekselverknadens styrke. Vidare presenterest ei utvida effektiv-verknadsframstilling av magnona i denne magnetiske strukturen, inkludert ledd som er avhengige av påført mekanisk tøying, for å undersøkje korleis grunntilstandens spin-kiralitet kan bytast mellom positive og negative konfigurasjonar. Det vart oppdaga at ei jamn tøying på 7.75 % er tilstrekkeleg til å tvinga materialet til å skifta kiralitet utan trong for termiske fluktuasjonar. Vidare fanst ei målbar, ikkje-forsvinnande netto magnetisering, eksklusiv for kagomé-antiferromagneten med negativ spin-kiralitet, som kan nyttast som eit middel for eksperimentell påvising av kiralitetsbyte. Denne magnetiseringas nærvær vart stadfesta gjennom ei mikroskopisk behandling av modellen. Til slutt undersøkjer vi korso desse to fenomena, nemleg spin-kiralitetsbyte og topologiske magnons nærvær, påverkar spin-Seebeck-straumens styrke. I det fyrste tilfellet, ved nytte av materialparametrane til Mn3Ir, fann vi at spinstraumen har ein konstant y-komponent som er karakteristisk for tilstanden med negativ kiralitet. Denne komponenten kan målast i det tilstøytande normale metallet som ei spenning indusert av den inverse spin Halleffekten. Vidare fann vi ein skilnad på om lag 81 % mellom spinstraumanes magnetfeltavhengige z-komponentar i dei to kirale grunntilstandane, som kan målast gjennom den anomale inverse spin Hall-effekten. Til slutt, for ein kagomé-antiferromagnet i det topologiske regime, fann vi at den topologiske kanttilstanden i det fyrste bandgapet bidreg med kring 28 % av den totale spin-Seebeck-straumen, som impliserer at topologiske tilstandars nærvær gjev ei tydeleg forsterking av spinstraumen.

Offshore crane lifting is among the most hazardous routine operations in the petroleum industry, much of the difficulty arising from the relative motion between a wave-excited vessel and the load it handles. This thesis develops and evaluates an autonomous pick-and-place system for a knuckle-boom crane mounted on a six-degree-of-freedom Stewart platform, which reproduces vessel motion on land, and establishes the simulation infrastructure to extend the system toward learning-based control. The previously separate crane and platform were integrated into a single operational unit coordinated by a distributed ROS~2 control architecture that bridges both Beckhoff PLCs to a high-level autonomous controller with active six-degree-of-freedom motion compensation and a vision-driven pickup pipeline. A physically consistent NVIDIA Isaac Sim model runs as a real-time digital shadow of the rig, structured to support future reinforcement-learning work. Evaluated against an optical motion-capture reference, full six-degree-of-freedom compensation reduced the boom-tip deviation by a factor of four to six at low platform intensity, holding the tool point steady enough for autonomous operation across calm and moderate sea-state analogues. On a stationary base the pipeline completed every cycle on a reachable payload with no autonomous pickup failures, and under emulated vessel motion the crane captured the payload in all twenty-one attempts. The digital shadow reproduced the tool-point trajectory to 8.3~mm three-dimensional RMS, tracking the live telemetry roughly twice as closely as that telemetry tracked physical reality. The result is a working, instrumented platform for studying offshore lifting on land, together with a validated digital shadow that prepares the system for the learning-based control it was built to enable.

Offshore crane lifting is among the most hazardous routine operations in the petroleum in- dustry, much of the difficulty arising from the relative motion between a wave-excited vessel and the load it handles. This thesis develops and evaluates an autonomous pick-and-place system for a knuckle-boom crane mounted on a six-degree-of-freedom Stewart platform, which reproduces vessel motion on land, and establishes the simulation infrastructure to ex- tend the system toward learning-based control. The previously separate crane and platform were integrated into a single operational unit coordinated by a distributed ROS 2 control architecture that bridges both Beckhoff PLCs to a high-level autonomous controller with active six-degree-of-freedom motion compensation and a vision-driven pickup pipeline. A physically consistent NVIDIA Isaac Sim model runs as a real-time digital shadow of the rig, structured to support future reinforcement-learning work. Evaluated against an optical motion-capture reference, full six-degree-of-freedom compen- sation reduced the boom-tip deviation by a factor of four to six at low platform intensity, holding the tool point steady enough for autonomous operation across calm and moderate sea-state analogues. On a stationary base the pipeline completed every cycle on a reachable payload with no autonomous pickup failures, and under emulated vessel motion the crane captured the payload in all twenty-one attempts. The digital shadow reproduced the tool- point trajectory to 8.3 mm three-dimensional RMS, tracking the live telemetry roughly twice as closely as that telemetry tracked physical reality. The result is a working, instrumented platform for studying offshore lifting on land, together with a validated digital shadow that prepares the system for the learning-based control it was built to enable.

Stive ledd brukes ofte i tradisjonelle design av robotmanipulatorer fordi de gir rask respons og høy posisjonsnøyaktighet. De er egnet for anvendelser der hastighet og nøyaktighet er de viktigste hensynene. Imidlertid er de ofte innelukket på grunn av den potensielle risikoen for menneskers helse og omgivelsene. På den annen side verdsettes elastiske ledd for sin iboende sikkerhet ved interaksjon med omgivelsene. De har imidlertid treg respons og kan avvike fra ønsket posisjon når en kollisjon oppstår. For å kombinere fordelene og redusere ulempene ved begge leddtyper kan robotmanipulatorer med blandede stive–elastiske ledd vurderes. Posisjonsreguleringsalgoritmer fungerer godt i frie rom, der robotmanipulatoren kan bevege seg fritt. Imidlertid kan posisjonsregulering føre til for store kontaktkrefter når en kollisjon oppstår. Kraftreguleringsalgoritmer fungerer derimot godt i begrensede rom, der robotmanipulatoren er i tett kontakt med omgivelsene. Imidlertid har kraftregulering dårlig posisjonsnøyaktighet. Impedansregulering er velkjent som en metode for å regulere forholdet mellom kontaktkraft og posisjonsfeil. Ved å ta i bruk impedansregulering kan sikkerheten forbedres ikke bare gjennom maskinvare, men også gjennom programvare. Hovedmålet med dette arbeidet er å forbedre sikkerheten i menneske–robotinteraksjon. Spesielt utvikles en robotmanipulator med blandede stive–elastiske ledd, og det utvikles reguleringsalgoritmer, inkludert impedansregulering. Arbeidet har tre hovedbidrag. For det første utledes den matematiske modellen for denne typen robotmanipulator. Egenskapene til den matematiske modellen analyseres. For det andre utvikles dedikerte reguleringsalgoritmer. Målet med de foreslåtte reguleringsalgoritmene er å sikre sikkerhet når robotmanipulatoren samhandler med omgivelsene. Til slutt designes en prototype av en robotmanipulator med blandede stive–elastiske ledd. Robotmanipulatoren har seks ledd. De tre første leddene er stive, og de tre siste leddene er elastiske. Delene til robotmanipulatoren er 3Dprintet og montert. Denne prototypen brukes til å teste de foreslåtte reguleringsalgoritmene. Dette arbeidet gir en systematisk utledning av den dynamiske modellen for robotmanipulatorer med blandede stive–elastiske ledd. Et annet viktig vitenskapelig bidrag ligger i utviklingen av reguleringsalgoritmer tilpasset manipulatorer med blandede stive–elastiske ledd. I motsetning til konvensjonell impedansregulering utviklet for fullt stive systemer, tar de foreslåtte strategiene eksplisitt hensyn til leddenes elastisitet i regulatorutformingen. Ved å forbedre sikkerheten under fysisk interaksjon støtter den foreslåtte tilnærmingen en bredere bruk av roboter i industri, helsevesen og hjemmemiljøer. Integrasjonen av elastisitet og impedansregulering bidrar til sikrere systemer for menneske–robot-interaksjon.

Energy demand has increased significantly following the pandemic and the geopolitical instability worldwide. According to the IEA report, the growth of renewable sources reached 34.5% by the end of 2025. Although remarkable efforts are being made to develop and deploy renewable energy sources to meet the demand, they remain well short of the annual growth target of 8%. Moreover, to accelerate annual energy growth, permanent magnet synchronous machines (PMSMs) are central to the sustainable development of large-scale energy production, energy transition, electrification, autonomous vehicles, and the aerospace industry. PMSMs are now becoming the preferred choice for large -scale wind turbines due to their high efficiency, precise control, torque density, and ease of maintenance. Despite the rapid growth of diagnostic technologies enabled by artificial intelligence, PMSMs are subject to electrical faults and mechanical and thermal stresses under dynamic operating conditions. Additionally, the scarcity of historical fault data in practical applications limits the effectiveness of data-driven approaches, which typically require large amounts of labeled data. Therefore, robust fault detection and identification (FDI) based on data-driven approaches are not only key to deploying reliable, safe energy systems but also to reducing labor and maintenance costs. This dissertation addresses these challenges through three key contributions. First, it improves FDI performance under low-severity faults and dynamic operations. Second, a framework is developed to integrate domain knowledge with data-driven approaches to enable explainable decision making across different time scales in predictive maintenance. Third, an autonomous online fault diagnosis framework is proposed to provide traceable decision trajectories and physical reasoning during dynamic operations. The proposed methods are described theoretically and validated through experimental setups. The results demonstrate that the models achieve robustness and explainability for each prediction in fault diagnosis, making them well-suited for deployment in predictive maintenance strategies for PMSMs. A brief summary of the models and their findings is presented in this dissertation, while detailed explanations and results are provided in the appendices.

This thesis investigates a layered diagnostic framework for identifying open-circuit faults (OCFs) in the three-level F-type inverter. The study evaluates whether a two-stage pipeline, comprising a statistical feature-based phase-leg Neural Network (NN) classifier followed by a Continuous Wavelet Transform (CWT) based Convolutional Neural Network (CNN) for intra-leg switch identification, can improve diagnostic accuracy relative to direct 13-class classification in previous literature. In the first stage, one-way ANOVA was used to rank time-domain statistical features extracted from windowed pole-voltage measurements. These features were applied to multiple machine-learning classifiers to determine the most effective model for phase-leg identification. In the second stage, two CWT image representations were generated from the same pole-voltage signals: magnitude-only colored scalograms and sign-preserving grayscale CWT coefficient maps. These images were used to train a custom CNN designed to discriminate among the four switch positions within each phase leg. The results show that decomposing the 13-class OCF problem into two serial 4-class decisions improves classification accuracy and reduces inter-class confusion. The grayscale signed CWT coefficient maps consistently provided higher discriminative power than standard CWT scalograms, particularly for faults with subtle differences in waveform distortions. The proposed two-stage architecture demonstrates that hierarchical fault localization is a viable strategy for complex inverter topologies and provides a solid foundation for future extensions toward online monitoring and multi-fault scenarios.

The purpose of this master thesis is to investigate how sensor measurements can be used to enhance the fatigue life estimates of an offshore crane. Sensor measurements combined with a fatigue calculation methodology can give information about the condition of a structure and support condition based maintenance. A study published by the National Institute of Standards and Technology (NIST) found that companies using more condition based maintenance had 18.5% less unplanned downtime and 87.3% fewer defects than companies using more fixed interval based maintenance. The master’s thesis project consists of developing a fatigue model that uses sensor measurements as input values to estimate accumulated fatigue damage. This model was made in Python, and the calculation methods were based on NOV design calculation report T5480-N-RD-002 and DNVGL-RP-C203. The goal for this project is to answer the following research question: “How can sensor measuSensor measurements can be used as input values in the fatigue model to describe how the crane moves and operates. Based on sensor measurements the fatigue model can calculate stress over time and estimate accumulated fatigue damage. Accumulated fatigue damage gives an estimate of how much of a structural detail’s fatigue life has been consumed, and how much remaining fatigue life it has. Sensor measurements that are important for calculating fatigue life are load, working radius, crane tip height, slew angle, main boom angle, knuckle boom angle, pitch angle, roll angle and crane lift mode. A high sampling rate for the sensor measurements should be used because it gives a more accurate stress-time history for the part being considered. The fatigue model can also be combined with calculations for fatigue crack growth and probability of crack detection to give a better prediction for when it is necessary to perform inspection and maintenance.rements from an offshore crane be used to estimate remaining fatigue life?”

Final version of my master’s thesis, which compares the suitability of first-life and second-life batteries for stationary energy storage applications.

This thesis investigates the influence of electrolyte volume and vinylene carbonate (VC) additive content on the electrochemical performance and degradation behavior of NMC811/graphite cylindrical lithium-ion cells. The effective pore volume of the cells was experimentally determined and used to define a reference pore filling ratio (PFR = 1.00). Based on this reference, cells with different electrolyte filling levels and adjusted VC contents were prepared to separately study the effects of electrolyte quantity and additive amount. Electrochemical characterization was performed using formation cycling, incremental capacity analysis (ICA), differential voltage analysis (DVA), electrochemical impedance spectroscopy (EIS), direct current pulse resistance measurements, and ultra-high precision coulometry (UHPC). The results demonstrate that electrolyte quantity strongly affects interfacial stability, impedance growth, and ageing behavior. Poorly filled cells showed lower Coulombic efficiency, lower discharge capacity, and significantly higher resistance growth, indicating non-uniform electrolyte wetting and transport limitations within the porous electrode structure. VC content was also found to strongly influence early-stage interfacial reactions and SEI formation. UHPC results further suggest that precision ageing parameters become less reliable for severely underfilled cells, likely due to local electrolyte-deficient regions and non-uniform electrolyte distribution. In contrast, direct electrochemical metrics such as discharge capacity, energy efficiency, and voltage hysteresis showed clearer degradation trends. Early ageing results indicated improved capacity retention for highly filled cells during moderate C-rate cycling, whereas underfilled cells degraded more rapidly. However, under high C-rate cycling, highly filled cells exhibited the poorest capacity retention, which is proposed to originate from electrolyte motion-induced salt inhomogeneity (EMSI). To support the electrochemical analysis, custom software tools for automated dV/dQ analysis were also developed. Overall, this work demonstrates that electrolyte optimization in high-energy lithium-ion batteries requires simultaneous control of electrolyte volume and additive quantity rather than additive concentration alone. The findings provide important insight into the relationship between electrolyte wetting, resistance growth, and degradation mechanisms in NMC811/graphite cells for future electric vehicle applications.

Early stage HVAC design in school buildings is subject to significant uncertainty in some areas, particularly regarding occupancy levels, which directly influences the ventilation demand and component sizing. Traditional sizing methods rely on deterministic peak loads with safety factors or probabilistic approaches, which primarily assess the continuous performance metrics such as energy use or airflow rates. These methods often overlook the discrete nature of practical design decisions, such as standard duct diameters and AHU capacity classes, where small variations in occupancy can trigger a stepwise system change. This thesis presents a threshold based analytical framework to assess the robustness of early stage ventilation design under occupancy uncertainty. The method defines discrete decision objects, then evaluate room and system level thresholds for duct sizing and AHU utilization, and quantifies the smallest occupancy variation that may require a categorical change. Derived indicators, including duct area, fan power, and ventilation heat loss, provide insight into the practical consequences. The framework is implemented in Python to enable an automated reproducible analysis, and integration of supplementary validation checks including baseline CO$_2$ concentration, although further refinements could enhance accuracy and applicability. Validation against IDA ICE simulations confirms that the method does identify threshold sensitive configurations and supports informed decision making without requiring detailed dynamic modeling. Applied to a representative school case, the framework classified classrooms as robust but close to the upper threshold, with a governing margin of 3 persons. Common areas showed elevated CO$_2$ concentrations up to 1093–1105 ppm in the baseline analysis. IDA ICE simulations for room 1065 indicated a ppm equal to 1469 ppm with a reduced thermal comfort at baseline occupancy (PPD = 49\%, PMV = 1.468), aligning with the methodology. Group rooms, in contrast, exhibited the most robust baseline performance, with sufficient margins and CO$_2$ levels below 1000 ppm, indicating a more stable and insensitive occupancy decisions. Overall, the framework allows designers to be able to distinguish robust from sensitive configurations, prioritize verification efforts and reduce the risk of under or over dimensioned systems, overall providing a practical tool for early stage HVAC design under uncertainty.

This master thesis investigates coarse-aggregate Ultra-High Performance Fiber-Reinforced Concrete (UHPFRC) as a thin overlay for the rehabilitation of normal concrete. The problem addressed is the deterioration of aging bridge decks caused by cracking, chloride ingress and reinforcement corrosion, and the need for repair materials that can improve both durability and structural performance. The research was conducted as an experimental laboratory study. UHPFRC mixtures with 2 and 3 vol.% steel fibers were produced and evaluated through fresh and hardened material testing, a 5◦ inclined casting assessment, as well as pull-off tests and four-point bending tests on UHPFRC-NC composite beams. Digital image correlation was used to analyze the flexural behavior and crack development. The results show that 3 vol.% steel fibers improved compressive strength, flexural resis- tance and crack distribution compared with 2 vol.%, although flowability was reduced. Both mixtures achieved sufficient bond, as failures mainly occurred in the normal concrete sub- strate rather than at the interface. The study concludes that coarse-aggregate UHPFRC is promising for bridge deck rehabilitation, and that the 3 vol.% mixture gave the best overall structural performance under the tested conditions. Practical application still requires care- ful control of flowability, viscosity, and consolidation, especially for inclined overlays utilizing viscosity modifying agents.

This master thesis investigates coarse-aggregate Ultra-High Performance Fiber-Reinforced Concrete (UHPFRC) as a thin overlay for the rehabilitation of normal concrete. The problem addressed is the deterioration of aging bridge decks caused by cracking, chloride ingress and reinforcement corrosion, and the need for repair materials that can improve both durability and structural performance. The research was conducted as an experimental laboratory study. UHPFRC mixtures with 2 and 3 vol.% steel fibers were produced and evaluated through fresh and hardened material testing, a 5 degree inclined casting assessment, as well as pull-off tests and four-point bending tests on UHPFRC-NC composite beams. Digital image correlation was used to analyze the flexural behavior and crack development. The results show that 3 vol.% steel fibers improved compressive strength, flexural resistance and crack distribution compared with 2 vol.%, although flowability was reduced. Both mixtures achieved sufficient bond, as failures mainly occurred in the normal concrete substrate rather than at the interface. The study concludes that coarse-aggregate UHPFRC is promising for bridge deck rehabilitation, and that the 3 vol.% mixture gave the best overall structural performance under the tested conditions. Practical application still requires careful control of flowability, viscosity, and consolidation, especially for inclined overlays utilizing viscosity modifying agents.

This master thesis investigates coarse-aggregate Ultra-High Performance Fiber-Reinforced Concrete (UHPFRC) as a thin overlay for the rehabilitation of normal concrete. The problem addressed is the deterioration of aging bridge decks caused by cracking, chloride ingress and reinforcement corrosion, and the need for repair materials that can improve both durability and structural performance. The research was conducted as an experimental laboratory study. UHPFRC mixtures with 2 and 3 vol.% steel fibers were produced and evaluated through fresh and hardened material testing, a 5 degree inclined casting assessment, as well as pull-off tests and four-point bending tests on UHPFRC-NC composite beams. Digital image correlation was used to analyze the flexural behavior and crack development. The results show that 3 vol.% steel fibers improved compressive strength, flexural resistance and crack distribution compared with 2 vol.%, although flowability was reduced. Both mixtures achieved sufficient bond, as failures mainly occurred in the normal concrete substrate rather than at the interface. The study concludes that coarse-aggregate UHPFRC is promising for bridge deck rehabilitation, and that the 3 vol.% mixture gave the best overall structural performance under the tested conditions. Practical application still requires careful control of flowability, viscosity, and consolidation, especially for inclined overlays utilizing viscosity modifying agents.

This master’s thesis is a feasibility study conducted as a part of the civil engineering, masers program at the University of Agder. The study is conducted in accordance with Grimstad municipality’s plans on constructing a new wastewater treatment plant. As a part of the treatment plant project, a need to construct a pumping station at Groosenestet in Grimstad Norway, was developed. The projects research question is: How can a feasibility study be used as a tool to shed light on the potential and effect of a place bound pump station design at Groosenestet, Grimstad? This study provides two polar design alternatives based on two postmodern architects, Christian Norberg-Schultz and Robert Venturi: The design process of alternative A, based on Christians Norberg-Schultz’s methodology, involved an attempt to visualise the spirit of the place through on cite observation, with additional historical and cultural qualities, combined and symbolised as one construction. The result is a pumpstation design as an overgrown industrial brick building with a historical rooftop view, an outdoor seating area and a universal accessible toilet. Alternative B, based on Robert Venturi’s concepts of messy vitality, inclusiveness and the difficult whole. The alternative address the potential of Groosenestet as a whole park. Taking inspiration from OMA’s Parc de la Vilette, the parks design is organised into four layers: the central plaza, activities, circulation and supporting elements. The pump station stands in the centre of the park as a traditional south Norwegian house with a playful twist while the park is design to serve under representative groups. The study concludes that this feasibility study effectively illustrates how a pumpstation can be designed to enhance the quality of Groosenestet in two different ways. The designs are to be used as basis for discussion about the future of Groosenestet.

This master’s thesis investigates how concrete can contribute to promoting marine biodiversity on Norwegian harbour quays, with a focus on nature-inclusive design (NID) in coastal infrastructure. The background for the study is the increasing loss of natural marine habitats as a result of coastal development, as well as the need for solutions that can combine technical requirements for strength and durability with ecological considerations. The main research question addressed is: How can concrete promote marine biodiversity on a Norwegian harbour quay? To answer this question, a combination of a literature review, laboratory work, and field-based experiments was conducted. The literature review covers existing research on the bioreceptivity of concrete, artificial substrates in marine environments, and the application of nature-inclusive design in maritime infrastructure. The experimental work consists of the production and testing of various concrete types, including traditional concrete (B45) with variations in colour and surface geometry, ultra-high performance concrete (UHPC), alkali-activated concrete (AAC), and permeable concrete. In total, 36 concrete tiles were exposed in seawater at a Norwegian harbour quay for a period of approximately 12 weeks. Biological establishment was assessed through visual observations and image analysis, where artificial intelligence was used as a supporting tool in the image analysis to estimate algal and biofilm coverage on the concrete surfaces. Laboratory tests were also conducted to assess the mechanical properties of the concrete. The results show that biological growth was established on all investigated concrete types during the exposure period. Brown algae and small amphipods dominated the early colonisation phase, and attached organisms such as barnacles were observed on most surfaces. The study indicates that local environmental conditions, particularly light availability and position in the water column, have a greater influence on early biological establishment than differences in concrete composition alone. Nevertheless, surface characteristics such as roughness, colour, and cavities may influence which organisms establish. Laboratory experiments show that the addition of colour pigments does not result in measurable negative effects on the compressive strength or resistance to chloride ingress of the concrete within the investigated parameters. The thesis concludes that concrete can function as a suitable substrate for marine biodiversity in harbour areas, and that nature-inclusive design can be integrated into concrete structures without compromising technical performance. Furthermore, long-term studies are recommended to investigate ecological development over time, as well as further development of standardised methods for measuring bioreceptivity. Interdisciplinary collaboration between engineering and marine biology is highlighted as crucial for realising sustainable solutions in future maritime infrastructure.

This report is the main thesis report in the subject “BYG508 – Masteroppgave”, completed as part of the civil engineering master’s degree at the University of Agder in Norway. This report is the final report of the master’s degree and was completed in the spring of 2026. The aim of the report was to answer the question «What is the effect of wood verying wood species on the mechanical behaviour of connections with wooden dowels». The research methodology used in this report is a combination of extensive literature review and experimental testing. The experimental tests have been on a connection design using oak, beech, birch and pine wood dowels. The world is in a climate crisis due to global warming. Most countries have goals to reduce climate emissions through treaties such as the Paris Agreement and the United Nations goals for sustainable development. However, it seems that most countries are not able to meet these goals, and there is need for more drastic measures to reduce climate emissions. The construction industry is one of the main contributors to global greenhouse gas emissions, mainly due to construction materials such as steel and concrete. In recent years, swapping steel fasteners with wooden fasteners has become a popular method of reducing the amount of steel in larger structures. However, wooden fasteners can not be designed according to the current design standards, as they are based on the behaviour of steel fasteners. Therefore, there is a need for more research on the behaviour of wooden fasteners and on methods to accurately predict the capacity of connections with wooden fasteners. By comparing the experimental results with the FprEN 1995 draft, this report shows that FprEN overestimates the ultimate capacity of connections with wooden dowel by 27-55 % depending on wood species. Furthermore, the results show that Miller’s method estimating yield capacity in failure mode V underestimated the yield capacity by 20-40 %. In comparison, the methods presented by Vilguts overestimated the yield capacity by around 25% for oak and pine dowels, and 1-6% for beech and birch dowels. Based on the experimental test results conducted, this report concludes with the new Eurocode 5 draft being somewhat insufficient in predicting the ultimate capacity of wooden dowels. This report suspects that the main reason for this insufficiency is due to the wooden dowels crushing from compression perpendicular to grain, which results in the diameter of the dowel being reduced. This reduction of the diameter in turns reduces the moment capacity of the dowel, which is the dominating factor of dowel capacity. This report has therefore suggested a new modified equation for estimating ultimate moment capacity of wooden dowels, by introducing an effective reduction in dowel diameter due to the crushing effect.

This master’s thesis investigates how concrete can contribute to promoting marine biodiversity on Norwegian harbour quays, with a focus on nature-inclusive design (NID) in coastal infrastructure. The background for the study is the increasing loss of natural marine habitats as a result of coastal development, as well as the need for solutions that can combine technical requirements for strength and durability with ecological considerations. The main research question addressed is: How can concrete promote marine biodiversity on a Norwegian harbour quay? To answer this question, a combination of a literature review, laboratory work, and field-based experiments was conducted. The literature review covers existing research on the bioreceptivity of concrete, artificial substrates in marine environments, and the application of nature-inclusive design in maritime infrastructure. The experimental work consists of the production and testing of various concrete types, including traditional concrete (B45) with variations in colour and surface geometry, ultra-high performance concrete (UHPC), alkali-activated concrete (AAC), and permeable concrete. In total, 36 concrete tiles were exposed in seawater at a Norwegian harbour quay for a period of approximately 12 weeks. Biological establishment was assessed through visual observations and image analysis, where artificial intelligence was used as a supporting tool in the image analysis to estimate algal and biofilm coverage on the concrete surfaces. Laboratory tests were also conducted to assess the mechanical properties of the concrete. The results show that biological growth was established on all investigated concrete types during the exposure period. Brown algae and small amphipods dominated the early colonisation phase, and attached organisms such as barnacles were observed on most surfaces. The study indicates that local environmental conditions, particularly light availability and position in the water column, have a greater influence on early biological establishment than differences in concrete composition alone. Nevertheless, surface characteristics such as roughness, colour, and cavities may influence which organisms establish. Laboratory experiments show that the addition of colour pigments does not result in measurable negative effects on the compressive strength or resistance to chloride ingress of the concrete within the investigated parameters. The thesis concludes that concrete can function as a suitable substrate for marine biodiversity in harbour areas, and that nature-inclusive design can be integrated into concrete structures without compromising technical performance. Furthermore, long-term studies are recommended to investigate ecological development over time, as well as further development of standardised methods for measuring bioreceptivity. Interdisciplinary collaboration between engineering and marine biology is highlighted as crucial for realising sustainable solutions in future maritime infrastructure.

This report is the main thesis report in the subject “BYG508 – Masteroppgave”, completed as part of the civil engineering master’s degree at the University of Agder in Norway. This report is the final report of the master’s degree and was completed in the spring of 2026. The aim of the report was to answer the question «What is the effect of wood verying wood species on the mechanical behaviour of connections with wooden dowels». The research methodology used in this report is a combination of extensive literature review and experimental testing. The experimental tests have been on a connection design using oak, beech, birch and pine wood dowels. The world is in a climate crisis due to global warming. Most countries have goals to reduce climate emissions through treaties such as the Paris Agreement and the United Nations goals for sustainable development. However, it seems that most countries are not able to meet these goals, and there is need for more drastic measures to reduce climate emissions. The construction industry is one of the main contributors to global greenhouse gas emissions, mainly due to construction materials such as steel and concrete. In recent years, swapping steel fasteners with wooden fasteners has become a popular method of reducing the amount of steel in larger structures. However, wooden fasteners can not be designed according to the current design standards, as they are based on the behaviour of steel fasteners. Therefore, there is a need for more research on the behaviour of wooden fasteners and on methods to accurately predict the capacity of connections with wooden fasteners. By comparing the experimental results with the FprEN 1995 draft, this report shows that FprEN overestimates the ultimate capacity of connections with wooden dowel by 27-55 % depending on wood species. Furthermore, the results show that Miller’s method estimating yield capacity in failure mode V underestimated the yield capacity by 20-40 %. In comparison, the methods presented by Vilguts overestimated the yield capacity by around 25% for oak and pine dowels, and 1-6% for beech and birch dowels. Based on the experimental test results conducted, this report concludes with the new Eurocode 5 draft being somewhat insufficient in predicting the ultimate capacity of wooden dowels. This report suspects that the main reason for this insufficiency is due to the wooden dowels crushing from compression perpendicular to grain, which results in the diameter of the dowel being reduced. This reduction of the diameter in turns reduces the moment capacity of the dowel, which is the dominating factor of dowel capacity. This report has therefore suggested a new modified equation for estimating ultimate moment capacity of wooden dowels, by introducing an effective reduction in dowel diameter due to the crushing effect.

With the growing interest in implementing circular economy principles through reuse of construction materials in Norway, reuse assessment and mapping becomes an important aspect and requirement in the country. There remains considerable opportunity for research to further develop reusemapping workflows. At the same time, BIM technologies provide opportunities for integrating reuserelated information directly into digital building models and could support improved accessibility, information management, and coordination throughout project development. This master’s thesis investigates how reuse potential analysis and mapping can be integrated into the BIM environment. To explore this, a BIM-based solution was developed utilizing Revit and Dynamo where a set of scripts was created for automated integration of reuse-related parameters into BIM models, and calculation and classification of reuse potential results of elements based on user definable weight factors and thresholds. Additional scripts were also made for creation of filters to visualize reuse potential results and EOL decisions. The workflow incorporates exportation of models with the reuse-related parameters and values through an open and standard IFC format. The developed BIM-based solution was evaluated of its application and potential in practice through a Norwegian case study. From the results of developing and evaluating the solution, it was concluded that BIM based digital tools could allow reuse mapping workflows to be performed directly within the BIM models, while also supporting automation of repetitive manual tasks. It provides great support for analysis, overview, and simulation of reuse potential and end-of-life scenarios, as well as improving data sharing and interoperability among project users through collaborative BIM platforms.

Wind energy is crucial for mitigating global greenhouse gas emissions, but the significant operation and maintenance (O\&M) costs associated with wind turbines necessitate advanced condition-based monitoring. While predictive maintenance utilizing Supervisory Control and Data Acquisition (SCADA) data shows great promise, existing data-driven models frequently struggle with concept drift, fleet-wide scalability, and the practical translation of theoretical accuracy into actionable industrial alarms. This thesis addresses these critical gaps by proposing a hybrid predictive maintenance framework that integrates a Long Short-Term Memory (LSTM) deep learning architecture with Statistical Process Control (SPC) methodologies. Operating under a Normal Behavior Modeling (NBM) paradigm, an optimized two-layer LSTM is trained exclusively on preprocessed, healthy SCADA data to function as a thermodynamic "Digital Twin" of a source wind turbine. To detect developing anomalies, the squared prediction residuals are continuously evaluated using three complementary SPC tools: Shewhart X charts, Cumulative Sum (CUSUM) charts, and a Fault Prognosis Indicator (FPI) utilizing an Exponentially Weighted Moving Average (EWMA). Tested on multi-year SCADA data from the onshore Phuong Mai 3 Wind Farm, the framework successfully identified thermal anomalies associated with internal cooling fan faults and converter cabinet overheating. The SPC integration effectively filtered out transient operational weather noise, providing substantial predictive lead times ranging from 18.5 hours to over 200 days prior to actual SCADA alarms. Economically, these extended lead times provide operators with the crucial ability to transition from reactive interventions to cost-effective scheduled maintenance. Furthermore, cross-turbine validation demonstrated successful model transferability to unseen turbines within the same fleet. While the study acknowledges theoretical limitations in transferring models across distinct geographical climates or heterogeneous hardware, the within-fleet scalability offers wind farm operators a robust, interpretable early-warning system without the need for computationally expensive, unit-specific retraining.

This thesis investigates the impact of the temporal resolution of photovoltaic (PV) generation and load data on the calculated hosting capacity of low-voltage (LV) distribution network. With the increasing integration of distributed PV systems, accurate hosting capacity assessment has become essential for ensuring reliable grid operation and avoiding technical violation such as overvoltage, thermal overloading, and reverse power flow. Traditional hosting capacity studies often rely on coarse temporal resolutions such as hourly data, which may fail to capture short-term fluctuations in pv production and electricity demand. To address this limitation, this study evaluates the influence of multiple temporal resolutions, including 60-minute, 10-minute, 5-minute, and 1-minute datasets, on hosting capacity estimation. A detailed LV distribution network model consisting of multiple building types connected to a single transformer is implemented using the pandapower simulation framework. Hosting capacity is determined using three deterministic approaches: Forward, Backward, and Even methods. In addition, a modified even method is developed, where the utility factor (UF) increases simultaneously across all buses while freezing at nodes that violate operational constraints, allowing further penetration at non-critical locations. Time-series simulations are performed by applying power flow analysis at each time step, considering voltage limits, transformer loading, and line loading constraints. High-resolution load profiles are generated from hourly data using the Ornstein-Uhlenbeck stochastic process, while PV production is based on measured one-minute data.

This thesis demonstrated that integrated and physically informed machine learning can be used to simultaneously predict HTL product yield distribution and biocrude quality across diverse biomass classes and operating conditions. A harmonized multi-source dataset containing 4,716 records from nine published studies was developed through systematic filtering, standardization, and biomass categorization, enabling modelling across multiple feedstock types including microalgae, lignocellulosic biomass, sewage sludge, food waste, macroalgae, and mixed wastes. The developed framework successfully predicted ten HTL product descriptors from a common input space comprising feedstock composition and operating conditions. The final SubA model achieved R² = 0.818 for biocrude yield prediction. For four-phase yield prediction, the category-aware SubB framework improved from a pooled mean R² of 0.804 to 0.824 after introducing the soft-constrained mass-balance correction layer, while simultaneously reducing physically inconsistent yield sums. For oil-quality prediction, the baseline SubC framework achieved a pooled mean R² of 0.704, which further improved to 0.706 after applying the elemental-closure constraint strategy while enforcing exact CHNO normalization. These results demonstrate that incorporating physics-based constraints can improve the robustness and physical consistency of HTL surrogate predictions. An additional contribution of this work is the development of the elemental-to-biochemical translation (E2B) module, which enables operation of the full prediction pipeline using elemental composition alone when detailed biochemical composition is unavailable. This substantially improves the practical applicability of the framework for literature-derived and industrially relevant biomass datasets. The SHAP- and PDP-based explainability analyses further confirmed that the surrogate models learned chemically meaningful and literature-consistent relationships. Lipid-rich feedstocks promoted biocrude formation, temperature enhanced gas-phase production and phase redistribution, and feedstock nitrogen strongly influenced oil nitrogen composition. The agreement between the explainability results and established HTL reaction chemistry increases confidence that the models captured physically interpretable thermochemical behavior rather than purely statistical correlations. Overall, this study demonstrates the feasibility of unified, multi-output, and physically informed surrogate modelling for HTL systems. The framework establishes a modular foundation for future developments involving catalytic HTL, uncertainty quantification, and integration with techno-economic or life-cycle assessment workflows for sustainable biomass conversion systems.