Wahiba Bounoua has followed the PhD Programme at Faculty of Engineering and Science, specialisation in Engineering Sciences, Scientific field of Mechatronics.
- Trial lecture starts at 10:15
- Public defence starts at 12:15
Title of trial lecture: Hybrid semi-empirical modelling: combining physics-inspired/phenomenological and data-driven modelling: Approaches and potential industrial application
Disputation chair: Jan Henriksen
Assessment committee:
- First opponent: Professor Adil Rasheed, NTNU
- Second opponent: Dr. Lidia Auret, Stone Three Digital, Stellenbosch University, Sør-Afrika
- Chair of assessment committee: Professor Souman Rudra, University of Agder
Supervisors in the doctoral work:
- Main supervisor: Førsteamanuensis Muhammad Faisal Aftab, University of Agder
- Co- supervisor: Professor Christian Omlin, University of Agder
Summary of thesis
In the era of Industry 4.0, where automation and digitalisation are transforming industrial landscapes, maintaining optimal control performance in industrial control systems (ICSs) is paramount. These systems are the backbone of efficient, reliable, and safe operations, directly impacting plant profitability, production quality, and environmental sustainability. However, the increasing complexity of ICSs and inconsistent controller installation practices often lead to control loop performance
deterioration, manifesting as oscillations, deviations from set points, and abnormal responses. This necessitates developing and implementing robust controller performance monitoring (CPM) strategies.
This study extensively investigates the challenges and remedies associated with control loop performance deterioration, with a specific emphasis on the detection and analysis of oscillations. Notably, it underscores the criticality of employing advanced signal processing techniques to achieve efficient noise reduction and feature extraction. Among these techniques, online detrended fluctuation analysis (DFA) and improved empirical wavelet transform (IEWT), including its multivariate variant known as the improved extended empirical wavelet transform (IEEWT), hold particular significance. Furthermore, the study highlights the crucial role of datadriven monitoring, utilising machine learning and data analytics, to facilitate realtime anomaly detection and enable early intervention measures.
Looking towards the future, the work identifies key areas for further research and development. These include the development of more sophisticated CPM algorithms for accurate and timely detection of performance issues, the integration of diverse data-driven approaches for improved fault detection analysis regardless of changing process dynamics, and the development of standardised methodologies for implementing advanced monitoring techniques in diverse industrial settings.
By addressing control loop deterioration challenges and implementing advanced monitoring techniques, industries can unlock significant productivity, safety, and sustainability improvements. This will allow ICSs to remain the mainspring of efficient and reliable operations in the ever-evolving landscape of Industry 4.0.
What to do as an online audience member
The disputation is open to the public. To follow the trial lecture and the public defence online, register on Zoom.
We ask online audience members to join no earlier than 10 minutes in advance. After these times, you can leave and rejoin the meeting at any time.
Opponent ex auditorio:
Deadline for the public to pose questions is during the break between the two opponents. Questions ex auditorio can be submitted to Emma Elisabeth Horneman.