Gå til hovedinnhold
0
Jump to main content

Online Condition Monitoring of Electric Powertrains using Machine Learning and Data Fusion

Jagath Sri Lal Senanayaka of the Faculty of Engineering and Science has submitted his thesis “Online Condition Monitoring of Electric Powertrains using Machine Learning and Data Fusion” and will defend the thesis for the PhD-degree Wednesday 20 May 2020. (Photo: Private)

All the proposed fault diagnosis schemes are tested with experimental data, and key features of the proposed solutions are highlighted with comparative studies.

Jagath Sri Lal Senanayaka

PhD Candidate and Assistant Professor

Jagath Sri Lal Senanayaka of the Faculty of Engineering and Science has submitted his thesis “Online Condition Monitoring of Electric Powertrains using Machine Learning and Data Fusion” and will defend the thesis for the PhD-degree Wednesday 20 May 2020.

He has followed the PhD-programme at the Faculty of Engineering and Science with Spesialisation in Mechatronics.

Summary of the Thesis by Jagath Sri Lal Senanayaka

Online Condition Monitoring of Electric Powertrains using Machine Learning and Data Fusion

Safe and reliable operations of industrial machines are highly prioritized in industry. Typical industrial machines are complex systems, including electric motors, gearboxes and loads.

Catastrophic failures and productivity losses

A fault in critical industrial machines may lead to catastrophic failures, service interruptions and productivity losses, thus condition monitoring systems are necessary in such machines.

The conventional condition monitoring or fault diagnosis systems using signal processing, time and frequency domain analysis of vibration or current signals are widely used in industry, requiring expensive and professional fault analysis team.

Further, the traditional diagnosis methods mainly focus on single components in steady-state operations.

Under dynamic operating conditions, the measured quantities are non-stationary, thus those methods cannot provide reliable diagnosis results for complex gearbox based powertrains, especially in multiple fault contexts.

Four main research topics

In this dissertation, four main research topics or problems in condition monitoring of gearboxes and powertrains have been identified, and novel solutions are provided based on data-driven approach.

  • The first research problem focuses on bearing fault diagnosis at early stages and dynamic working conditions.
  • The second problem is to increase the robustness of gearbox mixed fault diagnosis under noise conditions.
  • Mixed fault diagnosis in variable speeds and loads has been considered as third problem.
  • Finally, the limitation of labelled training or historical failure data in industry is identified as the main challenge for implementing datadriven algorithms.

To address mentioned problems, this study aims to propose data-driven fault diagnosis schemes based on order tracking, unsupervised and supervised machine learning, and data fusion.

All the proposed fault diagnosis schemes are tested with experimental data, and key features of the proposed solutions are highlighted with comparative studies.

Opponent ex auditorio:

The chair invites members of the public to pose questions ex auditorio in the introduction to the public defense, with deadlines. Questions can be submitted to the chair Geir Grasmo at e-mail geir.grasmo@uia.no.

The thesis is available here: https://uia.brage.unit.no/uia-xmlui/handle/11250/2653755

Disputation facts:

The Candidate: Jagath Lal Senanayaka (1982, Matara Sri Lanka) BSc in “Electrical and Information Engineering”, University of Ruhuna, Sri Lanka (2007), MBA in “Technology Management”, University of Moratuwa, Sri Lanka (2012). MSC in Renewable Energy from University of Agder (2014). He worked as a Telecommunication Engineer in Sri Lanka from 2007 to 2011

and as a Research Assistant in the University of Agder from 2014 to 2016. Currently, he is a temporary assistant professor with the university of Agder. His research interest covers various areas related to fault diagnosis and control of electric powertrains, renewable energy, data analytics, signal processing, and artificial intelligence.

The trial lecture and the public defence will take place at internet, via the Zoom conferencing app (link below) Wednesday 20 May 2020.

Professor Geir Grasmo, Department of Engineering Sciences, UiA, will chair the disputation.

Trial lecture at 10:15 a.m.

Public defense at 12:15 p.m.

Given topic for trial lecture"Overview on electrical machines technology used in drive trains: challenges and opportunities for machine learning algorithms"

Thesis Title: “Online Condition Monitoring of Electric Powertrains using Machine Learning and Data Fusion”

Search for the thesis in AURA - Agder University Research Archive, a digital archive of scientific papers, theses and dissertations from the academic staff and students at the University of Agder. (Or see link below)

Opponents:

First opponent: Professor Gerard-André Capolino, Université de Picardie Jules Verne, France

Second opponent: D.Sc, Senior Specialist Antti Laiho, EHE Electronics Ltd, Finland

Professor Joao Leal, Department of Engineering Scineces, UiA, is appointed as the administrator for the assessment commitee.

Supervisors were Professor Kjell Gunnar Robbersmyr (main supervisor) and Professor Van Khang Huynh (co-supervisor)