Please use this identifier to cite or link to this item:
http://hdl.handle.net/11375/30118
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Gadsden, S. Andrew | - |
dc.contributor.author | Wu, Yuandi | - |
dc.date.accessioned | 2024-08-30T18:07:50Z | - |
dc.date.available | 2024-08-30T18:07:50Z | - |
dc.date.issued | 2024 | - |
dc.identifier.uri | http://hdl.handle.net/11375/30118 | - |
dc.description.abstract | Physics-informed neural networks (PINNs) provide an alternative to traditional solvers for differential equations, specifically in tasks such as system identification and inverse problems. This study applies the recently popularized paradigm of PINNs for system identification and surrogate modelling in Magnetorheological (MR) dampers. A task hindered by the nonlinear behaviour, hysteresis effects and the presence of latent variables in certain interpretations of the MR damper dynamic model. An experimental setup was developed to analyze empirical data collected from MR dampers, incorporating a voltage-controlled MR damper, with motions actuated through a linear actuator, and various sensors for capturing damping forces and motion profiles. A data collection pipeline was developed and allows for synchronous data collection from a multitude of devices, and database storage. Collected data is useful for validating developed models, as well as setting up a foundation for experimental validation of novel methods in future work. A literature review was performed, highlighting the limitations of existing models and the potential of PINNs, cases of deployments, and innovations by authors within the literature. The research additionally constructs and validates a discretized state space model from estimated parameters. Overall, this research demonstrates proof of concept for the application of PINNs in mechanical systems modelled by differential equations. Results demonstrated satisfactory accuracy in parameter identification, with implications for system behaviour prediction, demonstrating the potential and limitations of PINNs in this context. | en_US |
dc.language.iso | en_US | en_US |
dc.subject | System Identification | en_US |
dc.subject | Machine learning | en_US |
dc.subject | Deep learning | en_US |
dc.subject | Physics-informed neural networks | en_US |
dc.title | PHYSICS-INFORMED MACHINE LEARNING FOR MODELING AND ANALYSIS OF MECHANICAL SYSTEMS | en_US |
dc.type | Thesis | en_US |
dc.contributor.department | Mechanical Engineering | en_US |
dc.description.degreetype | Thesis | en_US |
dc.description.degree | Master of Applied Science (MASc) | en_US |
dc.description.layabstract | This study presents an implementation of Physics-Informed Neural Networks to model complex mechanical systems with hidden variables, exemplified by Magnetorheological (MR) dampers. This research focuses on the development of an experimental setup to analyze real-world data, emphasizing data collection for validation of developed methods. Through Physics-Informed Neural Networks, known dynamics of the system may be incorporated into the training process of machine learning algorithms, allowing for predictions that adhere to physical principles. This is employed to resolve many of the difficulties associated with accurately modelling an MR damper. Overall, this research demonstrates proof of concept for the application and use case of PINNs in mechanical systems. | en_US |
Appears in Collections: | Open Access Dissertations and Theses |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
wu_yuandi_finalsubmission202408_mechanicalengineeringmasc.pdf | 20.17 MB | Adobe PDF | View/Open |
Items in MacSphere are protected by copyright, with all rights reserved, unless otherwise indicated.