Minimizing vibration in coupled electro-mechanical systems via motor control

Abstract

In a rapidly electrifying world, motor systems of different sizes, loads and speeds are implemented in a wide range of products, vehicles and industries. Permanent magnet synchronous motor (PMSM)-based systems, however, have to deal with the inherent cogging torque, a phenomenon caused by the interactions of magnetic flux within the motor. This undesired property causes ripples in the output torque and speed, and many methods have been proposed to mitigate this disturbance, both through structural design and through motor control. A lot of motor control methods, however, are complex in development and implementation, and may not be suitable for low-cost motor systems with underpowered control hardware. Simulation of cogging torque in existing simulation software is also not a straightforward process, requiring custom solutions or convoluted workarounds.
This project aims to develop simple cogging torque mitigation methods, both easy in development and deployment. It also aims to streamline cogging torque analysis and simulation.
The first aim was achieved by implementing a static biquad digital filter within the existing field-oriented control (FoC) system. Through static filtering of the q-axis current, cogging torque and speed ripples were mitigated effectively within a simulated environment. The simple structure of this approach offered ease of tuning with only one required parameter, the cut-off frequency, and could be implemented in low-computational-power systems with low-cost hardware, as the FoC strategy only requires minimal changes.
By taking inspiration from active noise control (ANC) acoustic systems, a FxLMS hybrid filtering approach was proposed, which successfully resolved the shortfalls of the static filter approach. Through the use of static filtering and active anti-noise generation, q-axis currents were injected with destructive interference, and residual cogging torque components were successfully mitigated, with very little impact to the system’s speed response. This approach was also simple in nature, requiring no knowledge of the system parameters and implemented with a simple C-script.
For the second project aim, the cogging torque of a chosen motor was derived from finite element analysis (FEA), then compared with the measured cogging values. These data were then implemented in PLECS to validate the effectiveness of the proposed mitigation approaches. Through the use of a C-script, a platform-agnostic cogging torque integration method is proposed for streamlined analysis of cogging torque.

Date of Award	18 Jul 2026
Original language	English
Awarding Institution	University of Nottingham
Supervisor	Dunant Halim (Supervisor), Giampaolo Buticchi (Supervisor) & Adam Rushworth (Supervisor)