Fuzzy logic control in linear switched reluctance motors

Abstract

During the last decade, the Linear Switched Reluctance Motor (LSRM) has become popular in motor industry due to its structural simplicity, robustness, high power density, low cost and applicability in high temperature environment. However, its significant torque ripple from the flux nonlinearity creates difficulty on precision motion control. This research project aims to develop a fuzzy logic control system to provide a high performance on LSRM based on switched reluctance actuator technology. The controller is designed to replace the conventionally used numerical approach on handling model nonlinearity problem. For the preparation of high speed motion tests, the rigidity of the LSRM prototype is improved with advance mechanical features and special assembly procedures. This practice is promoted in industry to enhance the motor productivity. The LSRM prototype is firstly investigated to study its force and current relationship. To shorten the real time computation time and simplify the control plant, a table look-up fuzzification is developed. With the help of software, LSRM motion tests are simulated before real experiment. The significant improvement on position control strongly proves the success of the proposal. After that, the experimental result applying on the real prototype closely matches the simulation result. In order to enhance the LSRM robustness and the position tracking responses, another fuzzy logic controller is newly designed and implemented to supervise the traditional Proportional-Differential (PD) control parameters. Combining the experimental results on inner control loop with current force relationship and outer control loop on PD parameter supervision, the fuzzy logic improves motor performance at least 20%. With the successful application of the fuzzy logic, the LSRM system in this project is very robust and capable to provide a high precision motion performance. Based on the features of magnet-free and flexibility under high temperature, LSRM is believed to become popular in precision machine industry with fuzzy logic controllers.