Design of a Fuzzy Linear Quadratic Regulator for the Speed Control of SPMSM

Surface-Mounted Permanent Magnet Synchronous Machines (SPMSM) operate in wide torque-speed envelopes which impose the necessity of designing an effective control system to prevent performance degradation. However, the conventional Cascade PI (CPI) control structure is still utilized in the vast majority of SPMSM applications even though it becomes inadequate against emerging nonlinearities in different operating regions. In this paper, we propose a Fuzzy Linear Quadratic Regulator (F-LQR) for the speed control of SPMSMs to maintain robustness while improving the speed control performance. We designed a Single input Fuzzy Logic Controller (SFLC) to manipulate the error compensation dynamics of the LQR to improve reference tracking and disturbance rejection performances. We employed a geometrical analysis approach when designing the SFLC which enables us to explicitly shape the fuzzy mapping with just two parameters. The practical design goals considered in the design phase of LQR and SFLC are determined by taking into account the physical limitations of the inverter board as well as the system dynamics of the SPMSM. We experimentally analyzed and compared the performances of the proposed F-LQR, LQR, and CPI on the real-world SPMSM test setup with two inclusive test scenarios. The comparative test results revealed that the proposed F-LQR improves the speed control performance while maintaining the robustness of the control system when compared to LQR and CPI.