Robust Model-Free Predictive Current and Speed Control of PMSM Drive Based on Nonlinear Disturbance Observer

Abstract

 Model predictive control has recently attracted the attention of researchers in control of PMSM drives due to advantages such as simple concepts and fast dynamics. However, it is highly sensitive to changes in motor parameters and unmodeled uncertainties. To solve this problem, a model-free predictive control based on an ultra-local model is proposed in this paper, which uses only the input and output of the system based on nominal parameters. To enhance the drive performance and simplification of implementation, current control, and speed control are performed simultaneously based on the optimization of a cost function without using any PI speed or PI current controller. In the proposed method named ‘Model-Free Predictive Current and Speed Control" (MFPCSC), to reduce the sensitivity to model parameters and uncertainties, a nonlinear disturbance observer (NDO) is used, which accurately estimates all uncertainties and the effect of parameters’ change. The proposed MFPCSC-NDO control method has good behavior in tracking the speed and current reference values and has high robustness to parameter changes. The performance of the proposed method is compared with two methods of model-based predictive current control (MBPCC) and model-free predictive current control (MFPCC) by simulation and its behavior is confirmed with experimental results.