Raspberry Pi-Based Rapid Control Prototyping for PMSM Drive System Using RBFNN on OPAL-RT Platform

This study addresses the challenge of improving speed control accuracy and noise immunity in permanent magnet synchronous motor (PMSM) drive systems, which are widely used in industrial applications requiring high drive efficiency. Conventional controllers such as PIDs exhibit limitations in handling nonlinear components, parameter uncertainties, and time-varying dynamics. To overcome these issues, a sliding mode control (SMC) strategy integrated with a radial basis function neural network (RBFNN) is proposed. The RBFNN adaptively estimates and compensates for unknown components in real time, while the overall stability of the system is rigorously proven using Lyapunov theory. The proposed control algorithm is developed in MATLAB/Simulink, validated through real-time simulation using OPAL-RT OP5707XG platform and implemented on Raspberry Pi hardware under the Rapid Control Prototype (RCP) framework. The control structure consists of a speed control loop enhanced by adaptive RBFNN and a flux-driven current control loop. Experimental results under multi-stage variable speed conditions demonstrate that the proposed controller achieves fast dynamic response, minimal steady-state error, reduced overshoot and high robustness to load variations and disturbances.