Journal of Applied Research in Electrical Engineering

Journal of Applied Research in Electrical Engineering

Constrained MIMO Reference-Based Offline Steady–State (RBOSS) NMPC for PMSM using Voltage Source Inverter

Document Type : Research article

Authors
Department of Electrical and Electronic Engineering, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana
Abstract
This research introduces a constrained MIMO Reference-Based Offline Steady-State (RBOSS) Nonlinear Model Predictive Control (NMPC) approach for Permanent Magnet Synchronous Motors (PMSM) to address the overlooked aspect of starting conditions in current and speed multi-objective controllers. By employing an offline predictive optimisation technique, the study aims to improve the starting condition for the optimisation process, thus facilitating faster convergence. The RBOSS NMPC precomputes steady-state values of PMSM motors across a range of reference speeds, which are then used as starting conditions via a lookup table during online optimisation. The approach significantly reduces computational burden by 30% compared to the usual successive linearization technique; it also enhances performance parameters such as rise time and settling time while optimizing current usage. This method demonstrates potential adaptability to other nonlinear systems, marking a significant advancement in NMPC controller optimisation techniques and suggesting further exploration to expand its applicability.

Highlights

  • Novel RBOSS NMPC Approach: Introduced a Reference-Based Offline Steady-State NMPC to optimize PMSM speed tracking and reduce computational burden by 30%.
  • Enhanced Stability: Achieved reduced steady-state error (0.82%) for improved robustness.
  • Precomputed Steady-State Values: Used offline optimization to provide optimized initial conditions, enabling faster convergence and better performance.
  • Improved Speed Metrics: Delivered faster rise time (0.019%) and settling time (0.04%) compared to successive linearization NMPC.
  • Multi-Objective Optimization: Utilized a receding horizon cost function to eliminate cascaded MPC schemes.

Keywords
Subjects

 
Volume 4, Issue 1
Winter and Spring 2025
January 2025
Pages 76-85

  • Receive Date 15 August 2024
  • Revise Date 16 May 2025
  • Accept Date 02 October 2025