# Sequential model predictive direct speed control of PMSM

## TL;DR

This paper introduces an improved control strategy for permanent magnet synchronous motors that reduces instability and improves performance.

## Contribution

An enhanced sequential direct speed predictive control method is proposed and experimentally validated for PMSM drives.

## Key findings

- The original sequential direct speed control shows good dynamic performance but causes instability and current ripples.
- The enhanced method effectively suppresses instabilities and improves speed response.
- Experimental validation was conducted using a 1.1 kW PMSM and the OP 5600 platform.

## Abstract

Finite control set model predictive control (FCS-MPC) has emerged as a powerful strategy for permanent magnet synchronous motor (PMSM) drives. However, its performance strongly depends on appropriately chosen weighting factors, which directly affect control quality and, in some cases, may even lead to instability. Despite the crucial role of weighting factors, there is no systematic or generally accepted procedure for selecting their values, which limits the robustness and practical applicability of conventional FCS-MPC methods. To overcome this limitation, this paper presents the experimental validation of a sequential direct speed predictive control strategy for PMSM. The individual cost functions are evaluated sequentially, thereby tuning is simplified and weighting factors are reduced. Experimental results show that the original version of sequential direct speed control, as proposed in the literature, exhibits promising dynamic performance but suffers from instability and current ripples under certain conditions. To address these issues, an enhanced version of the sequential direct speed predictive control is proposed in the paper. It effectively suppresses instabilities and enhances the speed dynamic response of the drive. The proposed approach was experimentally validated using the OP 5600 rapid control prototyping platform running RT-LAB software and a 1.1 kW PMSM machine.