Artificial Neural Network and Data Dimensionality Reduction Based on Machine Learning Methods for PMSM Model Order Reduction

Abstract

The present paper targets a solution for permanent motor synchronous machine (PMSM) model order reduction (MOR) using artificial neural networks and machine learning techniques for data dimensionality reduction. The neural networks are trained using data obtained from a series of electromagnetic Finite Element Analysis (FEA), conducted in conditions imposed by the data dimensionality reduction method. The workflow proposed to build the PMSM MOR, starts with data generation, goes further to its post-processing, and finishes with the model training and experimental validation. In the study, data dimensionality reduction procedure (adaptive data generation) is performed to increase the computational efficiency, also maintaining the model accuracy. Different data reduction approaches are compared from the computational cost's point of view and their ease of use. The obtained results are compared to those obtained from FEA seeking the best solution for building the dynamic model. The resulting ROM is included in a real-time control prototyping platform to characterize machine's performances. The model accuracy and its usability are proved in a comparative analysis with simulated versus experimental measurements.INDEX TERMS artificial neural networks, adaptive data generation, machine learning, permanent magnet synchronous machine, dynamic simulation, real-time.

An Artificial Neural Network for Compensation of MTPA Algorithm in Permanent Magnet Synchronous Machines

“…This approach is motivated by the increasing complexity of converters, which grows with the number of semiconductors and passive elements used. On the other hand, there are other ANNs that focus on supporting control by adjusting parameters or predicting variables [21][22][23]. The aim is to enhance the characteristics of the control utilized, as the models are generally simplified to facilitate implementation and design stages.…”

Speed-current composite loop SPMSM control based on ADR-SMC

“…However, the PI controller is difficult to maintain high-performance control under com-plex working conditions [4,7]. In order to improve the performance of PMSM control system under uncertain disturbance, such as sliding mode control algorithm (SMC) [8,10], adaptive control algorithm [11,14], neural network control algorithm [15,18], model predictive control algorithm (MPC) [19,21], active disturbance rejection control algorithm (ADRC) and so on are applied to PMSM control system. ADRC has attracted wide attention because it does not require accurate system model and excellent anti-interference ability [22,23].…”

A Real-Time Data Storage Method for AC Servo Motor Drives Using Micro SD Card

“…Furthermore, in recent years, the application of machine learning techniques in motor drive systems has gained significant attention. Machine learning algorithms have shown promise in optimizing motor drive systems, improving performance, and achieving higher levels of efficiency [13], [14], [15], [16]. However, harnessing the power of machine learning requires the collection of a substantial volume of motor control data.…”