A nonlinear full model of switched reluctance motor with artificial neural network

Saliency and nonlinear magnetization characteristics of switched reluctance motor (SRM) are causes for its poor performance regarding torque ripples. This paper presents an innovative methodology based on a water cycle algorithm (WCA) for SRM speed control with minimum torque ripples and maximum torque per ampere. WCA enhances the dynamic and steady-state performance of this motor by optimizing a multiobjective function consisting of three parts. The first part is the integral time absolute error (ITAE) of motor speed with respect to a desired reference speed, the second part is an index for a motor torque smoothness factor (TSF), and the third part is the torque per ampere ratio (TAR). The desired output of WCA is the appropriate gains of the PI speed controller and optimum turn on and off angles of the H-bridge driving circuit. The results obtained from WCA are used to construct an adaptive controller-based artificial neural network (ANN) to automatically tune the gains of the PI speed controller and the turn on and off angles. The ANN controller ensures superior steady-state and dynamic motoring operation over a wide range of speed commands and different loads. The results of WCA are verified by other well-known meta-heuristic optimized techniques, such as the Genetic algorithm (GA). The suggested method exhibits an excellent speed transient response in addition to minimum torque ripples and maximum TAR.

show abstract

“…The nonlinear model for 6/4 SRM is built in MAT-LAB/SIMULIK based on the following set of differential equations [8][9][10][11][12][13][14][15].…”

Section: Mathematical Model and Driving System Of Srmmentioning

confidence: 99%

“…Linear models in Refs [6,7] are readily constructed and simulated. In contrast, a large number of experimental tests are required to construct nonlinear models [8][9][10][11][12][13][14][15], or using a finite element method (FEM) analysis [16][17][18]. Some configurations of converters used in SRM drives are presented in Ref.…”

Section: Introductionmentioning

confidence: 99%

Optimal dynamic and steady‐state performance of switched reluctance motor using water cycle algorithm

El-Hay

Elkholy

2018

IEEJ Transactions Elec Engng

View full text Add to dashboard Cite

show abstract

“…Chapter 2 of reference [2] is devoted to the analytical calculation of the SRM flux linkages. Artificial neural networks (ANNs) are used to model reluctance machines in [3], [4] and [6]. A nonlinear model of a synchronous reluctance motor was defined by the magnetic flux density along the air gap in [3].…”

Section: Introductionmentioning

confidence: 99%

“…In [4], the motor parameters are estimated through the application of ANN. A model for estimating concatenated flux and torque as a function of stator current and rotor position is given in [6]. Already article [5] proposes a way to calculate the torque per SRM phase using ANN to interpolate motor magnetization data, measured without blocking the rotor, thus reducing costs.…”

Section: Introductionmentioning

confidence: 99%

Switched Reluctance Machine Modeling through Multilayer Neural Networks

Mamede¹,

Camacho²,

Araújo³

2024

RE&PQJ

View full text Add to dashboard Cite

The work deals with the application of artificial neural networks (ANNs) in the modeling of switched reluctance machines (SRMs). The performance of a SRM is determined by its geometry, materials used and levels of excitation. In this way, this work investigates the influence of the stator and rotor back iron thickness in the performance of SRM. A multilayer neural network is proposed to learn the nonlinear characteristics of the motor. Data of flux linkages and torque are obtained through simulations of finite elements and used for ANN training. The algorithm developed in Octave allows the user to adjust the network parameters. The results presented confirm the feasibility of using ANN to establish a predictive model of SRM performance, thus enabling further investigation in the future.

show abstract

“…Some recent publications have used different neural network methods to develop a non-linear SRM [2,4,13,20,21,25,26,28,29]. All these above methods have either faster computational speed or good accuracy but not both.…”

Section: Introductionmentioning

confidence: 99%

Non-linear flux linkage modeling of switched reluctance machine using MVNLR and ANFIS

Susitra

Paramasivam²

2014

Journal of Intelligent &Amp; Fuzzy Systems

View full text Add to dashboard Cite

A Non-linear flux linkage model of a Switched reluctance Machine (SRM) has been developed using two different real time applicable modelling techniques and presented in this paper. The techniques are based on Multivariate nonlinear regression (MVNLR) and adaptive neuro fuzzy inference system (ANFIS). The best features of MVNLR and ANFIS are utilized in this research to develop the computationally efficient flux linkage model for SRM. Mathematical models for the phase flux linkage ψ(i, θ) using MVNLR and ANFIS have been successfully arrived, tested and presented for various values of phase currents (Iph) and rotor positions (θ) of a non linear SRM. It is observed that MVNLR and ANFIS are highly suitable for flux linkage ψ(i, θ) modelling of SRM which is tested to be in good agreement with the training data used for modelling.

show abstract

A nonlinear full model of switched reluctance motor with artificial neural network

Cited by 27 publications

References 13 publications

Optimal dynamic and steady‐state performance of switched reluctance motor using water cycle algorithm

Optimal dynamic and steady‐state performance of switched reluctance motor using water cycle algorithm

Switched Reluctance Machine Modeling through Multilayer Neural Networks

Non-linear flux linkage modeling of switched reluctance machine using MVNLR and ANFIS

Contact Info

Product

Resources

About