Marine asynchronous propulsion motor parameter identification using dynamic particle swarm optimization

Siyuan Liu,Yancheng Liu,Chuan Wang,Junjie Ren

DOI: https://doi.org/10.4028/www.scientific.net/AMR.860-863.2211

2014-01-01

Advanced Materials Research

Abstract:This paper proposes a new application of dynamic particle swarm optimization (PSO) algorithm for parameter identification of vector controlled asynchronous propulsion motor (APM) in electric propulsion ship. The dynamic PSO modifies the inertia weight, learning coefficients and two independent random sequences which affect the convergence capability and solution quality, in order to improve the performance of the standard PSO algorithm. The standard PSO and dynamic PSO algorithms use measurements of the mt-axis currents, voltages of APM as the inputs to parameter identification system. The experimental results obtained compare the identified parameters with the actual parameters. There is also a comparison of the solution quality between standard PSO and dynamic PSO algorithms. The results demonstrate that the dynamic PSO algorithm is better than standard PSO algorithm for APM parameter identification. Dynamic PSO algorithm can improve the performance of ship propulsion motor under abrupt load variation.