基于改进粒子群算法的BP网络在DTC系统中的转速辨识

Rotor Speed Identification on DTC System Based on BP Neural Network of Improved PSO Algorithms

利用改进粒子群优化(PSO)算法优化BP神经网络的权值和阈值,有效地解决了BP算法易陷入局部极小值的缺点,能更快速的实现收敛,不仅具有广泛的映射能力,还明显提高了运算效率。通过对直接转矩控制(DTC)系统进行MATLAB/SIMULINK仿真研究,结果表明:基于PSO-BP神经网络构造的速度辨识器具有良好的辨识效果。

To optimize the parameters of BP neural network,a modified particle swarm optimization （PSO） algorithm can achieve convergence faster and is effective to solve the defect that other BP algorithms easily plunge into local solution. With comprehensive mapping ability,it also promotes the efficiency visibly.The simulation of Direct Torque Control （DTC） system based on PSO-BP neural network is performed using MATLAB/SIMULINK. The result proves that the performance of rotor speed identification is satisfactory.