改进布谷鸟算法的永磁同步电机多参数辨识

Multi-Parameter Identification of PMSM with Improved Cuckoo Search

为解决PI控制器依赖于准确的电机参数且传统PI控制器效率不高、稳态误差大的缺点,采用模糊PI控制器替代永磁同步电机矢量控制系统中的传统PI控制器,采用改进的布谷鸟算法,以实现对电机多参数的精确辨识。通过Logistic-Tent映射构造随机衰减惯性权重,并采用高斯分布改进步长因子和指数函数构造动态发现概率,采用多样性维持策略,在一定程度上改善了标准布谷鸟算法收敛速度慢和后期易陷入局部最优解的问题,从而使改进的布谷鸟算法对电机多参数的辨识更加精确。改进的布谷鸟算法只需采集模糊PI控制下电机的转速,定子电压和电流信号就可精确辨识电机参数。仿真实验表明:改进的布谷鸟算法对多参数的辨识收敛速度更快精度更高,对改善电机性能具有重要意义。

In order to solve the shortcomings of PI controller relying on accurate motor parameters and the traditional PI controller with low efficiency and large steady-state error,the fuzzy PI controller was used to replace the traditional PI controller in the vector control system of PMSM,and an enhanced cuckoo algorithm was proposed to realize the multi-parameter precise identification of motors.The random attenuation inertia weights were constructed by Logistic-Tent mapping,the Gaussian distribution was used to improve the growth factor and exponential function to construct the dynamic discovery probability,and finally the diversity maintenance strategy was adopted,which improves the slow convergence speed of the standard cuckoo algorithm and the problem of falling into the local optimal solution in the later stage,so that the enhanced cuckoo algorithm enable more accurately identify the multi-parameters of the motors.The improved cuckoo algorithm only needs to collect the speed of the motor under the control of fuzzy PI,and the stator voltage and current signals could accurately identify the motor parameters.The simulation experiments indicate that the enhanced cuckoo algorithm has higher accuracy and faster convergence speed for multi-parameter identification,which is of great significance for improving motor performance.