基于苍狼算法的感应电机改进模型参数辨识

Grey Wolf Optimizer for Parameters Identification of Induction Motor with Improved Model

为解决感应电机高性能控制中参数估计不准的问题,提出基于苍狼算法的感应电机参数辨识方法。该优化算法是一种新颖的元启发式算法,运用起来简单、灵活,需调节参数少。考虑到两种经典动态数学模型对不同参数辨识精度影响不同,提出了一种改进的感应电机参数辨识模型。相对于经典模型,仿真表明提出的模型使电阻特别是定子电阻的辨识效果得到较大提升,证明改进模型的有效性。在改进的模型下,将该算法与粒子群算法、遗传算法进行电机参数辨识的对比实验。实验结果表明苍狼算法具有较高的辨识精度,证明应用该算法辨识感应电机参数的可行性。

According to the problems of inaccurate parameters estimation of the induction motor in high-performance control, a grey wolf optimizer was used to identify the parameters of the induction motor. Grey wolf optimizer is a new meta-heuristic. It is simple and flexible to implement, and has fewer parameters to tune. Considering that two typical dynamic mathematical models have different identification precision on different parameters, the improved identification model of the induction motor was proposed. Compared with typical model, simulation results show that the proposed model obviously improves the identification performance of resistances especially stator resistance, verifying the validity of improved model. The algorithm was compared with particle swarm optimization and genetic algorithm for parameters identification of the induction motor with the improved model. Experimental results show that grey wolf optimizer has higher identification precision, demonstrating that parameters identification of the induction motor based on this algorithm is feasible.