基于混合生物地理学算法的非线性系统辨识

Identification of Nonlinear System Based on Hybrid Biogeography-Based Optimization Algorithm

针对非线性系统辨识问题,由于传统辨识方法存在精度低收敛慢等缺点,提出了一种采用混合生物地理学算法的非线性系统辨识方法。混合算法是在对生物地理学算法进行改进的基础上与差分进化算法相结合,通过适当地融合具有不同搜索能力的优化算法,使得混合算法的开采能力和探索能力得到更好的增强和平衡。通过对Wiener模型进行参数辨识,并与生物地理学算法和差分进化算法进行比较,仿真结果表明,利用混合生物地理学算法能够提高辨识精度并获得良好的辨识效果,验证了混合算法的有效性和可行性。

A Hybrid Algorithm （IDEBBO） is proposed based on Biogeography-Based Optimization for nonlinear identification problem. The hybrid algorithm is Improved Biogeography-Based Optimization combined with Differential Evolution Algorithm. By suitably fusing several optimization methods with different searching mechanisms, the exploration and exploitation abilities of the hybrid algorithm can be enhanced and well balanced. According to parameters identification of the Wiener model and compared with Biogeography-Based Optimization （BBO） and Differential Evolution Algorithm （DE）, the simulation results show that using IDEBBO algorithm can improve the identification accuracy and get good recognition results, which verifies the effectiveness and feasibility of the IDEBBO algorithm.