基于TB-PC算法的MISO系统辨识的研究

Research on MISO System Identification Based on TB-PC Algorithm

如何保持MISO系统辨识的精确度、收敛度、耗时以及跳出局部最优解,是当今研究的热点和难点。提出了一种基于协同进化策略和禁忌搜索的蝙蝠和粒子群混合算法(TB-PC),分析蝙蝠算法控制参数,提出了合理的脉冲频度和音强初值,将蝙蝠算法与粒子群算法的优势用协同进化结合起来,并引入了禁忌搜索。通过对四个测试函数和MISO系统实例的辨识仿真,验证了TB-PC算法具有稳定性能好、收敛精度高等优点,对MISO系统有优良的辨识效果。

How to maintain the accuracy, convergence, time consuming and jump out of the local optimal solution of MISO system identification is a hot and difficult problem in the research. It presented a bat algorithm cooperative with particle swarm optimization based on tabu search strategy （TB-PC） , analyzed control parameters, put forward reasonable frequency and initial pulse intensity, combined bat algorithm and particle swarm optimization with co-evolution together and the introduced tabu search. Through the identification and simulation of four test functions and MISO system, the results showed that the TB-PC algorithm has the advantages of good stability, high convergence precision and good identification effect on the MISO system.