中文摘要：

针对基本鱼群算法盲目搜索的问题,提出一种新的基于知识的带变异算子的人工鱼群算法。利用文化算法的框架,将鱼群算法嵌入到种群空间当中,构造适用于文化鱼群算法的新的影响函数。同时应用信念空间中的规范知识和情境知识通过影响函数指导人工鱼群算法中的进化步长和方向。通过高维多峰函数检验新算法的性能,最后将新算法应用于一台内置有执行器的鼠笼电机系统的参数辨识问题,得到了参数化的执行器-转子模型。仿真结果表明新算法与基本鱼群算法相比性能显著提高,并且能够有效地解决工程优化问题。

英文摘要：

A knowledge-based artificial fish-swarm optimization algorithm （AFA） with crossover operator （CAFAC） is proposed in this paper to combat with the blindness of search of the original AFA. The AFA was embedded into the population space based on the cultural framework. The influence function was constructed for the CAFAC. The normative knowledge and the situational knowledge stored in the belief space were utilized to guide the step size as well as the direction of the AFA evolution. High-dimensional and multi-peak functions were employed to investigate the proposed algorithm. Then the CAFAC was em- ployed to identify the parameters of a two-pole cage induction motor equipped with a built-in force actua- tor. A parametric model of the actuator-rotor system was obtained. Numerical simulation results demon- strate that the CAFAC can outperform the regular AFA, and shows effectiveness dealing with engineering optimization problems.