改进人工鱼群算法及其在时滞系统辨识中的应用

针对人工鱼群算法(AFSA)存在收敛速度慢和寻优精度低等问题,本文提出了一种改进人工鱼群算法(IAFSA).该算法中的人工鱼能够根据鱼群当前状态调整自身的视野和步长来平衡局部搜索和全局搜索.此外,算法中还加入了引导行为,即人工鱼在觅食行为未发现更优的位置时,当前人工鱼向最优人工鱼移动一步.仿真结果表明,改进人工鱼群算法在收敛速度、寻优精度和克服局部极值等方面有很大优势.本文将改进鱼群算法应用时滞系统的辨识中,辨识结果表明改进算法能获取被控对象的精准数学模型,并具有较强的抗干扰能力.

An improved artificial fish swarm algorithm and its application on system identification with a time-delay system

To remedy the low convergence rate and low optimization accuracy of the artificial fish swarm algorithm(AFSA),an improved artificial fish swarm algorithm(IAFSA)was proposed.In the improved algorithm,the artificial fish could adjust the vision and step and form a balance between the local search and global search by identifying the actual condition.Furthermore,when the artificial fish in the foraging behavior does not find a better position than the current location,it steps forward to the optimal artificial fish by introducing the guide behavior to improved algorithm.The results indicate that the improved algorithm has advantages such as convergence rate,optimization accuracy,and anti local extremum value.The improved algorithm was applied to the system identification with the time-delay model.This algorithm can obtain a precise mathematical model of the controlled object and acquire great identification accuracy in the case of external interference.