Abstract: When the Particle Swarm Optimization（PSO） algorithm is easy to concentrate near the optimal particle, the particle motion ability is lost and the population is stagnant, so the optimization effect is not ideal. In this case, a new hybrid optimization algorithm is developed to solve these problems by combining the aggregation and foraging behavior of the Artificial Fish Swarm Algorithm（AFSA） with the particle swarm optimization algorithm. Finally, simulation experiments show that the hybrid optimization algorithm has excellent performance in the optimization of high-dimensional functions.

Key words: optimization of high dimensional function, particle swarm optimization, artificial fish swarm algorithm, hybrid optimization algorithm

摘要： 基本粒子群算法（PSO）在面对高维多极值函数优化的问题时粒子容易快速集中到最优粒子附近，导致粒子运动能力丧失，种群陷入停滞，因此寻优效果并不理想。针对这种情况，通过引入人工鱼群算法（AFSA）中的聚群和觅食行为与粒子群算法相结合形成一种新的混合优化算法来解决这些问题。最终通过仿真实验证明该混合优化算法在面对高维函数的优化问题上具有优秀的寻优能力。

关键词: 高维函数优化, 粒子群算法, 人工鱼群算法, 混合优化算法