中文摘要：

粒子群优化算法（PSO）是一种群体智能计算方法,该算法精度高,收敛速度快,但在优化多峰函数的时候容易陷入早熟。加入启发性变异机制,可以在不破坏原算法高速收敛性质的同时,扩展算法的有效搜索区域。经过13个经典函数的测试证明,带启发性变异的粒子群优化算法（HMPSO）速度比原算法速度更快,精度更好,且不容易陷入局部最优。与其它带变异的粒子群优化算法相比,该算法收敛更快,在一些问题上有一定的精度优势。

英文摘要：

Particle swarm optimization （PSO） is an algorithm of swarm intelligence, which performs well in function optimizing area, with its high precision and fast convergence. However, PSO may be premature sometimes, especially for multimodal functions. Combined with heuristic mutation, PSO can expand the algorithm＇s search place to avoid being trapped. By experiments based on 13 classic benchmarks function, it is proved that the heuristic mutation particle swarm optimization （HMPSO）, which is seldom trapped, is faster and more precision than the standard one. Compared to other PSOs with mutation, HMPSO also converges faster, and gets some advantages.