Abstract: Grid task allocation is a typical NP complete problem. According to the essence of grid and based on Particle Swarm Optimization algorithm, this paper proposes a new algorithm called Grid-based Hybrid Particle Swarm Optimization（GHPSO）. This algorithm transforms and redefines the problem’s resolution space to make it more suitable to the problem-solving environment of PSO algorithm, achieves the optimal allocation of grid resources. The simulation results compared with the Discrete Particle Swarm Optimization algorithm and  Genetic Algorithm show that this algorithm has better performance.

Key words: grid task scheduling, Grid-based Hybrid Particle Swarm Optimization（GHPSO）, problem’s resolution space transformation

摘要： 网格任务分配是一个NP难问题，结合微粒群优化（Particle Swarm Optimization，PSO）算法，和网格自身的特性，提出了基于网格的混合微粒群算法。算法对问题的解空间进行变换、重定义，使之更加符合PSO算法的求解环境，实现了网格资源的优化分配。与离散微粒群（DPSO）算法和遗传算法进行了仿真比较，结果表明，新的PSO算法具有较好的性能。

关键词: 网格任务分配, 微粒群优化算法, 解空间变换