Abstract: To address the competition problem in grid resource allocation，a resource allocation method is proposed based on dynamic mechanism of evolutionary game.The replicated dynamic mechanism is used to produce evolutionary stable point of strategy selection of grid consumer.Then，grid consumers learn and adjust strategies through repeated games.Particularly，the effects of four typical valuation functions on evolutionary stable point are discussed.Finally，the performance of the evolutionary algorithm is evaluated through grid simulator.The results show that the proposed evolutionary algorithm is convergent and generates better utility results compare to classic algorithm.

Key words: grid, resource allocation, replicated dynamic, valuation function

摘要： 针对网格资源分配中的竞争问题，提出了一种利用进化博弈的动态机制研究资源分配的方法。该方法利用复制动态方程求解网格使用者策略选择比例的进化稳定点，通过反复博弈使得网格使用者学习并调整出价策略，并讨论了四种典型的使用者评估函数对进化稳定点的影响。最后利用网格模拟器进行了实验评估，结果表明提出的进化博弈方法是收敛的，且在网格使用者的总体效用方面优于传统算法，从而实现了网格资源的优化分配。

关键词: 网格, 资源分配, 复制动态, 评估函数