基于多Agent遗传算法的水电厂机组优化组合研究与实现

摘要

Based on multi-Agent system theory, a MAS based system for unit optimization combination is built.For the optimization problem mentioned above, a novel MAS based genetic algorithm is proposed, which solves the efficiency problem for genetic algorithm with large-scale computation.As far as the nonlinear water consumption characteristic of generating units is concerned, the BP neural network which can be managed and maintained by Agent is proposed.Based on multi-agent middle ware JADE platform which is fully compliant with FIPA, a detailed implementation system is presented.The simulation results exhibit the reasonability and feasibility of the model proposed in this paper.This work is supported by National High Technology Research and Development Program(863)(No.2007AA05Z428).%基于多Agent系统理论,构建了一个多Agent机组优化组合系统.对于问题优化模型,提出了一种新颖的多Agent遗传算法,解决了大规模遗传算法的效率问题.对于机组的非线性耗水量特性,提出了一种由Agent动态地管理与维护的神经网络.基于与FIPA兼容的多Agent中间件JADE平台,给出了一个详细的具体实施系统.仿真结果验证了所提出的优化模型与实施方案的合理性和可操作性.