基于RAGA的PPPCA模型在水资源配置方案中的应用

Water Resources Allocation Schemes Evaluation Based on Projection Pursuit Principal Components Analysis Model

为了解决日益突出的水资源供需矛盾并得到较为合理的水资源配置方案,将投影寻踪主成分分析模型(PPPCA)与加速遗传算法(RAGA)相结合,即利用PPPCA对配置过程中涉及的指标特征变量较多等问题进行降维处理,采用RAGA解决高维全局寻优问题,并将其应用于台兰河灌区水资源配置方案中。结果表明,采用加速遗传算法与投影寻踪主成分分析模型得出的配置方案合理、可行。

In order to solve the increasingly prominent contradiction between supply and demand of water resources and to get more reasonable water resources allocation schemes evaluation, this paper combines the projection pursuit principal components analysis （PPPCA） model with water resources allocation schemes. By using this model that involves more index characteristic variables to reduce the dimension in the process of water resources allocation problem, real-co- ded accelerating genetic algorithm （RAGA） is adopted to solve high-dimensional global optimization problems. The mod- el can effectively solve multiple correlation among indices in the problem, which achieves the water resources allocation decisions in low dimensional space. Taking water resources allocation of Tailanhe irrigation area for an example, application results show that the allocation scheme obtained by PPPCA with RAGA is reasonable and feasible.