投影寻踪模型在水资源工程方案优选中的应用

Application of Projection Pursuit Model to Opotimal Choice of Water Resources Engineering Schemes

提出了水资源工程方案优选的投影寻踪(PP)模型的新方法。利用PP模型可把方案多维评价指标值综合成一维投影值,投影值越大表示该方案优选,根据投影值的大小就可对方案集进行优选。采用实码加速遗传算法进行PP建模,简化了投影寻踪技术的实现过程,克服了目前投影寻踪技术计算复杂、编程实现困难的缺点,为投影寻踪技术在水资源系统工程中的广泛应用提供了新的有力工具。应用实例的结果说明,直接由样本数据驱动的PP模型用于水资源工程方案优选简便可行,PP模型的投影值比较分散、易于决策,适用性和可操作性强,不需确定评价指标的权重,优选结果较为客观。

A new method, named projection pursuit (PP) model is presented in this paper to optimize choice of water resources engineering schemes. By using PP model, many evaluation indexes values of each scheme can be synthesized projection value with one dimension which indicates advantage of each scheme, so the schemes sample can be optimized choice according to the projection value of each scheme. The modeling of PP based on real coded accelerating genetic algorithm can predigest the realized process of projection pursuit technique, can overcome the shortcomings of large computation amount and difficulty of computer programming in traditional projection pursuit metods, and can give a new trenchancy tool for wide applications of projection pursuit technique to water resources systems engineering. The results of applied samples show that applying PP model driven directly by samples data to optimizing choice of schemes is both simple and feasible, that its projection values are relatively decentralization and profit decision-making, that its applicability and maneuverability are high, and that its results are impersonality because it does not require determining the weights of evaluation indexes values of each scheme.