针对树状管网布置中较多依赖设计人员经验的特点。提出了一种基于整数编码遗传算法的树状管网两级优化方法。第一级优化是根据树状管网单点供水的原则,建立了融合工程设计经验的树状管网优化布置整数编码遗传算法模型,克服了传统二进制编码方法易产生不可行解的问题,可快速寻找出一组符合工程实际情况的管网布置方案。第二级优化在确定管网布置方案组的基础上采用整数编码的遗传算法,以投资最小为目标,建立了管径优化模型与算法。编制了灌溉管网两级优化设计Matlab程序,进行了工程实例验证。与单亲遗传算法(SPGA)和管网布置经验设计方法进行了比较,表明本文提出的基于整数编码的管网优化设计方法可方便地将设计经验融合到优化计算过程中,能降低管网优化设计的复杂性和求解难度,快速有效求解符合工程实际的管网优化方案。
A two-step optimal design method based on the integer coding genetic algorithm for tree-type pipe network is proposed. At the first step, the optimal layout model based on integer coding genetic algorithm cording according to the one-point-water-supply principle is established by combining with designer' s experiences, which can avoid the difficulty in searching the solution of the optimal model based on binary coding genetic algorithm and easily to find a set of pipe network layout for the practical project. On the basis of first step, the second step is to establish the optimal pipe diameter model adopting the integer coding genetic algorithm with the minimal investment regarded as the objective. The proposed method is compared with the single parent genetic algorithm according to the application to a project design. The result shows that by using this method the optimal pipe network design plan can easily be worked out.