基于最大水回用规则的遗传算法用水网络优化

Optimal design of water networks by using genetic algorithm based on criterion of maximal water reuse

超结构用水网络规模大、结构复杂、求解比较困难。本文提出了基于最大水回用规则的遗传算法用水网络优化设计方法,该方法按出口浓度的单调性进行排列,使进口浓度或出口浓度达最大值以确定用水量,简化了用水网络超结构及数学模型。以最小新鲜水量为目标,以过程之间回用水量和废水量为基因,对简化后的用水网络运用遗传算法进行了优化设计。遗传算法采用浮点数编码方法,运用了算术交叉技术和种群间交叉技术。计算实例表明,本文所提方法可行,能利用最大水回用规则简化用水网络,并快速求出其最优解。

In order to simplify the superstructures of water networks to obtain their solutions, an approach to water network design by using genetic algorithm （GA） based on the criterion of maximal water reuse was proposed. In this method, the superstructure of water network was simplified according to the monotony of outlet concentration of each process. When the inlet concentration or the outlet concentration reached its maximum constraint, the fresh water usage was determined. The minimum fresh water usage of water network was taken as the optimal objective, the internal flow rate and waste water flow rate of each process were selected as the genes. This simplified water network was optimized by using GA. The coding in GA was the floating point number coding, the arithmetic cross technique and the cross technique among different groups were used. Case studies showed that this method worked well, the water networks could be simplified by using the criterion of maximal water reuse and the solution of global optimum of water network could be obtained by applying GA.