环状管网优化设计的双重编码混合遗传算法

以投资最小为目标函数,建立了注水系统环状管网优化数学模型。针对环状管网优化的特点,遗传算法采用二进制编码和整数编码相结合的双重编码,同时对管网连接方式和管径进行优化,并结合了模拟退火算法,形成了混合遗传算法。操作过程中采用随机多父辈适应函数值加权交叉,提出了与优化问题相适应的多种变异方法,给出了约束条件的处理方法,减少了不可行解的产生,使该算法的优化性能得到了提高。大庆油田某采油厂一新建注水区块的优化算例表明,管网投资由1160万下降到850万元,说明该优化方法有效且实用。

Dual coding hybrid genetic algorithm for optimization design of looped pipe network

An optimization model for pipe network of water injection system is established, in which the least cost is taken as objective function. Considering the characteristics of looped pipe network optimization, the dual coding combining binary coding with integer coding is adopted in the genetic algorithm to optimize the connection mode and pipe diameter simultaneously, and simulated annealing algorithm is combined with, then forming the hybrid genetic algorithm. In the course of operation, random parent-number fitness-weighted cross is adopted, and many mutation methods compliant with the problem are proposed, the process method of constraint conditions is presented, the number of infeasible solutions is reduced, and the optimum performance of the algorithm is improved. The example shows that the algorithm is efficient and practical.