基于改进遗传算法的给水管网最短路径求解

Based on Improved Genetic Algorithm for Solving Shortest Path to Supply Water Network

研究城市供水系统问题,为降低城市给水管网扩建成本,提高城市给水管网通过水质量的安全性,提高综合效益,研究了网络地理信息系统(WebGIS)中的遗传算法(GA),针对遗传算法在寻优过程中存在的问题,提出了一种隔离小生境自适应与遗传算法结合,设计自适应交叉函数和自适应变异函数,自适应调整交叉率和变异率策略,克服了简单遗传算法易陷入局部极小和早熟收敛的缺陷,并经过仿真表明,提高了算法的搜索效率、精度和稳定性,对于求解城市给水管网最短路径问题效果有明显的改进。

To reduce the city water supply network expansion costs,improve the urban water security of quality pipe network of supply water.a genetic algorithm is improved for solving the shortest path for the pipe network and the network Geographic Information System（WebGIS） of genetic algorithms（GA） is studied.A separate niche adaptive and genetic algorithm is presented.By using niche technology and designing adaptive crossover function and mutation function,the improved genetic algorithm can adjust the crossover rate and the frequency of mutation each individual,and avoid local minimum values and premature convergence which often occurs in simple genetic algorithms.It has higher efficiency,precision and stability.