基于遗传算法和高德地图API实现地震预警台站巡检路径自动规划

1.云南省地震局,昆明 650224;2.中国地质大学(北京)信息工程学院,北京 100083

台站巡检;路径规划;遗传算法;高德API;旅行商问题

Realization of Automatic Planning of Inspection Paths of Earthquake Early Warning Stations Based on Genetic Algorithm and Amap API
LYU Shuai1,2,LIU Pengfei1,AN Xiaowei1,PENG Yuxiang1,SU Yi1

1.Yunnan Earthquake Agency, Kunming 650224, China;2.School of Information Engineering , China University of Geosciences , Beijing 100083, China

Seismic station inspection; Route planning; Genetic algorithms; Amap API; Traveling salesman problem

DOI: 10.13512/j.hndz.2023.03.08

备注

由于地震预警台站分布较广,而地震监测中心站的运维人员较少,如何设计合理的巡检路径,以更少的巡检成本完成更多的巡检任务成为中心站面临的一个问题。文中采用高德地图API,基于遗传算法使用python编写程序实现地震预警台站巡检路径近似最优解的自动求解。通过输入云南57个地震预警基准站位置进行测试后得出,系统在样本个数为60,迭代次数为2500h,得到近似最优路径,总行驶距离4869km,过路费439元。相较于其他几次收敛结果,最优巡检路径收益在百公里以上,因此认为,利用遗传算法规划的台站巡检路径可以获得较低的巡检成本。
Due to the wide distribution of earthquake early warning stations and the small number of operation and maintenance personnel in seismic monitoring central stations, how to design a reasonable inspection path to accomplish more inspection tasks with less inspection cost becomes a problem for central stations. Based on the Amap API and genetic algorithm,this paper uses python to write a program to realize the automatic solution of the approximate optimal solution of the inspection path of the earthquake early warning station. After testing by inputting 57 EEW benchmark station locations in Yunnan,it is concluded that the system obtains the approximate optimal path with a total driving distance of 4869 km and a toll fee of 439 yuan RMB when the number of samples is 60 and the number of iterations is 2500. Compared with other convergence results, the optimal inspection path gains more than a hundred kilometers, so the inspection cost of the station inspection path planned by using genetic algorithm is lower.
·