利用机器学习中的模拟退火和K均值聚类算法优化了地震台站的巡检路径规划。以重庆市为例,通过模拟退火算法和百度地图API计算实际路径距离和时间,找到近似最优路径。在分组巡检时,引入K均值聚类算法先对台站进行分组再进行路径规划,提高了巡检效率。结果表明,在台站数量较多的情况下,机器学习算法相较于人工规划更能有效地找到最优路径,提高工作效率。
This study optimizes the inspection path planning of seismic stations using the simulated annealing and K-means clustering algorithms in machine learning. Taking Chongqing as an example, the simulated annealing algorithm and Baidu Maps API are employed to calculate actual path distances and times, successfully identifying an approximate optimal path. The introduction of the K-means clustering algorithm in group inspections categorizes the stations, which enhanced inspection efficiency through optimized path planning. The results indicate that, especially with a large number of stations, machine learning algorithms are more effective than manual planning in finding the optimal path, thereby improve overall efficiency.
2024,45(4): 91-98 收稿日期:2024-3-14
DOI:10.3969/j.issn.1003-3246.2024.04.012
基金项目:中国地震局地震监测、预报、科研三结合课题(项目编号:3JH-202401084);重庆市地震局科技创新课题(项目编号:CQ2024004)
作者简介:易江(1986—),男,工程师,主要从事地震监测与运维工作。E-mail:620049400@qq.com
*通讯作者:陈凯(1989—),男,高级工程师,主要从事地震监测与预警工作。E-mail:124753884@qq.com
参考文献:
陈凯,易江,孙国栋,等. 基于Android平台的地震台站运维系统的设计与实现[J]. 华北地震科学,2018,36(4):79-83.
戴波,张扬,缪发军,等. 江苏省地震台站巡检系统架构[J]. 地震地磁观测与研究,2017,38(6):110-116.
李元昊,段鹏飞,郭绍义,等. 船舶全局路径规划相关算法研究综述[J]. 船舶标准化工程师,2022,55(5):26-30.
吕帅,刘鹏飞,安小伟,等. 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[J]. 华南地震,2023,43(3):63-69.
向玉云,高爽,陈云红,等. 百度、高德及Google地图API比较研究[J]. 软件导刊,2017,16(9):19-21.
杨俊成,李淑霞,蔡增玉. 路径规划算法的研究与发展[J]. 控制工程,2017,24(7):1 473-1 480.
杨俊闯,赵超. K-Means聚类算法研究综述[J]. 计算机工程与应用,2019,55(23):7-14.
叶威惠,张飞舟. 真实路况下的快递配送路径优化研究[J]. 计算机工程与科学,2017,39(8):1 530-1 537.
于莹莹,陈燕,李桃迎. 改进的遗传算法求解旅行商问题[J]. 控制与决策,2014,29(8):1 483-1 488.
张广林,胡小梅,柴剑飞,等. 路径规划算法及其应用综述[J]. 现代机械,2011,(5):85-90.
张昊,陶然,李志勇,等. 基于自适应模拟退火遗传算法的特征选择方法[J]. 兵工学报,2009,30(1):81-85.
张润,王永滨. 机器学习及其算法和发展研究[J]. 中国传媒大学学报(自然科学版),2016,23(2):10-18.
Applegate D L, Bixby R E, Chvátal V, et al. The traveling salesman problem: a computational study[M]. Princeton University Press, 2007.
Bektas T. The multiple traveling salesman problem: an overview of formulations and solution procedures[J]. Omega, 2006, 34(3): 209-219.
Erdiwansyah E, Gani T A, Away Y. Hibridisasi Simulated Annealing dengan Algorithm Evolutionary dalam Penyelesaian Travelling Salesman problem (TSP)[J]. Kitektro, 2016, 1(1): 1-5.
Gao H C, Feng B Q, Zhu L. Reviews of the meta-heuristic algorithms for TSP[J]. Control and Decision, 2006, 3(21): 241-247.
Guntsch M, Middendorf M, Schmeck, H. An ant colony optimization approach to dynamic TSP[C]//Proceedings of the 3rd Annual Conference on Genetic and Evolutionary Computation. San Francisco, CA, United States: Morgan Kaufmann Publishers Inc, 2001.
Li J. An improved dynamic programming algorithm for Bitonic TSP[J]. Applied Mechanics and Materials, 2013(347/348/349/350): 309-3 098.
Nazeer K A A, Sebastian M P. Improving the accuracy and efficiency of the k-means clustering algorithm[C]//Proceedings of the World Congress on Engineering 2009 Vol I. London, 2009.
Sapkal S D, Kakarwal S N, Revankar P S. Analysis of classification by supervised and unsupervised learning[C]//Proceedings of International Conference on Computational Intelligence and Multimedia Applications (ICCIMA 2007), Sivakasi, India: IEEE, 2007.
Ye C, Yang Z C, Yan T X. An efficient and scalable algorithm for the traveling salesman problem[C]//Proceedings of 2014 IEEE 5th International Conference on Software Engineering and Service Science. Beijing, China: IEEE, 2014.
Yu H, Zhang K. Optimizing the Greedy algorithm used in the TSP abstract problems[J]. Applied Mechanics and Materials, 2014, 651/652/653: 2 352-2 355.