无人机在炸药库房区域智能巡检的应用研究

doi:10.16265/j.cnki.issn1003-3033.2025.S1.0012

摘要/Abstract

摘要： 为提升煤矿炸药库房的安全管理水平,解决传统人工巡检方式存在的总行程距离长以及总行程耗时长等问题,提出采用无人机智能巡检炸药库房区域。首先,采用三维栅格化建模方法分析炸药库房的复杂空间环境,构建炸药库房三维环境模型;然后,综合巡检效率与巡检成本2个关键指标,利用无人机智能巡检目标函数,分析炸药库房区域智能巡检的核心问题;最后,在炸药库房区域无人机巡检过程中,利用贪心遗传混合算法在炸药库房三维模型空间内对目标函数进行寻优求解,检验炸药库房区域无人机的巡检质量,并采用所提方法进行试验。结果表明:该设计方法的总航行距离较试验对比方法分别减少0.14、0.13 km;总航行时间较试验对比方法分别缩短6.0、4.0 min,即完成巡检的总航行距离最短、总航行时间最低,表明该设计方法的巡检效率高。此外,采用该设计方法的无人机能在高风险环境下进行非接触式巡检,显著提升了安全性和巡检效率。

关键词: 无人机, 炸药库房区域, 智能巡检, 三维环境模型, 目标函数, 贪心遗传混合算法

Abstract:

In order to further improve the safety management level of explosive warehouses in coal mines and solve various problems, such as long total navigation distance and long total navigation time in traditional manual inspection methods, the application of unmanned aerial vehicles in the intelligent inspection of explosive warehouse areas was studied. Firstly, the complex spatial environment of the explosive warehouse was analyzed using a three-dimensional rasterization modeling method, and a three-dimensional environment model of the explosive warehouse was constructed. Then, based on the two key indicators of comprehensive inspection efficiency and inspection cost, the core issues of intelligent inspection in the explosive warehouse area were analyzed using the objective function of intelligent inspection for unmanned aerial vehicles. Finally, during the unmanned aerial vehicle inspection process in the explosive warehouse area, a greedy genetic hybrid algorithm was used to optimize and solve the objective function in the three-dimensional model space of the explosive warehouse area, thereby testing the inspection quality of the explosive warehouse area by unmanned aerial vehicles. The experiment was conducted using the proposed method, and the results show that the total navigation distance of the design method is reduced by 0.14 km and 0.13 km, respectively compared to the experimental method; the total navigation time is shortened by 6.0 min and 4.0 min, respectively compared to the experimental method, indicating that the design method in the article has the shortest total navigation distance and the lowest total navigation time for completing inspections and that the inspection efficiency of the design method in the article is high. In addition, the unmanned aerial vehicle using the design method described in the article can conduct non-contact inspections in high-risk environments, significantly improving safety and inspection efficiency.

Key words: unmanned aerial vehicle, explosive warehouse area, intelligent inspection, three-dimensional environmental model, objective function, greedy genetic hybrid algorithm

中图分类号:

X934

李波. 无人机在炸药库房区域智能巡检的应用研究[J]. 中国安全科学学报, 2025, 35(S1): 71-77.

LI Bo. Research on application of intelligent inspection by unmanned aerial vehicles in explosive warehouse area[J]. China Safety Science Journal, 2025, 35(S1): 71-77.

图/表 4

图1

表1

表2

图2

参考文献 11

[1]	魏明, 孙雅茹, 孙博, 等. 基于深度强化学习的无人机线路及航迹协同规划[J]. 中国安全科学学报, 2023, 33(8):68-76. doi: 10.16265/j.cnki.issn1003-3033.2023.08.0038
	WEI Ming, SUN Yaru, SUN Bo, et al. UAV distribution route and flight path collaborative planning based on deep reinforcement learning[J]. China Safety Science Journal, 2023, 33(8):68-76. doi: 10.16265/j.cnki.issn1003-3033.2023.08.0038
[2]	赵恩来, 苏鑫磊, 董衍旭, 等. 基于遗传算法的输电线路无人机巡检任务智能分配协同控制系统设计[J]. 自动化与仪器仪表, 2024(9): 64-68.
	ZHAO Enlai, SU Xinlei, DONG Yanxu, et al. Design of cooperative control system for intelligent assignment of transmission line UAV inspection task based on genetic algorithm[J]. Automation & Instrumentation, 2024(9): 64-68.
[3]	朱定赟. 未知环境下基于机器视觉的巡检机器人自主导航控制方法[J]. 自动化与仪器仪表, 2024(9): 221-224,229.
	ZHU Dingyun. Autonomous navigation control method for inspection robot based on machine vision in unknown environment[J]. Automation & Instrumentation, 2024(9): 221-224,229.
[4]	崔子鹏, 汪雨洋, 张磊, 等. 基于优化跳点搜索和车道采样的巡检机器人路径规划方法[J]. 控制工程, 2024, 31(9): 1 634-1 643.
	CUI Zipeng, WANG Yuyang, ZHANG Lei, et al. Path planning method for inspection robot based on optimized jump point search and lane line sampling[J]. Control Engineering of China, 2024, 31(9): 1 634-1 643.
[5]	张文献. 无人机倾斜摄影测量在水库巡检中的应用[J]. 测绘与空间地理信息, 2024, 47(9): 172-174.
	ZHANG Wenxian. Application of UAV oblique photogrammetry in reservoir patrol inspection[J]. Geomatics & Spatial Information Technology, 2024, 47(9): 172-174.
[6]	侯伟, 陈雅, 宋承继, 等. 基于改进YOLOv5算法的无人机巡检图像智能识别方法[J]. 微型电脑应用, 2024, 40(9): 26-30,36.
	HOU Wei, CHEN Ya, SONG Chengji, et al. Intelligent recognition method for UAV inspection images based on improved YOLOv5 algorithm[J]. Microcomputer Applications, 2024, 40(9): 26-30,36.
[7]	王晓晖, 兰建辉, 韩玉鑫, 等. 基于巡检机器人的燃气发电主厂房异常问题图像识别报警方法[J]. 自动化技术与应用, 2024, 43(9): 64-68.
	WANG Xiaohui, LAN Jianhui, HAN Yuxin, et al. Image recognition and alarm method for abnormal problems of gas power plant based on inspection robot[J]. Techniques of Automation and Application, 2024, 43(9): 64-68.
[8]	王骋昊, 贺凌飞. 输电线路电场分布特征驱动下的无人机巡检避障策略分析[J]. 技术与市场, 2024, 31(9): 74-78.
	WANG Chenghao, HE Lingfei. Analysis of obstacle avoidance strategies for unmanned aerial vehicle inspection driven by the distribution characteristics of electric fields in transmission lines[J]. Technology & Market, 2024, 31(9): 74-78.
[9]	高阳, 杨磊, 刘枫琪, 等. 基于各向异性滤波的无人机红外图像超分辨率重建方法[J]. 红外技术, 2024, 46(8): 933-939.
	GAO Yang, YANG Lei, LIU Fengqi, et al. Super-resolution reconstruction of thermal images by UAV based on anisotropic filtering[J]. Infrared Technology, 2024, 46(8): 933-939.
[10]	胡小兵, 卢泽, 李航, 等. 基于实时备降安全约束的无人机风电场巡检路径规划[J]. 中国安全科学学报, 2025, 35(2):28-39. doi: 10.16265/j.cnki.issn1003-3033.2025.02.0865
	HU Xiaobing, LU Ze, LI Hang, et al. Unmanned aerial vehicle wind farm inspection path planning based on real-time potential landing safety constraints[J]. China Safety Science Journal, 2025, 35(2):28-39. doi: 10.16265/j.cnki.issn1003-3033.2025.02.0865
[11]	冯欧阳, 马奇, 移强平, 等. 基于改进蚁群算法的无人机在配电网线路巡检中的研究与应用[J]. 电气时代, 2024(7): 53-57.

求解步骤	操作
问题定义与目标函数设定;编码与初始化种群	明确无人机巡检的目标,并设定目标函数; 采用二进制编码表示无人机的巡检路径,并初始化种群,生成多个可能的巡检路径作为初始解
适应度函数计算	根据目标函数计算每个个体的适应度值
贪心算法局部优化	使用贪心算法局部优化当前最优个体,以提升其适应度值。贪心算法在每1步都选择当前最优的决策,以期望达到全局最优
遗传操作	选择操作采用轮盘赌选择等策略,交叉操作采用基于模板的交叉等策略,变异操作则随机改变个体的某些基因
迭代与收敛	重复步骤3—步骤5,直到达到预设的迭代次数或适应度值不再显著提升为止。此时,认为算法已经收敛到最优解或近似最优解

货位	位置信息/m	货位	位置信息/m
1	(35,42)	22	(44,17)
2	(47,54)	23	(56,08)
3	(39,22)	24	(75,19)
4	(75,26)	25	(62,24)
5	(23.17)	26	(28,40)
6	(64.29)	27	(19,56)
7	(49,02)	28	(22,41)
︙	︙	︙	︙
20	(17,11)	40	(19,14)
21	(26,39)	—	—

[1]	邹彪, 朱晓康, 任伟达, 高煜博, 方腾. 强电磁扰动环境变电站无人机自主高精巡检研究[J]. 中国安全科学学报, 2025, 35(S1): 239-245.
[2]	李聪, 侯乐乐, 金衍科, 武卓琦, 周睿, 姚一娜. 基于Bentley Hammer的系留无人机载竖直管缆水锤效应研究[J]. 中国安全科学学报, 2025, 35(7): 122-132.
[3]	李航, 韩凤, 胡小兵. 城市无人机坠落对地面交通的风险评估方法[J]. 中国安全科学学报, 2025, 35(6): 157-164.
[4]	吴立舟, 李华, 张鹏, 钱虹燕, 康飞, 谭光丽. 实景三维数据下城市洪涝模拟及风险评估[J]. 中国安全科学学报, 2025, 35(5): 237-245.
[5]	张健, 赵嶷飞, 卢飞, 黎宗孝, 罗鑫悦. 基于Event模型的城市物流无人机同高度交叉运行间隔研究[J]. 中国安全科学学报, 2025, 35(5): 99-105.
[6]	耿鹏, 杨豪杰, 薛芳琳, 柳艳. 基于多无人机协同的林火安全探测及人员疏散[J]. 中国安全科学学报, 2025, 35(4): 43-50.
[7]	李聪, 何泽群, 杨高豪, 许文博, 王悦朋. 灭火弹爆炸过程及灭火剂抛洒特性研究[J]. 中国安全科学学报, 2025, 35(3): 107-114.
[8]	胡小兵, 卢泽, 李航, 周航. 基于实时备降安全约束的无人机风电场巡检路径规划[J]. 中国安全科学学报, 2025, 35(2): 28-39.
[9]	柳青, 沈恬. 基于贝叶斯网络的生物样本无人机运输风险评估[J]. 中国安全科学学报, 2025, 35(1): 16-24.
[10]	耿增显, 陈俊宇. 基于模糊贝叶斯网络的低空无人机运行风险评估[J]. 中国安全科学学报, 2024, 34(8): 53-60.
[11]	胡焱松, 王长君, 郑金子, 褚宇航. 面向交通事故现场三维实景建模的无人机航拍参数[J]. 中国安全科学学报, 2024, 34(7): 194-201.
[12]	江松, 李研博, 何旭乾, 何润丰, 张超, 张存良. 基于无人机影像深度学习的滑坡灾害智能识别[J]. 中国安全科学学报, 2024, 34(7): 229-238.
[13]	龚英, 刘晓聪, 周愉峰, 杨维. 面向灾情侦察的有优先级多无人机协同调度优化[J]. 中国安全科学学报, 2024, 34(11): 202-212.
[14]	靳慧斌, 刘希, 黄俊, 陈怿淳. 基于定位、速度双误差的物流无人机碰撞概率分析[J]. 中国安全科学学报, 2024, 34(10): 64-70.
[15]	李聪, 许文博, 金衍科, 汪陈徽, 程海涛. 基于中继-巡视无人机协同作业的输电线路山火隐患特征分析[J]. 中国安全科学学报, 2023, 33(S2): 238-244.