面向低并发设备的矿山边缘计算任务卸载仿真与优化

doi:10.16182/j.issn1004731x.joss.25-0840

摘要/Abstract

摘要：

随着矿山智能化的深入，任务卸载技术已成为矿山边缘计算(mine-edge-computing, MEC)的核心技术。针对部署矿山低并发设备(mine low-concurrency devices, MLCD)的边缘计算系统，为解决任务处理时延与矿山边缘服务器(mine-edge-server, MES)负载均衡的协同优化难题，构建了一种面向MLCD的MEC任务卸载离散事件仿真模型。通过动态仿真任务的排队、传输与计算的过程，以实现任务处理时延最小化。提出一种改进的演化算法ISBT-EA (task-feature-driven initialization strategy and bottleneck-task-driven evolutionary algorithm)，融合了任务特征驱动的初始化策略与局部搜索算子，在仿真推演中提升了策略寻优效率。仿真结果表明：与传统方法相比，ISBT-EA的平均时延至少降低了4.77%，为解决MLCD场景下的时延优化与负载均衡问题提供了可靠的仿真与优化方案。

关键词: 矿山边缘计算, 矿山低并发设备, 任务卸载, 演化算法, 时延优化

Abstract:

With the advancement of mine intelligence, task offloading technology has become a core technology of mine edge computing (MEC). For edge computing systems deployed with mine low-concurrency devices (MLCD), a discrete-event simulation model for MEC task offloading oriented to MLCD was constructed to solve the collaborative optimization problem of task processing latency and load balancing of mine edge servers (MES). By dynamically simulating the queuing, transmission, and computation processes of tasks, the objective of minimizing task processing latency was achieved. An improved evolutionary algorithm, ISBT-EA, was proposed, which integrated a task-characteristic-driven initialization strategy and local search operators, enhancing the efficiency of strategy optimization in simulation-based evaluation. The simulation results show that, compared with traditional methods, ISBT-EA reduces the average latency by at least 4.77%, providing a reliable simulation and optimization solution for addressing latency optimization and load balancing problems in MLCD scenarios.

Key words: mine edge computing, mine low-concurrency device, task offloading, evolutionary algorithm, latency optimization

中图分类号:

TP391.9

郭肇禄,刘千辉 . 面向低并发设备的矿山边缘计算任务卸载仿真与优化[J]. 系统仿真学报, 2026, 38(7): 2091-2104.

Guo Zhaolu,Liu Qianhui . Simulation and Optimization of Task Offloading in Mine Edge Computing for Low-concurrency Devices[J]. Journal of System Simulation, 2026, 38(7): 2091-2104.

图/表 6

图1

图2

表1

仿真参数设置

参数	取值
任务数量 $I$	50~300
任务 $t i$ 的数据量 $s i$ /MB	[1, 25]
任务 $t i$ 的计算量 $c i$ /cycles	[100, 500]
任务 $t i$ 所属设备编号 $u i$	1~T
MLCD数量 $R$	5~15
MLCD的并发数 $q$	3~6
$M M L C D, k$ 计算能力 $F k$ /(cycle/s)	[300, 600]
最大运行停滞次数 $c c o u n t, m a x$	3
$β$	$2 × 107$
$w r a t i o$	0.15
MES负载均衡最大容忍度 $σ m a x$	0.3
$M E S, l$ 的计算能力 $f l$ /(cycle/s)	[1 000, 1 500]
MES数量 $L$	5~10
传输功率 $P u p l o a d$	2~5
通信带宽 $B$	10
背景噪声功率 $ω ¯$	0.1
信道增益 $g c$	1
$α$	1
$χ$	$10 - 3$
$N P$	30

表1

表2

表3

图3

参考文献 24

[1]	丁恩杰, 廖玉波, 张雷, 等. 煤矿信息化建设回顾与展望[J]. 工矿自动化, 2020, 46(7): 5-11.
	Ding Enjie, Liao Yubo, Zhang Lei, et al. Review and Prospect of Coal Mine Informatization Construction[J]. Journal of Mine Automation, 2020, 46(7): 5-11.
[2]	钟一鸣, 仲文斌, 王阳培华, 等. 边缘计算技术在智慧矿山建设中的应用[J]. 现代矿业, 2025, 41(6): 29-35.
	Zhong Yiming, Zhong Wenbin, Wang Yangpeihua, et al. Application of Edge Computing Technology in Intelligent Mine Construction[J]. Modern Mining, 2025, 41(6): 29-35.
[3]	李智勇, 王琦, 陈一凡, 等. 车辆边缘计算环境下任务卸载研究综述[J]. 计算机学报, 2021, 44(5): 963-982.
	Li Zhiyong, Wang Qi, Chen Yifan, et al. A Survey on Task Offloading Research in Vehicular Edge Computing[J]. Chinese Journal of Computers, 2021, 44(5): 963-982.
[4]	张伟, 帅根来, 白洪亮, 等. 矿山环境下的矿车自主导航与避障算法探析[J]. 数字技术与应用, 2024, 42(5): 208-210.
[5]	贾永刚, 张志成, 卓晓军, 等. 基于专利文献检索的深海采矿环境影响监测与保护技术发展趋势分析[J]. 矿冶工程, 2025, 45(3): 1-7.
	Jia Yonggang, Zhang Zhicheng, Zhuo Xiaojun, et al. Analysis of Technology Trends for Environmental Monitoring and Protection in Deep-sea Mining Based on Patent Literature Retrieval[J]. Mining and Metallurgical Engineering, 2025, 45(3): 1-7.
[6]	刘向举, 李金贺, 方贤进, 等. 移动边缘计算中计算卸载与资源分配联合优化策略[J]. 计算机工程与科学, 2024, 46(3): 416-426.
	Liu Xiangju, Li Jinhe, Fang Xianjin, et al. A Joint Optimization Strategy for Compute Offloading and Resource Allocation in Mobile Edge Computing[J]. Computer Engineering & Science, 2024, 46(3): 416-426.
[7]	丁昱华, 陈力, 卫国. 面向6G的低时延高可靠边缘计算架构[J]. 中国科学院大学学报:中英文, 2025, 42(1): 126-133.
	Ding Yuhua, Chen Li, Wei Guo. Ultra-reliable Low-latency Edge Computing Architecture for 6G[J]. Journal of University of Chinese Academy of Sciences, 2025, 42(1): 126-133.
[8]	田真真, 蒋维, 郑炳旭, 等. 基于服务器集群的负载均衡优化调度算法[J]. 计算机科学, 2022, 49(增1): 639-644.
	Tian Zhenzhen, Jiang Wei, Zheng Bingxu, et al. Load Balancing Optimization Scheduling Algorithm Based on Server Cluster[J]. Computer Science, 2022, 49(S1): 639-644.
[9]	Yang Guisong, Hou Ling, He Xingyu, et al. Offloading Time Optimization via Markov Decision Process in Mobile-edge Computing[J]. IEEE Internet of Things Journal, 2021, 8(4): 2483-2493.
[10]	Wu Yuan, Qian Liping, Ni Kejie, et al. Delay-minimization Nonorthogonal Multiple Access Enabled Multi-user Mobile Edge Computation Offloading[J]. IEEE Journal of Selected Topics in Signal Processing, 2019, 13(3): 392-407.
[11]	Fan Qiang, Ansari N. Towards Workload Balancing in Fog Computing Empowered IoT[J]. IEEE Transactions on Network Science and Engineering, 2020, 7(1): 253-262.
[12]	Li Guangshun, Yao Yonghui, Wu Junhua, et al. A New Load Balancing Strategy by Task Allocation in Edge Computing Based on Intermediary Nodes[J]. EURASIP Journal on Wireless Communications and Networking, 2020, 2020(1): 3.
[13]	Sun Yuxuan, Guo Xueying, Song Jinhui, et al. Adaptive Learning-based Task Offloading for Vehicular Edge Computing Systems[J]. IEEE Transactions on Vehicular Technology, 2019, 68(4): 3061-3074.
[14]	沈俊杰, 彭江, 郭坤银, 等. 车联网中基于位置信息映射和相关性评估的进化多任务优化算法[J]. 电子学报, 2025, 53(5): 1661-1676.
	Shen Junjie, Peng Jiang, Guo Kunyin, et al. Location Mapping and Correlation Assessment Based Evolutionary Multi-task Optimization Algorithm in the Internet of Vehicles[J]. Acta Electronica Sinica, 2025, 53(5): 1661-1676.
[15]	丁转莲, 胡曦, 曹略, 等. 采用进化计算的高维多模态特征选择[J]. 西安交通大学学报, 2024, 58(9): 117-128.
	Ding Zhuanlian, Hu Xi, Cao Lüe, et al. High-dimensional Multimodal Feature Selection Based on Evolutionary Computation[J]. Journal of Xi'an Jiaotong University, 2024, 58(9): 117-128.
[16]	Holland J H. Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence[M]. Cambridge: The MIT Press, 1992.
[17]	Deb K, Pratap A, Agarwal S, et al. A Fast and Elitist Multiobjective Genetic Algorithm: NSGA-II[J]. IEEE Transactions on Evolutionary Computation, 2002, 6(2): 182-197.
[18]	贺毅朝, 王熙照, 刘坤起, 等. 差分演化的收敛性分析与算法改进[J]. 软件学报, 2010, 21(5): 875-885.
	He Yichao, Wang Xizhao, Liu Kunqi, et al. Convergent Analysis and Algorithmic Improvement of Differential Evolution[J]. Journal of Software, 2010, 21(5): 875-885.
[19]	刘会宇, 韩继红, 袁霖, 等. 基于双变异策略的自适应骨架差分进化算法[J]. 通信学报, 2017, 38(8): 201-212.
	Liu Huiyu, Han Jihong, Yuan Lin, et al. Self-adaptive Bare-bones Differential Evolution Based on Bi-mutation Strategy[J]. Journal on Communications, 2017, 38(8): 201-212.
[20]	Derrac Joaquín, García Salvador, Molina Daniel, et al. A Practical Tutorial on the Use of Nonparametric Statistical Tests as a Methodology for Comparing Evolutionary and Swarm Intelligence Algorithms[J]. Swarm and Evolutionary Computation, 2011, 1(1): 3-18.
[21]	Karaboga Dervis. An Idea Based on Honey Bee Swarm for Numerical Optimization[EB/OL]. (2005-10-09) [2025-08-12]. .
[22]	Dorigo M, Gambardella L M. Ant Colony System: A Cooperative Learning Approach to the Traveling Salesman Problem[J]. IEEE Transactions on Evolutionary Computation, 1997, 1(1): 53-66.
[23]	Storn Rainer, Price K. Differential Evolution-A Simple and Efficient Heuristic for Global Optimization over Continuous Spaces[J]. Journal of Global Optimization, 1997, 11(4): 341-359.
[24]	Kennedy J, Eberhart R. Particle Swarm Optimization[C]//Proceedings of ICNN'95 - International Conference on Neural Networks. Piscataway: IEEE, 1995: 1942-1948.

实例	ISBT-EA-R	ISBT-EA-A	ISBT-EA-B	ISBT-EA
T1 (50)	3.007 0	2.843 0	2.946 0	2.829 0
T1 (50)	0.065 0 (-)	0.025 2 (-)	0.053 3 (-)	0.029 3
T2 (100)	5.825 3	5.368 2	5.719 4	5.323 1
T2 (100)	0.111 1 (-)	0.055 4 (-)	0.084 5 (-)	0.039 6
T3 (150)	9.177 7	8.037 8	9.046 2	7.916 2
T3 (150)	0.167 4 (-)	0.115 2 (-)	0.158 7 (-)	0.081 0
T4 (200)	12.384 8	10.721 5	12.228 9	10.597 7
T4 (200)	0.183 8 (-)	0.129 0 (-)	0.205 1 (-)	0.124 5
T5 (250)	16.965 4	15.440 9	16.718 4	15.082 7
T5 (250)	0.310 1 (-)	0.281 8 (-)	0.351 0 (-)	0.318 5
T6 (300)	19.243 1	16.859 5	19.099 1	16.579 8
T6 (300)	0.339 7 (-)	0.223 2 (-)	0.287 1 (-)	0.249 3
+/-/=	0/6/0	0/6/0	0/6/0

实例	ISBT-EA	ABC	ACO	DE	PSO
T1 (50)	2.829 0	2.835 8	2.979 0	3.043 5	2.907 6
T1 (50)	0.029 3	0.032 1 (-)	0.032 7 (-)	0.055 4 (-)	0.088 4 (-)
T2 (100)	5.323 1	5.497 9	5.670 4	5.901 4	5.343 4
T2 (100)	0.039 6	0.063 9 (-)	0.073 7 (-)	0.100 4 (-)	0.132 4 (-)
T3 (150)	7.916 2	8.841 2	8.849 5	9.238 7	8.445 6
T3 (150)	0.081 0	0.102 6 (-)	0.119 1 (-)	0.175 6 (-)	0.241 1 (-)
T4 (200)	10.597 7	11.851 8	12.033 3	12.425 3	11.439 6
T4 (200)	0.124 5	0.152 1 (-)	0.141 0 (-)	0.222 6 (-)	0.390 5 (-)
T5 (250)	15.082 7	16.276 7	16.549 4	17.027 1	15.444 5
T5 (250)	0.318 5	0.273 4 (-)	0.195 9 (-)	0.350 6 (-)	0.528 4 (-)
T6 (300)	16.579 8	18.736 7	18.911 5	19.304 9	18.332 2
T6 (300)	0.249 3	0.274 4 (-)	0.242 0 (-)	0.306 6 (-)	0.653 9 (-)
+/-/=		0/6/0	0/6/0	0/6/0	0/6/0