Dynamic Optimization Method for Automated Testing Strategy of Power Internet of Things Based on Double Deep Q-Network

Gu Zhiming; Li Bo; Dai Zelin; Zhu Mengmeng; Zhang Li; Bao Mingkang

doi:10.11959/j.issn.1000-0801.DXKX260018

您当前的位置：

首页 >

文章列表页 >

Dynamic Optimization Method for Automated Testing Strategy of Power Internet of Things Based on Double Deep Q-Network

更新时间：2026-05-19

- Dynamic Optimization Method for Automated Testing Strategy of Power Internet of Things Based on Double Deep Q-Network
- Telecommunications Science (2026)
- 作者机构：
  
  1.云南电网有限责任公司电力科学研究院，云南昆明 650217
  云南省绿色能源与数字电力量测及控保重点实验室，云南昆明 650217
  3.云南电网有限责任公司，云南昆明650011
- 作者简介：
- 基金信息：
- DOI：10.11959/j.issn.1000-0801.DXKX260018
  CLC：
- Received：08 January 2026，
  
  Revised：2026-02-21，
  
  Accepted：18 May 2026，
- 稿件说明：
移动端阅览
Gu Zhiming, Li Bo, Dai Zelin, et al. Dynamic Optimization Method for Automated Testing Strategy of Power Internet of Things Based on Double Deep Q-Network[J/OL]. Telecommunications Science, 2026.
DOI：

Gu Zhiming, Li Bo, Dai Zelin, et al. Dynamic Optimization Method for Automated Testing Strategy of Power Internet of Things Based on Double Deep Q-Network[J/OL]. Telecommunications Science, 2026. DOI： 10.11959/j.issn.1000-0801.DXKX260018.

摘要

本文提出一种基于深度强化学习的电力物联网自动化测试策略优化方法，核心采用双深度 Q 网络（Double Deep Q-Network，DDQN）架构，通过目标网络与经验回放机制提升策略学习的稳定性与收敛效率。该方法依托双深度 Q 网络的决策能力，使智能体在与电力物联网环境的实时交互中，基于奖励反馈动态调整测试动作（如测试用例选择、参数配置优化），实现对电力物联网设备和系统的高效覆盖测试。实验结果表明，相较于传统随机策略与顺控策略，基于双深度 Q 网络的优化方法在智能电表、继电保护装置等典型设备测试中，缺陷检测率平均超 90%，测试覆盖率稳定在 93% 以上，且平均测试时间缩短 3-7 秒；在高负载、故障注入等动态场景下仍保持优异性能，有效应对电力物联网环境的不确定性，为电力物联网的稳定运行提供有力技术保障。

Abstract

This paper proposes an optimization method for automated testing strategies in the Power Internet of Things (Power IoT) based on deep reinforcement learning

with the core adoption of a Double Deep Q-Network (DDQN) architecture. Through the target network and experience replay mechanism

the stability and convergence efficiency of strategy learning are enhanced. Relying on the decision-making capability of the DDQN

this method enables the intelligent agent to dynamically adjust testing actions (such as test case selection and parameter configuration optimization) based on reward feedback during real-time interaction with the Power IoT environment

thereby achieving efficient coverage testing of Power IoT devices and systems. Experimental results show that

compared with traditional random strategies and sequential control strategies

the optimization method based on DDQN achieves an average defect detection rate of over 90% and a stable test coverage rate of more than 93% in the testing of typical devices such as smart electricity meters and relay protection devices. Additionally

the average testing time is shortened by 3-7 seconds. The method still maintains excellent performance in dynamic scenarios such as high load and fault injection

effectively addressing the uncertainties of the Power IoT environment and providing strong technical support for the stable operation of the Power IoT.

关键词

Keywords

references

张宁，杨经纬，王毅，等 . 面向泛在电力物联网的5G通信：技术原理与典型应用 [J ] . 中国电机工程学报， 2019 , 39 ( 14 ): 4015 - 4024 .

ZHANG Ning , YANG Jingwei , WANG Yi , et al . 5G Communication for the ubiquitous internet of things in electricity: technical principles and typical applications [J ] . Proceedings of the CSEE , 2019 , 39 ( 14 ): 4015 - 4024 (in Chinese) .

余翔，张海波，杨路 . 混合 D2D 蜂窝网络中基于模拟退火算法的资源调度策略 [J ] . 电信科学， 2017 , 33 ( 04 ): 71 - 77 .

YU Xiang , ZHANG Haibo , YANG Lu . Resource scheduling strategy based on simulated annealing algorithm in hybrid D2D cellular networks [J ] . Telecommunications Science , 2017 , 33 ( 04 ): 71 - 77 .

严兴煜，高赐威，陈涛，等 . 数字孪生虚拟电厂系统框架设计及其实践展望 [J ] . 中国电机工程学报， 2023 , 43 ( 2 ): 604 - 618 .

YAN Xingyu , GAO Ciwei , CHEN Tao , et al . Framework design and application prospect for digital twin virtual power plant system [J ] . Proceedings of the CSEE , 2023 , 43 ( 2 ): 604 - 618 .

伏玉笋，杨根科 . 无线超可靠低时延通信：关键设计分析与挑战 [J ] . 通信学报， 2020 , 41 ( 8 ): 187 - 203 .

FU Yusun , YANG Genke . Wireless ultra - reliable and low - latency communication: key design analysis and challenges [J ] . Journal on Communications , 2020 , 41 ( 8 ): 187 - 203 .

邵振国，张承圣，陈飞雄，等 . 生成对抗网络及其在电力系统中的应用综述 [J ] . 中国电机工程学报， 2023 , 43 ( 3 ): 987 - 1003 .

SHAO Zhenguo , ZHANG Chengsheng , CHEN Feixiong , et al . A review on generative adversarial networks for power system applications [J ] . Proceedings of the CSEE , 2023 , 43 ( 3 ): 98 - 1003 .

肖金星，徐冰雁，叶影，等 . 基于生成对抗网络的配电网与多微网随机调度 [J ] . 科学技术与工程， 2023 , 23 ( 05 ): 1997 - 2006 .

XIAO Jingxing , XU Binyan , YE Ying , et al . Stochastic scheduling of distribution network and multi - microgrids based on generative adversarial networks [J ] . Science Technology and Engineering , 2023 , 23 ( 05 ): 1997 - 2006 .

邓炳光，徐成义，张泰，等 . 基于多智能体深度强化学习的D2D通信资源联合分配方法 [J ] . 电子与信息学报， 2023 , 45 ( 4 ): 1173 - 1182 .

DENG Bingguang , XU Chengyi , ZHANG Tai , et al . A joint resource allocation method of D2D communication resources based on multi - agent deep reinforcement learning [J ] . Journal of Electronics & Information Technology , 2023 , 45 ( 4 ): 1173 - 1182 .

康迎业，蔡泽祥，曾兴，等 . 考虑即裁成本的配电网能量管理业务协同调度方法 [J ] . 南方电网技术， 2021 , 15 ( 09 ): 61 - 68 .

KANG Yingye , CAI Zexiang , ZENG Xing , et al . An offloading cost - eeding collaborative scheduling method for power distribution network energy management applications . [J ] . Southern Power System Technology , 2021 , 15 ( 09 ): 61 - 68 .

姚楠，刘子全，秦剑华，等 . 基于电力物联网的边缘计算任务卸载优化 [J ] . 科学技术与工程， 2022 , 22 ( 16 ): 6577 - 6584 .

YAO Nan , LIU Ziquan , QIN Jianhua , et al . Offloading Optimization of the Edge Computing Task Based on the Power Internet of Things [J ] . Science Technology and Engineering , 2022 , 22 ( 16 ): 6577 - 6584 .

周振宇，王照，廖海君，等 . 电力物联网5G云-边-端协同框架与资源调度方法 [J ] . 电网技术， 2022 , 46 ( 5 ): 1641 - 1651 .

ZHOU Zhenyu , WANG Zhao , LIAO Haijun , et al . 5G cloud-edge-end collaboration framework and resource scheduling method in power internet of things [J ] . Power System Technology , 2022 , 46 ( 5 ): 164 - 1651 .

蒲天胜，杜帅，李烨，等 . 面向隐私保护基于联邦强化学习的分布式电源协同优化策略 [J ] . 电力系统自动化， 2023 , 47 ( 8 ): 62 - 70 .

PU Tiansheng , DU Shuai , LI Ye , et al . Collaborative optimization strategy of distributed generators based on federated reinforcement learning for privacy preservation [J ] . Automation of Electric Power Systems , 2023 , 47 ( 8 ): 62 - 70 .

仝杰，乔宇蒙，蒲天胜，等 . 电力物联网边缘智能：概念、架构、技术和应用 [J ] . 中国电机工程学报， 2024 , 44 ( 14 ): 5473 - 5496 .

TONG Jie , QIAO Yumeng , PU Tiansheng , et al . Edge intelligence to power Internet of Things: concept, architecture, technology and application [J ] . Proceedings of the CSEE , 2024 , 44 ( 14 ): 5473 - 5496 .

刘林，祁兵，李彬，等 . 面向电力物联网新业务的电力通信网需求及发展趋势 [J ] . 电网技术， 2020 , 44 ( 8 ): 3114 - 3128 .

LIU Lin , QI Bing , LI Bin , et al . Requirements and developing trends of electric power communication network for new services in electric internet of things [J ] . Power System Technology , 2020 , 44 ( 8 ): 3114 - 3128 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Multi-service routing algorithm for satellite networks based on deep Q-network with long short-term memory

Research on heterogeneous task offloading methods based on CA3C in satellite-ground edge computing networks

A layered collaborative resource allocation algorithm for cell-free RAN aimed at efficient resource utilization

Research on optimization of data transmission strategies based on deep reinforcement learning

Automatic modulation classification in dynamic noise based on BYOL and real complex domain fusion

Related Author

Pan Chengsheng

Wei Debin

Zhang Yi

Wen Jinglong

Pan Chengsheng

Wang Lin

Jin Xiaoyong

Liu Zhiguo

Related Institution

National Local Joint Engineering Research Center for Communication and Network Technology, Nanjing University of Posts and Telecommunications

College of Information Engineering, Dalian University

School of Electronics and Information Engineering, Nanjing University of Information Science and Technology

School of Environmental and Chemical Engineering, Dalian University

Key Laboratory of Communication and Network, Dalian University

⁰