蚁群优化算法驱动的DQN多任务协同调度方法研究  

Research on Ant Colony Optimization Algorithm Driven DQN Multi-task Collaborative Scheduling Method

作　　者：李华峰 王富臣 江广冀 宋悦 Li Huafeng;Wang Fuchen;Jiang Guangji;Song Yue(State Grid Huainan Power Supply Co.,Ltd.,Huainan Anhui 232007,China;Hefei Thermal Power Group Co.,Ltd.,Hefei Anhui 230001,China)

机构地区：[1]国网淮南供电公司,安徽淮南232007 [2]合肥热电集团有限公司,安徽合肥230001

出　　处：《电气自动化》2025年第1期5-7,共3页Electrical Automation

基　　金：国网淮南供电公司科技项目(2612F01800C0)。

摘　　要：针对传统调度方法存在调度效率低下、资源利用率低等问题,提出了蚁群优化算法驱动的深度Q网络方法。通过结合蚁群优化算法的全局搜索能力和深度Q网络在学习调度决策方面的优势,能够在复杂环境下实现高效多任务调度。仿真试验结果表明,与普通的Q学习和深度Q网络相比,所提方法在收敛速度和效率方面更有优势。A deep Q-network method driven by ant colony optimization algorithm was proposed to address the problems of low scheduling efficiency and resource utilization in traditional scheduling methods.By combining the global search capability of ant colony optimization algorithm with the advantages of deep Q-networks in learning scheduling decisions,the efficient multi task scheduling can be achieved in complex environments.The simulation test results show that compared with ordinary Q-learning and deep Q-networks,the proposed method has more advantages in convergence speed and efficiency.

关 键 词：深度Q网络 蚁群算法 Q学习 任务调度 

分 类 号：TN929.5[电子电信—通信与信息系统]