Intelligent operation and maintenance scheduling configuration method and device based on deep reinforcement learning

By adopting an intelligent operation and maintenance scheduling configuration method based on deep reinforcement learning, the problems of flexibility and efficiency of traditional operation and maintenance scheduling methods are solved, the optimal scheduling plan is generated, and the efficiency and accuracy of operation and maintenance scheduling are improved.

CN122453045APending Publication Date: 2026-07-24GUANGZHOU DIGITAL ENERGY TECHNOLOGY RESEARCH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN Β· China
Patent Type
Applications(China)
Current Assignee / Owner
GUANGZHOU DIGITAL ENERGY TECHNOLOGY RESEARCH CO LTD
Filing Date
2026-04-30
Publication Date
2026-07-24

AI Technical Summary

Technical Problem

Traditional operation and maintenance scheduling methods are unable to flexibly cope with multi-dimensional and dynamically changing scheduling needs. The rigid rules, high human-computer interaction threshold, and strong data dependence lead to low efficiency and insufficient accuracy.

Method used

An intelligent operation and maintenance scheduling configuration method based on deep reinforcement learning is adopted. By converting users' scheduling request text into a set of scheduling rules, a deep reinforcement learning model is used for analysis and decision-making to generate the optimal scheduling plan and to optimize it with rewards and penalties.

Benefits of technology

It improved the efficiency and accuracy of operation and maintenance scheduling, simplified complex processes, reduced labor costs, and enhanced the system's flexibility and adaptability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122453045A_ABST
    Figure CN122453045A_ABST
Patent Text Reader

Abstract

The application discloses an intelligent operation and maintenance scheduling configuration method and device based on deep reinforcement learning, and the method comprises the following steps: inputting the scheduling demand text of a user into a preset scheduling rule configuration model for conversion to obtain a configuration file of a scheduling rule set of the user, wherein the scheduling rule set comprises a plurality of sub-scheduling rules; inputting the scheduling rule set into a deep reinforcement learning model, analyzing the scheduling rule set, and obtaining a reward function corresponding to the scheduling rule set; making a decision on a scheduling task through an agent of the deep reinforcement learning model and according to the scheduling rule set to obtain a scheduling scheme after the decision; and rewarding and punishing the scheduling scheme according to the reward function, and outputting an optimal scheduling scheme for the scheduling demand of the user after optimization. It can be seen that the application can solve the technical problems of complex and low-efficiency operation and maintenance scheduling, thereby being favorable to improving the efficiency and accuracy of operation and maintenance scheduling.
Need to check novelty before this filing date? Find Prior Art