Method of training reinforcement learning agent for industrial process system, and system for training reinforcement learning agent for industrial process system

JP2025174914APending Publication Date: 2025-11-28ABB (SCHWEIZ) AG
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
JP2025081037
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Priority Date
2024-05-15
Filing Date
2025-05-14
Publication Date
2025-11-28

Smart Images

  • Figure 2025174914000001_ABST
    Figure 2025174914000001_ABST
Patent Text Reader

Abstract

To provide a method of training a reinforcement learning (RL) agent for an industrial process system.SOLUTION: A method according to the present invention has a step 110 of training a RL agent with using plant history data of an industrial process system, and a step 120 of re-training the RL agent with using the plant history data and a low-fidelity simulator of the industrial process system. The step 120 has a step 130 of analyzing the plant history data so as to discriminate a process state which was not searched during the training of the RL agent and a white spot as a region of dynamic behavior, and a step 140 of re-training the RL agent by searching with priority with using information obtained from the white spot and simulated data obtained by simulating the industrial process system by using the low-fidelity simulator.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Intelligent agent reinforcement learning method and device based on iterative strategy constraint

    CN116681142A

  • Simulator system and simulator method

    JP2023069415A

  • Multi-fidelity simulated data for machine learning

    US20210357692A1