一种固液传热系统的建模和温度控制方法
By establishing a variable state time-delay ODE-Heat coupling model and a reinforcement learning temperature control method, the problems of model uncertainty and actuator saturation in solid-liquid heat transfer systems are solved, achieving more accurate modeling and stable temperature control.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- BEIJING INST OF TECH
- Filing Date
- 2023-10-09
- Publication Date
- 2026-07-17
AI Technical Summary
In solid-liquid heat transfer systems, existing technologies struggle to establish accurate models, resulting in time delay effects and actuator saturation problems, which degrade the performance of the control system.
A variable state time-delay ODE-Heat coupling model is established, and a reinforcement learning temperature control method with actuator saturation is adopted. The controller is trained through Markov decision process and Actor-Critic network to design a control strategy that can adapt to uncertainty.
This improved the modeling accuracy of the heat transfer process, reduced the requirements for prior knowledge of the system, ensured that the controller input was within the available range of the actuator, avoided system instability, and achieved stable control of the system.
Smart Images
Figure CN117130281B_ABST