An adaptive cooling control method for industrial personal computer based on thermal response neural network
By constructing a coupled control LTCs model and optimizing the heat dissipation control parameters of the industrial control computer, the problems of temperature control lag and energy waste under multiple heat source coupling conditions were solved, and stability and energy efficiency were improved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- KONGXUN TECH (SHENZHEN) CO LTD
- Filing Date
- 2026-03-30
- Publication Date
- 2026-06-12
AI Technical Summary
Existing industrial computer heat dissipation control methods lack unified modeling of the coupling relationship of multiple heat sources, making it difficult for the temperature control process to reflect the actual thermal evolution law. The control strategy has lag and energy waste, and lacks a continuous optimization mechanism.
A coupled control LTCs model based on thermal sensing neural network is constructed. By using the continuous-time thermal state input sequence and the heat source coupling structure, dynamic time constant and thermal evolution trajectory are generated, heat dissipation control parameters are optimized, and continuous updating thermal state evolution calculation is realized.
It improves the foresight and stability of temperature control, reduces the amplitude of temperature fluctuations, enhances energy efficiency, and improves the response continuity and reliability of heat dissipation control through a closed-loop regulation mechanism.
Smart Images

Figure CN122195233A_ABST