Method for determining the temperature of a medium
By combining FP-AI with PDE and PINN to process temperature sensor data, the problem of response delay in the transient phase of non-invasive sensors is solved, enabling accurate estimation of the temperature of the medium and inner surface inside the pipe. This improves the real-time performance of temperature monitoring and the responsiveness of the control system, making it particularly suitable for pipelines transporting corrosive fluids.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ABB (SCHWEIZ) AG
- Filing Date
- 2026-01-14
- Publication Date
- 2026-07-17
AI Technical Summary
Existing non-invasive temperature sensors cannot provide real-time temperature estimation in the transient phase, resulting in excessively long response times for traditional methods during rapid temperature changes. This affects the accuracy of temperature monitoring and the rapid response capability of the control system, especially in pipelines transporting corrosive fluids, where the thermalization process of the sensor leads to delays and inaccuracies.
First-principles artificial intelligence (FP-AI) combined with partial differential equations (PDE) is used to process data from two temperature sensors. The temperature of the outer surface of the tube and the ambient temperature is measured in a non-invasive manner. The heat conduction equation is solved using a physical information neural network (PINN) to achieve accurate estimation of the temperature of the medium inside the tube and the inner surface.
It significantly shortens the temperature response time, provides instant and accurate temperature estimation, improves the temperature monitoring accuracy and control system responsiveness during dynamic operation, is suitable for pipelines made of low thermal conductivity materials, and ensures process safety and efficiency.
Smart Images
Figure CN122408981A_ABST