A heat exchanger heat transfer coefficient identification method based on grey box model and pattern clustering

By using gray box model and pattern clustering method, combined with Gaussian mixture model and genetic algorithm to optimize heat transfer coefficient, the deployment problem of heat exchanger heat transfer coefficient identification is solved, realizing dynamic reflection of performance changes and accurate identification, which is suitable for real-time control of district heating systems.

CN122432741APending Publication Date: 2026-07-21TIANJIN UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
TIANJIN UNIV
Filing Date
2026-04-27
Publication Date
2026-07-21

AI Technical Summary

Technical Problem

Existing methods for identifying the heat transfer coefficient of heat exchangers have problems such as being unable to be directly deployed, ignoring thermal inertia and heat transfer delay, and failing to dynamically reflect performance changes.

Method used

A gray box model and pattern clustering method is adopted to calculate the drag coefficient by collecting data, perform clustering using Gaussian mixture model and Bayesian information criterion, and optimize the heat transfer coefficient and water volume by combining genetic algorithm to establish a dynamic heat transfer model, construct a heat transfer coefficient spectrum and embed it into the control system.

Benefits of technology

It enables deployment without cumbersome calibration and large amounts of tag data, can dynamically identify changes in heat exchanger performance, improves the accuracy and stability of heat transfer coefficient identification, and is suitable for real-time prediction and control.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122432741A_ABST
    Figure CN122432741A_ABST
Patent Text Reader

Abstract

The present application relates to the technical field of performance evaluation and fault diagnosis of heat exchangers in district heating systems, and discloses a heat exchanger heat transfer coefficient identification method based on a grey-box model and pattern clustering. The method introduces a differential equation of water heat capacity dynamics and adopts Taylor expansion linearization and discretization to establish a grey-box heat transfer model with physical mechanism and calculation efficiency, so that the model can accurately reflect the influence of terminal load change and water heat capacity on the heat transfer process when in use, thereby being suitable for real-time prediction and control. The Darcy formula is introduced to calculate the hourly local resistance coefficient, and combined with the Gaussian mixture model and the Bayesian information criterion automatic clustering, the historical data is divided into multiple typical patterns, so that the clustering result can objectively reflect the actual operating condition distribution of the heat exchanger, and the global nonlinear problem is decomposed into multiple approximately linear sub-problems, thereby eliminating the need for manual setting of the number of patterns, and further achieving the beneficial effects of improving the heat transfer coefficient identification accuracy and stability.
Need to check novelty before this filing date? Find Prior Art