Boiler heating surface tube wall state online analysis method and system based on multi-source data fusion

By using a multi-source data fusion method and leveraging CFD proxy models and deep learning regression models, real-time data on the local micro-environment of the boiler is obtained, solving the problem of real-time prediction and accurate forecasting of boiler heating surface faults, and realizing online analysis and predictive maintenance.

CN122154513APending Publication Date: 2026-06-05SHANDONG HUADIAN ENERGY CONSERVATION TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SHANDONG HUADIAN ENERGY CONSERVATION TECHNOLOGY CO LTD
Filing Date
2026-01-06
Publication Date
2026-06-05

AI Technical Summary

Technical Problem

Existing technologies cannot achieve real-time prediction and accurate forecasting of boiler heating surface failures, resulting in a lack of scientific maintenance strategies, failure to meet online analysis requirements, and failure to utilize the correlation between wear rate and operating parameters in real time.

Method used

By employing a multi-source data fusion method, and through a CFD proxy model and a deep learning regression model, real-time local micro-environment data of the boiler is acquired and fused with macro-data to achieve online quantitative analysis of the boiler's heating surface tube wall condition.

Benefits of technology

It significantly improves the accuracy and reliability of wear rate prediction, meets real-time requirements, and realizes the transformation from post-event analysis to in-event early warning, providing a scientific basis for predictive maintenance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122154513A_ABST
    Figure CN122154513A_ABST
Patent Text Reader

Abstract

The present application belongs to the technical field of boiler heating surface early warning, and specifically proposes a multi-source data fusion boiler heating surface tube wall state online analysis method and system. It includes collecting real-time operation data of the boiler; inputting the real-time operation data of the boiler into a CFD proxy model to obtain real-time local field data of the boiler combustion; inputting the real-time operation data of the boiler and the real-time local field data of the boiler combustion into a deep learning regression model to obtain the predicted wear of the current boiler heating surface tube wall, thereby realizing online analysis of the boiler heating surface tube wall state. The present application effectively fuses local microscopic environment and global macroscopic data, provides high-precision, physically meaningful direct input features for the wear model, significantly improves the prediction accuracy, greatly improves the accuracy and reliability of the wear rate prediction, makes the model more physically interpretable, and meets the real-time requirements.
Need to check novelty before this filing date? Find Prior Art