A method and device for monitoring and diagnosing faults in float glass production process

By using the fault monitoring method based on attention mechanism TCN and TS-DKCVA models in the production process of float glass, we collaboratively analyze the spatiotemporal information in the production process of float glass, solving the problem of low accuracy of fault detection in the existing technology, and achieving efficient fault monitoring and diagnosis.

CN118396404BActive Publication Date: 2025-04-04UNIV OF SCI & TECH BEIJING
2 Cites 0 Cited by

Patent Information

Application Number
CN202410433444.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-04-10
Publication Date
2025-04-04
Estimated Expiration
2044-04-10

AI Technical Summary

Technical Problem

The existing fault monitoring methods for floating glass production process fail to effectively coordinate the analysis of time and space information, resulting in a low accuracy of fault detection.

Method used

The time convolution network (TCN) and spatiotemporal distributed kernel specification variable analysis (TS-DKCVA) model are adopted based on attention mechanism, and combined with Bayesian fusion algorithm, fault monitoring and root cause diagnosis methods are constructed. This method can realize fault monitoring and diagnosis by capturing the causal relationship between process variables and collaborating on the spatiotemporal information.

Benefits of technology

Improves the accuracy and efficiency of fault monitoring, saves computing resources, and can accurately identify the root cause and propagation path of faults.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The present invention discloses a method and device for monitoring and root cause diagnosis of float glass production process faults, which belongs to the technical field of industrial process monitoring. The method comprises: obtaining float glass historical production process data; based on the historical production process data, using a preset causal discovery framework to capture the causal relationship between each process variable, and constructing a causal relationship diagram; based on the historical production process data, constructing a float glass production process fault monitoring model; obtaining float glass real-time production process data, and using the float glass production process fault monitoring model to perform fault monitoring on the current float glass production process according to the real-time production process data; when a fault is detected, determining the cause of the fault based on the monitoring result and the causal relationship diagram. The present invention can improve the accuracy of float glass production process fault monitoring, reduce the false alarm rate, accurately identify the root cause and propagation path of the fault, and improve production efficiency by modeling the float glass production process.
Need to check novelty before this filing date? Find Prior Art