Method and device for detecting air leak in sintering machine pallet

The method employs a machine learning prediction model to differentiate between operational and blending state-induced fluctuations and actual air leaks in Dwight Lloyd sintering machines, enhancing the accuracy of air leak detection and reducing false positives.

WO2024236889A1PCT designated stage expired Publication Date: 2024-11-21JFE STEEL CORP
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
PCT/JP2024/008644
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-05-18
Filing Date
2024-03-07
Publication Date
2024-11-21

AI Technical Summary

Technical Problem

Existing methods for detecting air leaks in Dwight Lloyd sintering machines fail to accurately distinguish between fluctuations in oxygen concentration caused by changes in the operating state or raw material blending and actual air leaks, leading to false detection and reduced productivity.

Method used

A method and device using an oxygen concentration meter and a machine learning prediction model to calculate the predicted oxygen concentration based on operational and blending state information, allowing for accurate detection of air leaks by subtracting the predicted value from measured oxygen concentrations.

Benefits of technology

Enables high-accuracy detection of air leaks by isolating variations due to operational and blending state changes, reducing Mean Absolute Error and improving the reliability of air leak identification.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure JP2024008644_21112024_PF_FP_ABST
    Figure JP2024008644_21112024_PF_FP_ABST
Patent Text Reader

Abstract

A method for detecting an air leak in a sintering machine pallet according to the present invention comprises: a measurement step for using an oxygen concentration meter, installed in a wind leg directly below a plurality of rotating pallets, to measure, for each revolution of each of the pallets, the oxygen concentration in the wind leg, when each of the pallets passes directly above the wind leg; a prediction step for calculating a predicted value of the oxygen concentration by inputting information at the time of processing to a prediction model in which pieces of information indicating the operation state of a Dwight-Lloyd sintering machine and the blending state of sintering raw materials are input variables and the predicted value of oxygen concentration is an output variable; and a detection step for subtracting the predicted value of oxygen concentration calculated in the prediction step from the oxygen concentration measured for each revolution of each of the pallets in the measurement step, and using the difference to detect a pallet in which an air leak has occurred.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Method for identifying air-leaking part in sintering machine pallet

    JP2012036477A

  • History management method of sintering machine pallets

    JP2014122380A

  • All-directional sintering process intelligent management and control system

    CN113253618A

  • Method for measuring wind leakage ratio of sintering machine

    JP1985228620A

  • Detection of wind leakage in sintering installation

    JP1986195929A