A method, device and equipment for predicting the residual life of an oxygen lance and a storage medium

By constructing health indicators based on historical oxygen flow and pressure data under oxygen lance operating conditions, and conducting T-tests and training linear degradation models, the data quality problem in predicting the remaining lifespan of oxygen lances in existing technologies has been solved, achieving more accurate predictions.

CN122333933APending Publication Date: 2026-07-03INSPUR YUNZHOU (SHANDONG) IND INTERNET CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
INSPUR YUNZHOU (SHANDONG) IND INTERNET CO LTD
Filing Date
2025-01-03
Publication Date
2026-07-03

AI Technical Summary

Technical Problem

Existing technologies require various environmental factors as input when predicting the remaining life of oxygen lances. The data quality is generally poor, the preprocessing is cumbersome, and the remaining life of oxygen lances in metal smelting environments cannot be accurately predicted.

Method used

By constructing health index data based on historical oxygen flow and oxygen pressure data under the working state of the oxygen gun, a T-test and linear degradation model training were performed. Prior and posterior model parameters were obtained by using weight allocation and noise variance processing for prediction.

Benefits of technology

It improves the accuracy of oxygen lance remaining life prediction, reduces the impact of invalid data on model training, and makes the prediction results closer to the true value.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122333933A_ABST
    Figure CN122333933A_ABST
Patent Text Reader

Abstract

This application discloses a method, apparatus, device, and storage medium for predicting the remaining life of oxygen lances, relating to the field of metal smelting technology. The method includes: constructing health index data based on historical data of oxygen flow rate and oxygen pressure corresponding to each oxygen lance in its working state; performing a first T-test on initial training data obtained from the health index data to obtain target training data that meets a first preset linearity condition; training an initial linear degradation model using the target training data; obtaining prior model parameters by weighting the obtained model parameters; determining posterior model parameters based on cumulative state variables and prior model parameters obtained from real-time data corresponding to oxygen flow rate and oxygen pressure; and predicting the remaining life of the oxygen lance using the posterior model parameters. Training the initial linear degradation model using the target training data corresponding to oxygen flow rate and oxygen pressure reduces the impact of invalid data on model training, thereby improving the accuracy of the prediction.
Need to check novelty before this filing date? Find Prior Art