Mechanical properties prediction of tire cord steel via multi-stage neural network with time-series data
Long Chen, Fei He
Abstract
ABSTRACTCord steel is a kind of high-quality wire, whose mechanical properties will affect the safety and service life of tire. Therefore, the prediction model of mechanical properties during production process is very important to ensure the quality stability. In the paper, the Multi-Stage Neural Network with Time-Series data (MSNNTS) is proposed to mine the rich information of high-resolution time-series data and represent multistage process to achieve accurate mechanical properties prediction. According to the results, the best mean relative error, for tensile strength prediction, is about 1.25% and the hit rate with 3% error limit is about 98% on the testing set. It also obtains good results in predicting reduction of area. The results show that the method is of great significance to improve the quality stability and uniformity of cord steel.KEYWORDS: Time-series dataneural networkdeep learningdata miningmechanical propertiescord steelsteel rollingmultistage process Disclosure statementNo potential conflict of interest was reported by the author(s).Additional informationFundingThis research is supported by the National Key Technology R&D Program of the 12th Five-year Plan of China [grant number 2015BAF30B01]. This research is supported by the Fundamental Research Funds for the Central Universities [grant number FRF-AT-20-04].