Dry clutch temperature prediction method based on dynamic neural network time sequence prediction

• Main Authors: ZHANG LIJIE, XIANG ZHILI, CHEN CAI, GONG YUBING, ZHOU HONGDA, ZHENG XIANLING, YIN YUTIAN
• Format: Patent
• Language: Chinese; English
• Published: 28.05.2021
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; PHYSICS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR

The invention discloses a dry clutch temperature prediction method based on dynamic neural network time series prediction, and the method employs the clutch temperature and time historical sample data to carry out modeling and prediction of the clutch temperature through employing the dynamic neural network time series prediction method. Firstly, data acquisition is carried out; data is trained, and a dynamic neural network time sequence model is created. clutch temperature is predicted on a future time sequence, a prediction result error is analyzed and reverse normalization is performed on prediction data; finally, a clutch temperature prediction model and a predicted value on a time sequence are acquired. Compared with a traditional test method and a finite element numerical simulation method for obtaining the temperature of the clutch, the invention has the advantages of being easy and convenient to implement, high in precision, low in cost and the like, has a memory function, is very suitable for process