Prediction of Torque Parameters in Automobile Rear Axle Assembly Based on Long Short-Term Memory

摘要

The effective prediction of assembly parameters in automobile rear axle assembly process can reduce the time of fault treatment and improve the efficiency of rear axle assembly, which is of great significance to enterprises. Based on the historical data of assembly quality parameters, this paper proposes a method for predicting the assembly quality parameters of automobile rear axle based on Long Short-Term Memory, including network structure design, network training and prediction process implementation algorithm. The values of torque parameters at multiple time steps of the station are predicted, and by comparing it with the normal range of torque. Thus, the quality of assembly products in multiple time steps of the station is predicted.