Research on risk prediction of automobile friction noise based on neural network
Friction noise was an important factor causing abnormal noise in the entire vehicle,which could affect the quietness of the vehicle.To effectively determine the risk of friction noise,loudness data was obtained through material to material friction noise testing.The loudness classification for the loudness time curve was set,and a characteristic friction noise risk parameter SLCP value was obtained through numerical integration.Using this parameter as the input layer and the subjective evaluation of abnormal noise risk during friction testing as the output layer,a neural network model was established,and a method for predicting friction abnormal noise risk was formed through sample data training.The validation results indicated that the prediction model was effective and could ensure a certain degree of accuracy.
NETL
NSTL
万方数据
维普
