Application of Grey Linear Regression Combined Model in Predicting the Motor Oil Wear Particles for Passenger Cars

Abstract

: A new model is established by combining the grey model and the linear regression model to synthesize the advantages of the two models, and then the number of oil wear particles in passenger cars is predicted. The three models are used to predict and compare the particle content of different levels of passenger car oil. The prediction results of wear particles in the SJ oil for No.1 passenger car show that the prediction accuracy of the grey linear regression combined model are higher than the linear regression model (1.85%) and the grey model (0.29%), and for the SL oil are 1.34% and 0.45%, respectively. For No.2 passenger car, the prediction accuracy is increased by 2.86% in SJ oil and 1.28% in SL oil for the linear regression model, and 0.12% in SJ oil and 2.62% in SL oil for the grey model. The results indicated that the combined model has better prediction effect, and it can be applied to the prediction of oil wear particles in passenger cars. Through the prediction of combined model and the judgment of cleanliness grade, it can provide the basis for automobile to replacement oil by quality.