Study of the lubrication oil consumption prediction of linear motion guide through the grey theory and the neural network

To determine the lubrication oil consumption change of linear motion guide under some mileage, this study designs a test machine to fasten the linear motion guide. The grey prediction model GM(1,1) and neural network are employed for comparison and exploration. Through this study we can understand the differences in prediction of lubrication oil consumption between neural network and grey theory. Experiment results will serve as reference for manufacturers and users for the purpose of quality improvement and selection of better linear motion guides. Based on fewer measurement data, the outcome can be more accurately predicted, and that with a nondestructive test can accurately predict the lubrication oil consumption of the linear motion guide. The outcome indicates that the prediction model of neural network is superior to the grey theory model GM(1,1). The average prediction error of neural network prediction is around 2% - showing a very high accuracy level.