Grinding is an important part of high-precision process,and the factors affecting size accuracy of grinding workpiece are complex.Aiming at the problem that the traditional prediction model cannot be used to accurately predict the trend change or the prediction effect is poor,and the prediction accuracy is not high,through the analysis of the grinding process,the applicability of the size prediction technology was studied,and a prediction model combining wavelet transform and time series analysis was proposed.It is verified by experiments that the average prediction error of the wavelet time series model is less than 1 μm,the average absolute error is 0.105,the root mean square error is 0.185,the average absolute percentage error is 0.159.The experiment proves the accuracy and feasibility of the grinding size prediction technology based on wavelet time series model.