Prediction research for surface topography of internal grinding based on mechanism and data model

Grinding surface topography is an important evaluation index of grinding surface integrity. In order to realize the prediction of the grinding surface topography, a three-dimensional theoretical model of the surface topography of internal grinding was established in terms of the mutual movement between the grinding wheel and the workpiece and a data model for the surface roughness of internal grinding was set up based on the grey wolf algorithm-support vector machine (GWO-SVM). Taking the inner ring of rolling bearing as an example, the grinding experiment was carried out to analyze the influence of technological parameters on the surface topography and surface roughness. Three traditional data models were compared with the GWO-SVM data model, the comparison results showed that the GWO-SVM data model had a small change in the relative error of the predicted value of the test sample, the maximum relative error was 5.42%, and the average relative error was 3.95%. Experimental results were compared with the results of theoretical model and data model, the results show that the error of both the theoretical model and the data model is within 10%. This work should be helpful in reflecting the formation mechanism of the surface topography of the internal grinding and provide theoretical and technical support for high-precision and high-efficiency manufacturing of high-end rolling bearing.