On-line Intelligent Prediction Model of Surface Roughness in Cylindrical Grinding

摘要

On-line measuring workpiece surface roughness is still a key issue for grinding now. A new intelligent prediction model is developed in this paper. This model bases on the theory of roughness during cylindrical grinding and the theory of fuzzy-neural network. The inputs for the model are the grinding conditions, such as feed and speed, and the vibration data. An accelerometer is used to gather the vibration signal in real time. The model is used in the grinding experiment, and the accuracy is 98.81%. This verifies the feasibility of the proposed model.