Cutting surface roughness prediction method

The invention discloses a cutting machining surface roughness prediction method, which comprises the following steps of: firstly, acquiring vibration signals of a tool and a workpiece in a machining process, then performing noise reduction by using a third-order Symmlet wavelet generating function, and acquiring signals after wavelet packet noise reduction treatment; secondly, time domain feature extraction is carried out on the signals after noise reduction processing, twelve time domain feature values are obtained, then a Relief-F feature selection algorithm is used, vibration signal feature value screening is carried out on the basis of surface roughness value influence weights, and two vibration signal feature values with the maximum influence weights are screened out; thirdly, forming training data by using the screened feature vectors and the corresponding surface roughness, constructing an SVM regression model, introducing a Gaussian RBF kernel function, inputting the training data into the RBF-SVM reg