Research On Key Technology Of Milling Cutter Breakage Detection Based On Vibration And Permanent Magnet Perturbation Sensor

During the milling process,due to the influence of cutting factors,tool and workpiece materials,the milling cutter often occur brittle damage,and affect the accuracy and efficiency of processing.Therefore,it is necessary to detect the damage of the milling cutter,and if it is found,take measures such as changing the knife in time to avoid further losses.Since the vibration signal in the milling process can better reflect the state of the milling cutter,this paper uses the indirect detection method based on the vibration sensor to detect the real-time damage of the milling cutter during the machining process.This paper also proposes a direct detection method based on permanent magnet disturbance sensor,which can detect the cutter breakage in the unmilled machining interval,and can also make the second judgment on the result of the indirect detection to reduce the false positive rate.In the detection method of the cutter damage based on the vibration sensor,collect vibration signals in the three directions of machine tool spindles X,Y,and Z.This paper proposes to use the variational mode decomposition to decompose the signal and extract the center frequency and energy of each mode as modal characteristics,and combine the extracted time domain,frequency domain features and energy characteristics of wavelet packet decomposition,and use the feature selection method based on random forest model to select important features.The selected features are trained by using support vector machine and ensemble learning model.The optimal parameters of the model are determined by combining K-fold cross validation and grid search.Comparing the two models,the random forest model with integrated learning method has higher prediction accuracy.In the detection method based on permanent magnet perturbation sensor,a feature extraction method based on frequency of frequency band is proposed,and the energy value of each frequency band corresponding to the frequency multiplier is extracted.Furthermore,the detection model based on the energy entropy of the frequency-doubled frequency band is proposed.The experiment shows that the method can detect the damage of the milling cutter and achieve better results.According to the cutter damage detection method of this paper,the corresponding cutter damage detection software is developed.The software has data acquisition,time domain and frequency domain oscilloscope,eigenvalue display,command information display,milling cutter damage detection and status display,data saving,screenshots and other functions.