Prediction Of Tool Wear In Ultrasonic Vibration-assisted Drilling Of CFRP

Ultrasonic vibration-assisted drilling(UVAD)is an effective processing method for Carbon Fiber Reinforced Polymer(CFRP).It can reduce the torque and cutting force during machining,reduce the surface roughness value of the machined surface,reduce machining defects such as burrs,and relieve tool wear.But like traditional drilling,UVAD also inevitably suffers from tool wear.The wear of the tool is closely related to the machining quality,which directly affects the probability of machining defects such as burrs,fiber tearing,and hole wall scratches.Therefore,predicting the wear of the tool during the making hole process plays an important role in improving the quality of the making hole,using the tool fully and reasonably,improving the production efficiency,reducing the production cost,and reducing the accidents in the production process.Combining the advantages of particle filter in machining process modeling and tool wear prediction performance,this thesis proposes a tool wear prediction method based on the theoretical model of particle filter algorithm and data fusion to solve the problem of ultrasonic-assisted drilling CFRP tool wear prediction.The main research contents of the full text are as follows:1)UVAD tool wear state monitoring experiment.Based on the basic understanding of the interaction between the tool,workpiece,and chip in the process of UVAD of CFRP,the causes of different tool wear profiles are analyzed,and the wear mechanism of the flank face is emphatically analyzed.The average wear width of the flank is used as a measure of tool wear.On this basis,an experimental platform for the UVAD tool wear experiment was built,and the tool wear experiment under the UVAD condition was carried out.2)Theoretical model of UVAD tool wear prediction.Aiming at the problem that the current wear model fails to consider the special motion state and force of the tool under UVAD,through the analysis of the kinematics and mechanical properties of the cutting edge of the tool in the UVAD,a suitable twist drill is established.The semianalytical theoretical model for predicting the wear width of the tool flank is analyzed and verified with the experiment.3)A data-driven model for UVAD tool wear prediction.To overcome the interference of ultrasonic vibration on the sensor-based data model,the relationship between the wear amount of the tool flank and the axial force and vibration signal was established.Firstly,noise reduction and sensitive area extraction are performed on the monitoring signals in the above experiments.Then,feature extraction is performed on the original signal of the sensitive area and the correlation coefficient method is used to remove the features with low correlation with tool wear.Finally,the multi-layer perceptron machine learning method is used to predict the tool wear through the extracted features,and a relatively stable prediction effect is obtained.4)Tool wear fusion prediction method based on particle filter algorithm.Theoretical models fail to reflect all wear mechanisms,while data-driven inference models do not consider theoretical properties and rely excessively on data quality and sample size.To improve the tool wear estimation results,the theory-based model is regarded as the state equation,the sensor data-based data-driven model is regarded as the observation equation,and a method of fusing the results of the two is proposed using particle filtering.By comparing the results of the fusion model with those of the theoretical model and the data-driven model,the prediction error is significantly reduced,verifying the feasibility and superiority of the fusion method.In this paper,through the combination of theoretical analysis and experimental research,the problem of tool wear prediction in UVAD of CFRP is deeply studied,and a tool wear prediction method based on the fusion of theoretical model and data-driven model based on particle filter algorithm is obtained.It is of great significance to make a more accurate prediction of the tool life in the UVAD of CFRP,to make full and reasonable use of the tool under the condition of ensuring the machining quality.