Research On Spindle State Perception And Evaluation Technology Under Cutting Condition

CNC machine tools are the main equipment for manufacturing complex and precise parts.As the key component of CNC machine tools,the performance of the spindle will affect the overall reliability of the machining center.At present,most of the research on the spindle is carried out under idling conditions.There is a big difference between the state of the spindle and the idling under the condition of the spindle,so this paper mainly studies the spindle under the cutting condition.Due to the complex internal structure of the spindle,it is often difficult to monitor its running state.At the same time,affected by factors such as the processing environment,a single monitoring signal often cannot effectively identify different faults of the spindle.Therefore,this paper studies the rotor sensing sensor to monitor the spindle state,and studies the multitype spindle monitoring signals to identify the spindle fault.The main research contents are as follows:(1)The dynamic model of the spindle system is established by the transfer matrix method,and the natural frequency is calculated by programming the spindle system with MATLAB software.A comparative analysis of the variation of the natural frequency for three different cases of spindle only,toolholder mounted and tool mounted.Then use the finite element software ANSYS to calculate the natural frequency.Errors were compared between the results obtained by the transfer matrix method and the finite element method,and the critical rotational speed of the spindle was also analyzed.Finally,with the help of hammering modal experiment and finite element modal analysis technology,the dynamic characteristics of the tool holder after the rotor sensing sensor is installed are evaluated.(2)Research on the spindle rotation accuracy under idling and cutting conditions based on the axis trajectory is carried out.First,the spindle signal is collected by the eddy current displacement sensor,and then the rotation accuracy value is evaluated after error separation.At the same time,a new rotor sensing sensor is used.The sensor can be directly installed on the spindle and rotate with the rotation of the spindle,which can better sense the state signal of the spindle,and evaluate the rotation accuracy of the spindle through signal analysis.Finally,in the idling condition,the rotation accuracy values calculated by the two sensors are compared with the values measured by the American Lion instrument to verify the effectiveness of the method.At the same time,two kinds of sensors are used to measure the rotation accuracy of the spindle in cutting state.(3)Use multi-source signals to identify spindle faults,extract features from the signals,and screen the selected features to screen out features that can clearly distinguish different spindle faults,and then reduce the dimensions of the selected features to form features The set input is input into the neural network for training and testing,and the analysis results show that the multi-source signal can more accurately identify the different faults of the spindle than the single signal.(4)The spindle state evaluation system is built using LABVIEW and MATLAB software.The main functions include signal acquisition,signal processing,error separation,rotation accuracy evaluation,and fault identification.Using this system,the rotation accuracy value of the main shaft can be observed in real time,and the fault of the main shaft can also be diagnosed.The feasibility of the system is verified by comparing with the American Lion instrument through experiments.