Research On Key Technologies Of Acoustic Emission Feature Recognition In Laser Assisted Machining State

Zirconia engineering ceramics have a wide range of applications in the fields of machinery.For example,aerospace,defense,automotive,tools,chemical engineering and casting,etc.The demand for zirconia ceramics is increasing in the market,and the quality requirements are getting higher and higher.The traditional zirconia ceramic processing method is mostly grinding,but the material removal rate of the process is low and there are different degrees of micro cracks.It is difficult to obtain high surface quality.With the emergence of laser assisted machining technology,high-efficiency plastic removal of zirconia ceramics and high-quality surface can be achieved.However,how to monitor the plastic state of its processing is still a scientific problem that needs to be solved internationally.This topic proposes to use acoustic emission detection technology to solve this problem and realize processing condition monitoring.By analyzing the influence of process parameters and different processing conditions on acoustic emission signals,the theoretical support is provided for the monitoring and automation of the processing conditions of zirconia ceramics processed by laser assisted machining.The main work done is as follows:(1)By consulting the literature,the development and application of AE detection technology and laser assisted machining technology,which are closely related to the subject,are sorted out;(2)The experimental platform was built according to the experimental requirements,including the cutting system,heating system and acoustic emission signal detection system.The model number and parameter configuration of the main equipment of each system are introduced,and the design of the experimental scheme is expounded,including the preliminary experimental exploration and the selection of the experimental parameters of the subject.The processing effect of different cutting states of zirconia ceramics is analyzed in detail;(3)Analyzing the feasibility of using acoustic emission detection technology to monitor the machining state in the process of laser assisted machining of zirconia ceramics briefly.By analyzing the acoustic emission signals corresponding to different process parameters,the law of acoustic emission characteristics changing with process parameters is found from the time domain and frequency domain.By analyzing the acoustic emission signals of different processing states in the time domain and frequency domain,the acoustic emission characteristics related to the processing state of zirconia workpieces are found.And three acoustic emission characteristics that can be used for processing state identification are extracted: RMS voltage,rise time and energy spectrum coefficient of 187.5～250KHz;(4)Combining the extracted acoustic emission characteristics,the change trend of the surface roughness of the workpiece accompanied by the change of acoustic emission characteristics under plastic working condition is analyzed.And the ideal acoustic emission characteristic value is given based on the guaranteed plastic working state.(5)Taking the extracted signal RMS,rise time and 187.5～250KHz energy spectrum coefficient as input parameters,the decision tree model learning classification model is used to accurately identify the plastic machining state and non-plastic machining state of the workpiece.Based on the BP neural network,the acoustic emission characteristics are used as input parameters to predict the surface roughness of the workpiece and achieve better results.The establishment of classification models and predictive models has accumulated valuable experience for the automated processing of zirconia ceramics.