Research On Audio Signal And Tool Wear Based On Turning

With the continuous development of science and technology,the rapid development of the automotive industry and machinery manufacturing industry provides an opportunity.Due to the rapid development of the industry,higher requirements are putting forward for the quality of production,production efficiency,complexity of structure and performance of various products.Tool wear condition monitoring technology in NC turning is one of the core parts of intelligent manufacturing technology.By studying and analyzing the sound signals generated in the actual turning process,the specific wear condition of the tool can be judged.This method can be used to measure the actual wear condition of the tool and work-piece surface in real time without stopping the machine.On-line monitoring plays a decisive role in improving the processing quality and production efficiency of mechanical products.Intelligent manufacturing technology is realized through three aspects: first,remote control of machine tool processing and detection of different states of NC machining by means of communication and sensor equipment;second,adjustment of processing parameters and tool compensation and tool change by monitoring tool wear,so as to ensure product processing quality and addition.The third is to collect the characteristic signals of tool under different wear states independently,and fuse these signals through neural network and multi-sensor technology,so as to realize the monitoring of tool wear and breakage.The specific research work of this paper is as follows:(1)Establishing the database of typical no-load state of machine tools.Firstly,the speed signal under no-load condition is segmented,then the speed signal is analyzed,and finally the database system under no-load condition is established.(2)Establishing a monitoring system of audio signal and cutting force signal.Firstly,the advantages and disadvantages of different monitoring methods are analyzed,and the application of this method is analyzed.Then,how to study tool wear through these methods at home and abroad is analyzed.Finally,according to the analysis and research of their respective research methods,it is proposed to analyze the sound signals of tool wear at different wear stages to realize tool wear condition monitoring.Measurement.(3)Designing the overall scheme of hardware and software of tool monitoring system.Firstly,the overall scheme is designed,and then the selection and installation of hardware,including the selection and installation of machine tools,cutting tools,processing materials,sensor devices and data acquisition card,are carried out.Then,the software is designed by using LABVIEW,including the design of sound acquisition module and analysis module.(4)Analyzing the correlation between tool wear status and cutting audio signal and force signal.In order to ensure the accuracy and practicability of the analysis,it is necessary to eliminate the environmental noise in the collected actual cutting sound signals.The processed sound signals are then analyzed in time and frequency domains,and the collected cutting forces are analyzed in X and Y directions.Finally,the cutting audio signals are obtained.Relativity with cutting force signal and tool wear.(5)Extracting the feature vectors related to tool wear state.Firstly,the processed audio signal is decomposed by wavelet transform,and the feature vectors of the audio signal are extracted.Then the cutting force in X and Y directions collected by Instrument for measuring force is taken as the feature vectors.Finally,the feature vectors of the audio signal and the cutting force signal are extracted.(6)Setting up the mathematical model of tool wear state,audio signal and cutting force signal.In this paper,BP neural network is used to fuse the extracted feature vectors,and the mathematical model of the comprehensive situation is established to realize the monitoring and discrimination of tool wear.