Research On Bit Wear Monitoring Technology Based On GA-LVQ Network

When laminated materials are used for drilling,the particularity of the materials makes the bit wear very severe,which has a great impact on the quality of the workpiece.Therefore,accurately judging the wear state of the bit in the process of processing is the key to ensure the quality of processing.In this paper,according to the characteristics of drilling and machining of laminated materials,the mechanism of bit wear and tool monitoring technology are theoretically analyzed.Acoustic emission signal and temperature signal are selected as monitoring signals of bit wear,and an experimental platform is set up to carry out experiments and collect processing signals of the whole life cycle of bit.Secondly,wavelet packet decomposition algorithm is used to decompose acoustic emission signal in four layers to extract relevant energy eigenvalues.The eigenvectors are composed of rotational speed,feed,amplitude,average temperature,and frequency band energy.The eigenvectors are processed by principal component analysis to obtain the input vectors of the monitoring model.Finally,BP and LVQ neural network were used to establish the tool wear monitoring model,and the genetic algorithm and its improved algorithm were used to optimize the initial parameters of the tool wear monitoring model,so as to improve the operation speed and prediction accuracy of the model.The experimental results show that the operation speed and accuracy of LVQ network model are 2.3 times and 7.5%higher than those of BP network model,which is more suitable for tool wear monitoring model of drill bit.The improved annealing genetic algorithm is used to improve the stability of the model after optimization.Meanwhile,the training speed is increased by 7.2 times and the accuracy by 12.5%,which can better realize the on-line monitoring of tool wear status.