Research On Tool Wear Condiction Regniton Based On Acoustic Emission And Neral Network

In manufacturing, tool is one of the most basic important factors, which directly impacts the quality of workpiece, efficiency of production and the cost of manufacture etc. Therefore, tool condition monitoring has become one of the most significant parts in the whole manufacturing process. Some figures show that tool condition monitoring can reduce halt time caused by human and technology mistake 75%, increase effective machining time from non-monitoring system's 10% to 65% and increase availability of lathe to more than 50%.In this paper, Acoustic Emission (AE) signal was used as access to study tool wear monitoring. By experiment, a large numbers of AE signal were acquired. According to the characteristics of AE, wavelet packet decompose was proposed to extract the information related to tool wear from the data, and the analysis result shows that wavelet packet decompose is an ingenious way to AE analysis, as the extracted information rightly reflects the condition of the tool.Neural network is one of the main methods of pattern recognition. In this paper, neural network was built to reflect the relationship between tool wear and AE signal, and finally, implement the recognition of tool wear. Meanwhile, the number of hidden layer, neuron and network training were also discussed.LabVIEW is a sort of graphic program language, which grows fast and is largely applied in virtual instrument area. In this paper, a plane based on LabVIEW was built to carry out AE analysis and BP network recognition. The plane calls Matlab script to combine LavVIEW with Matlab, and makes tool wear recognition much simpler, more convenient and integrated.