Research On Tool Condition Recognition Based On Acoustic Emission In Milling

From the machine structure, there are many parts or structures Influence the accuracy of the vertical milling complex machining center, however the Spindle and Cutting Tools are the most directly parts. Just for the cutting tools, the method of the use of Motor current Signal and AE signal to monitor the state of the tools are the most effective method。So According to the requirements of the program of the tutor, the focal point study of the paper is discussing the tooling monitor system which based on AE signal.AE signal is the physical phenomenon of the some materials. AE as an effective Non-destructive testing (NDT) technique has apply to all kinds of the inspect devices. With the advantages of the AE just fit the online test of the tools, such as high sensitivity and without stop during the works. We will make use of the AE signals to study state the tool abrasion based on the researcher of the technology of the inspect of machine tool with the high speed. So we will carried out our works form several aspects as following:(1) Build up the tool wear monitoring experiment system in milling. Including introduce how to construct the system's hardware and software environment with detailed. We do some preparations so that we can gather the AE signals which we will analyze and process in the following steps.(2) Detect and collect the acoustic emission signals of different wear stages through using AE sensor and data-acquisition card, in the environment of the LABVIEW, then the AE signals were analyzed by statistics analysis and power spectrum. In order to search characteristic component which changes with the state of tool wearing that existing in the AE signals. At the same time we will prove that it is feasible to use AE signals to monitor the tool states online.(3) According to the system of the hardware and soft ware, we will analysis the AE signals which gathered by the system. So that can get level information of the influents of the tools factors. So that can do some conclusions(4) Because of wavelet transform possess the function of analyze the high-frequency signal, so we will use the multi-resolution wavelet decomposition to analyze and process the AE signals. Then do energy statistics to the part of the frequency band. So that can extract the features frequency that reflected the tools wear state. And using the results of this step as the inputs the tools states. After analyzing of the characteristic of the AE signals, we use the RBF neural network to build the system to the tool wear states. And conclude the result on the all system。...