Research On Cutter Condition Monitoring Based On The Signal Parameters Database

In the milling process,the tools signal is obviously affected by many factors,suchas tools,workpiece material,cutting parameters. It is different to find a fixed thresholdto determine the tool condition in complex milling conditions. However,the millingcondition in aviation unit is relatively centralized.The tool generally is hard alloytool, the workpiece generally is aluminium-alloy, and the three factors of cutting aremore optimized and concentrated. In this background, the program of proposing afixed threshold in a variety of cutting parameters is feasible. So the milling toolsdiagnosis based on signal features database is proposed in this paper.First, we divide the degree of the tool wear according to the amount of tool wear,determine the type of experiment (the sensor mounting position on the signalcharacteristics of the experiment, the tool wear monitoring experiment, random testingexperiment),and then determine the cutting parameters in each type of experiment.Second, we set up data acquisition system that acquisition vibration signal andacoustic emission signal in different status of tool. The system is erected by graphicalprogramming software LabVIEW.Third, we analyze the vibration signal and acoustic emission signal from theirtime domain and frequency domain.The result of the analysis is that the status of toolwear is related to RMS value of the time-domain and wavelet band energy,so we makethe RMS value and wavelet band energy to be threshold that monitor the status oftool.However,the cutting parameters have influence on signal features.The element incutting parameters that has the biggest influence on vibration signal feature is theamount of back cutting tool.The element in cutting parameters that has the biggestinfluence on acoustic emission signal feature is spindle speed. So the amount of backcutting tool and spindle speed need to be classified. In each amount of back cuttingtool there is a fixed RMS value of vibration signal as threshold. In each spindle speedthere is a fixed RMS value of acoustic emission signal as threshold. In thetime-frequency domain,those three frequency bands of vibration signal that bestreflect the tool condition are used as feature frequency band. In the acoustic emissionsignal,the second frequency band that best reflect the tool condition are used as featurefrequency band. The value of this frequency band is used as threshold that judge thestatus of tool. Those thresholds in any cutting parameters should be deposited intoAccess database in order to establish signal feature database, then that signal featuredatabase can provide the supporting of thresholds.Finally, through the test of milling experiment, according to the cutting conditionthe monitoring system recalls the corresponding threshold,and then compares thesethresholds with signal features that are analyzed, the result indicates that this diagnosisbased on signal feature database is effective on judging the status of tool.In this paper, there is active exploration in designing of system platform,thedesigning of experiment, the extraction of vibration signal feature and acoustic emission signal feature. It has enhanced practicability and reliability of data, hasimproved accuracy of smart monitoring. Therefore, it has enhanced productionefficiency and security.