Researches Of Gear Fault Feature Extraction And Classification Using Multi-dimension Multi-scale

Gear fault diagnosis is the key part of the signal characteristics of the two extractionand pattern recognition. The comprehensive analysis of the research status issues at homeand abroad for gear fault diagnosis technology, based on the vibration signalcharacteristic vibration mechanism and gear on gear fault, we propose amulti dimensional and multi scale feature extraction algorithm based on fuzzy clusteringC The fault classification algorithm. Specific contents are as follows:(1) In summing up the gear vibration signal characteristics under differentconditions, based on the use of simulation signal, analogThe gear normal wear and broken teeth vibration signal characteristics, more intuitivegrasp of the vibration signal characteristics of different faults, but also for faultclassification provides a simulated signal.(2) proposed a multi dimensional multi scale feature extraction algorithm. Signal tobe processed through the singular value of the firstSolution, the one dimensional signal re constitute multidimensional signal, highlightingthe more feature information; then using singular value decomposition, thedecomposition into different scales AM FM signal, take the product of high energysummation function reconstruction; Finally, morphological difference between the filterfeature information extracted through simulation and gear fault simulation resultsdemonstrate the effectiveness of this method.(3) On the basis of feature extraction, based on the fuzzy C clustering algorithm isintroduced to fault classification, from the extractionFeature information in choosing the right amount of features, through simulation andgear fault simulation experiments to verify the fuzzy C clustering algorithm is anefficient fault classification algorithm.This paper presents an efficient fault feature extraction algorithms and fault classificationalgorithms, but also to verify its effectiveness, provide an effective and accurate methodof fault diagnosis gears.