Fault Diagnosis Of Mine Gearbox Based On Multi Feature Fusion And IGWO-MSVM

With the deepening of China's industrialization,mechanical equipment,as an indispensable tool for industrial production,have been paid more and more attention.In the operational process of coal mining,in addition that gas permeation and water disasters can have a great impact on the mine and even the entire coal mining enterprise,the smooth operation of mining mechanical equipment is also the key to safe production of coal mining.As the main mechanical equipment in the coal mining process,the safe operation of mine gearbox is the basis for the normal operation of the whole mine and the safe production of the whole coal mine enterprise.Therefore,from the perspective of management and maintenance,this thesis diagnoses the fault of the mine gearbox through the vibration signal of the running state of the gearbox,so that the coal mining workers can do the maintenance and prevention well in advance,which is of great significance for the safe mining of the coal mine.First of all,due to the poor operating environment of mine gearbox,vibration signal contains a lot of noise during operation,which has the characteristics of nonlinearity,instability and uncertainty.The original signal collected needs to be preprocessed.Through the analysis of the internal mechanism of the fault of the mine gear box,the wavelet packet decomposition and reconstruction are used to reduce the noise of the gearbox fault signal and remove the unrelated external signals in the vibration signal so that the vibration signal is closer to the vibration produced by its own operation,which provides the basis for the extraction of the signal characteristics of the mining gearbox.Secondly,through the fault signal after noise reduction,the high and low frequency coefficients of the fault signal determined in the wavelet packet decomposition process are used to calculate the energy entropy of the fault vibration signal.Then the sample entropy of the reconstructed frequency band of each signal is calculated,and the multi-feature of the vibration signal is extracted by the mean squared error of the signal.The multi-feature vector of mining gearbox extracted by energy entropy,sample entropy and statistics can turn the fault signal of the gearbox into a quantifiable feature structure,which can be used as the input of the diagnosis model and the key to the construction of fault diagnosis model.Finally,a mining gearbox fault diagnosis model based on improved grey wolf optimization algorithm and optimized multi-classification support vector machine is established.The gray wolf group location updating is improved by differential evolution algorithm to improve the global search ability of gray wolf optimization algorithm.Then the improved gray wolf algorithm and the multi classification support vector machine are combined to find the optimal parameters of the multi classification support vector machine by improving the global search ability of the gray wolf optimization algorithm,and then improve the accuracy of the fault diagnosis model.Then the case analysis is used to verify that the fault diagnosis model is applied to the fault diagnosis of the mining gear box.The research results of this thesis not only enrich the research theories of mine gearbox fault diagnosis,but also provide an effective theoretical guidance for the safe production and operation of coal mine machinery.And it helps to improve the operating performance of coal mining machinery and has a practical significance for the prevention and maintenance of the faults of coal mining mechanical equipment.