Research On Bearing Fault Diagnosis Method Of Coal Mine Main Fan Based On Vibration Data Analysis

In the safety production of coal mines,as the core equipment of the mine ventilation system,the main fan of the coal mine is responsible for the discharge of underground gas,mine dust and dirty gas,and its normal operation is an important guarantee for coal mine safety production.Therefore,the condition monitoring and fault diagnosis of the main ventilator in the coal mine are extremely important.As mankind enters the era of "big data",it has become an inevitable trend to use "big data" technology to perform state detection and fault diagnosis on mechanical equipment.At present,the key to the fault detection and diagnosis of mechanical equipment lies in the selection of appropriate fault diagnosis algorithms / fault detection equipment.At the same time,how to make full use of the monitored massive data,that is,to find and eliminate abnormalities in the operation of the equipment in time from the monitoring data,has also become a key research direction in the field of mechanical failure.While the traditional fault diagnosis method mainly relies on manual extraction of mechanical fault features,the success rate of fault diagnosis mainly depends on the technical level of professionals.Obviously,it has been unable to adapt to the current "big data" analysis requirements.Therefore,in this paper,based on the actual production of the coal mine,based on the vibration data,combined with the neural network and deep learning related theoretical technology,the fault diagnosis method of the main fan of the coal mine is studied.This paper studies the development status of fault diagnosis technology for coal mine main ventilator,studies vibration signal processing technology,analyzes the main fault types and vibration mechanism of coal mine main ventilator bearing,and on this basis,the vibration of coal mine main ventilator bearing The data was collected and processed.Researched neural network and deep learning related theories,focusing on two major technologies in the field of deep learning-CNN convolutional neural network and RNN recurrent neural network represented by LSTM long and short-term memory network,and constructed CNN and LSTM networks respectively,Calculation charts,mathematical calculation expressions,etc.were analyzed.Finally,using CNN's feature extraction capabilities and LSTM's time series data analysis capabilities,the two are combined to build a diagnostic model to realize the fault diagnosis of the coal mine main fan bearing,and combined with the actual vibration data,the proposed fault diagnosis model is verified Feasibility in the fault diagnosis of coal mine main fan bearing.The results show that the fault diagnosis method proposed in this study can effectively realize the fault diagnosis of the main ventilator bearing of the coal mine,and the diagnosis result has good accuracy and can be applied in practical systems.Finally,combined with actual vibration data,the proposed fault diagnosis model is verified.The results show that the fault diagnosis method researched in this paper can effectively realize the fault diagnosis of the rolling bearing of the main fan of the coal mine,and the diagnosis result has good accuracy,and the method can be applied in practical engineering.