Data-Driven Mine Hoist Gearbox Fault Diagnosis Methods And Applications

Mine hoist gearbox fault detection technology is of great significance to ensure the reliable operation of mine hoisting system equipment.It is difficult to obtain fault data samples of mine hoist gearboxes,and how to train the health status data of gearboxes(gears and bearings)to detect faults is the key.In this thesis,we use existing priori knowledge to design highly robust fault diagnosis methods to achieve accurate fault diagnosis under different types of historical data(such as health history data and fault history data)and under different operating conditions.The main research of the thesis is as follows.(1)For the problem of detecting the operating status of mine hoist gearboxes without any historical data,this thesis conducts a differential model analysis for the real-time operating status data of gearboxes,and proposes an automatic fault detection method for local gearboxes under different operating conditions based on the statistical similarity of the interaction feature data between gearbox gear teeth,so as to use the gearbox gear operating status feature data The difference degree of the probability density function is realized to detect local gearbox faults.(2)For the problem of detecting the operating state of mine hoist gearboxes where health state history data exists,this thesis extends the difference analysis detection technique to gearbox variable condition diagnosis by analyzing the applicability of time-frequency analysis technique and discrete model in the difference analysis,and then uses narrowband feature extraction method to extract new type of information from the vibration signal based on the historical data of attenuation signal in the vibration signal to enhance the signal in the fault-related information to achieve accurate gearbox fault diagnosis.In this thesis,two solutions are proposed and a prototype system is constructed to further verify its feasibility through two conditions in mine hoist gearbox fault detection.The test results show that the proposed method can enhance the robustness of hoist gearbox fault detection under different working conditions and improve the fault detection accuracy.