Research On Fault Diagnosis Of Main Shaft Device Of Mine Hoist

Main shaft device, a core part of one mine hoist which is known as the “throat” of the mining production system, carries almost all of the static-dynamic loading casued by the mine hoist which is lifting minerals, equipments or workers. As the key rotation mechanism of the mine hoist, the main shaft device will appear some mechanical faults after continuous operating under high load operation for a long time. As long as these faults occur, the main shaft device will also product unnormal mechanical vibration that will influence the mine hoist works smoothly and make the whole system’s performance degradation. What’s more, the undesired mechanical vibration will also bring some huge potential security security problems. For these traditional regularly maintenance and breakdown maintenance, they will not only waste a lot of manpower and financial resoures but also increase the risk of overterm service of these parts which have been seriously degraded. Therefore, we need to monitor the states of the mine hoist in real time and judge whether this system has fault and where the fault occurs by analyzing these monitoring data. Then we can ensure that the mine hoist works reliably and safely based on dynamic grasp of the mine hoist’s performance.Supported by National High-tech Research and Development Projects “Intelligent Diagnosis of Malignant Accident of Mine Hoist”（No. 2009AA04Z415）, National Natural Science Foundation of China “Healthy Monitoring of Super Deep Vertical Shaft Hoist System based on WSN”（No. 51275513） and the Scientific Research and Innovation Project for College and University Graduates in Jiangsu Province of China（No. CXZZ13₀926）, this thesis designs a vibration signal analysis based fault diagnosis method for the research object, i.e., the main shaft device, to provide theoretical support and technical solution for the safe operation of mine lifting system. This novel method is based on signal processing technology, feature extraction technology, statistical theory and fault identification technology, and which includes three parts: signal denoising analysis, feature extraction and fault identification.Firstly, design an approach about how to cancel the noise in the vibration signals collected from the main shaft device of mine hoist. As we know, mine hoist operats in a harsh environment where has storng noise interference and which will overwhelm the useful information about the fault when obtaining vibration signals with sersors. Address this issue, a denoise method based on ISVD and EMD is presented to effectively remove the noise from the collected vibration signals. In this approach, ISVD is short for the improvement singular value decomposition denoise method.Secondly, research how to extract features from vibration signals of main shaft device. Feature extraction is an important step of mechanical fault diagnosis. A random statistical average algorithm is designed by statistical analysis technology, and which unites with multidimensional data dimension reduction technique to form a new feature extraction method based on PCA and random statistical average analysis. At the same time, another feature extraction method based on improvement EEMD and random statistical average analysis is also presented. And these two feature extraction methods can obtain useful information to expose chatacteristics of mechanical faults.Finally, to carry out fault diagnosis method based on space vector statistical analysis research. The obtained feature samples are some special vectors in fact. After introducing spatial geometry theory to explore the relationship between various feature samples of different conditions of main shaft device and considering statistical analysis technology, a fault identification method based on zero space observer and a fault identification method based on self-zero space prejection analysis are presented to implement diagnose occurring know faults and warn the unknow faults without complex model and large number of computations.