Research On Pattern Recognition Of Vertical Hoist Steelwork Guide Typical Faults

There are various kinds of typical faults of the vertical hoist steelwork guide as radius off-set joint, abnormal protrude and inclination with the work time increasing. The status of the guide has a very important influence on the stability and security of the mine hoisting system. Therefore, it’s significant for hoisting safety to optimize the testing system, choose reasonable methods to process the signals and extract fault features, then research on the pattern recognition of the typical faults of steelwork guide. The dissertation was based on the research of dynamics mode of the hoisting rope-hoist container and steelwork guide-hoist container as well as the experiments of the steelwork guide typical faults. The fault features were extracted by the method of wavelet transform modulus maxima and EMD. Finally the patterns of the typical faults of steelwork guide were recognized on SVM.Specific contents are:The vertical dynamics mode of hoist container-hoisting rope has been built and the influence on acceleration has been researched. The multi degree dynamic model between hoist container and steelwork guide was established and simulated to analyze the output property in the excitation.The experimental platform was built up on virtual instrument. The normal condition, radius off-set joint, abnormal protrude and inclination were settled to collect the signal of vibration acceleration and inclination.The vibration signals wer denoised by the method of adaptive redundant second generation wavelet. The fault features were extracted by the method of wavelet transform modulus maxima for vibration acceleration signals and EMD for inclination signals.Using SVM to recognize the four patterns of vertical hoist steelwork guide as normal condition radius off-set joint, abnormal protrude and inclination. The recognition rate was analyzed to certify the availability of SVM to recognize the patterns of hoist steelwork guide.