Jiang, X., Namokel, M., Hu, C., Tian, R. (2021). Research on Elevator Fault Information Extraction and Prediction Diagnosis. In: Wang, Y., Martinsen, K., Yu, T., Wang, K. (eds) Advanced Manufacturing and Automation X. IWAMA 2020. Lecture Notes in Electrical Engineering, vol 737. Springer, Singapore.
Abstract
Nowadays, most of the fault diagnosis methods are based on the collected data, which can not realize the timely prediction of fault diagnosis. By extracting the vibration information that can reflect the different operation states of elevator, the signal is preprocessed by the multi-threshold denoising method of wavelet packet, and the predicted data is extracted by feature information. The vibration acceleration signals of elevator in X, Y and Z directions are measured by the elevator vibration tester EVA-625, and preprocessed by the method above. Then, the maximal peak value, peak-peak value and A95 value of the signals are calculated to judge whether the elevator has faults or not and the comfort of elevator ride according to the relevant national standards, combined with the historical vibration state of elevator operation, the analysis and comparison are carried out to realize prediction of mechanical faults and the prediction diagnosis.