Online Incipient Fault Detection Method Based on Improved ?₁ Trend Filtering and Support Vector Data Description

Poor model generalization, missing or false alarms, and heavy dependence on expert's experience are some of the major problems which exist in traditional incipient fault detection (IFD) methods. An IFD rolling bearing application method based combination improved ? 1 trend filtering (L1TF) support vector data description (SVDD) is proposed. First, spectral distance index multi-scale dispersion entropy normal vibration data, sensitive to faults, extracted. The filter (IL1TF) employed for processing feature values obtaining a factor with less fluctuation better indication ability. Then, after determining kernel function bandwidth SVDD by analyzing characteristics training suitable offline trained. Finally, faults identified estimating between real-time center hypersphere model. This employs full performance detect abnormal files, while reducing influence files via IL1TF. Furthermore, increases discrimination data. By utilizing Intelligent Maintenance Systems University Cincinnati laboratory Chinese petrochemical company's centrifugal pump engineering effectiveness constructed demonstrated. In addition, proposed compared against existing representative results indicate that this paper can solve alarms failure more accurately without depending external experience. great significance providing guidelines enterprises employ predictive maintenance techniques.