Online Incipient Fault Detection Method Based on Improved ℓ₁ Trend Filtering and Support Vector Data Description

Abstract

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

A Method for Constructing Automatic Rolling Bearing Fault Identification Model Based on Refined Composite Multi-Scale Dispersion Entropy

“…The data length processed by RCMDE should be greater than 2000. In order to meet the application requirements of engineering data for different equipment and working conditions, τmax should not be too large [39]. Finally, in order to improve normal or fault bearing state characterization of RCMDE values, parameters in this paper are selected as: m= 3, c= 6, d= 1, and τmax= 15.…”

“…The second group of experimental bearings has four acquisition channels. The incipient fault detection method proposed by Wang et al suggests that the fault occurred at 532 nd data file [39]. The first channel is selected to collect the outer ring fault data after a fault occurs and to gather the normal data when no fault occurs.…”

“…It was contributed to a fault in the inner ring. The method in [39] is used to determine the data file of fault. Normal data without failure and the inner ring data following the failure are screened as the test set data.…”

Health State Identification Method of Nuclear Power Main Circulating Pump Based on EEMD and OQGA-SVM

“…A multi-sensor data fusion method based on adaptive weighting strategy using analytic hierarchy process algorithm and cross-correlation function fusion algorithm is proposed in literature [12]. In literatures [13][14][15][16], aiming at the problems of data anomalies and high-dimensional data, corresponding state recognition methods are presented. In literature [13], a method is proposed to locate abnormal vibration data based on correlation coefficient.…”

“…Literature [14] develops a new deep CNN model called a multireceptive field denoising residual convolutional network (MF-DRCN). In literature [15], a fault detection method based on support vector data description (SVDD) is proposed. In literature [16], the influence of high-amplitude blade pass frequency (BPF) vibration on rotor fault detection is analyzed.…”