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 $\lambda _{1}$ 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 $\lambda _{1}$ 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.
Highlights
As one of the most commonly employed parts in rotating machinery, rolling bearings operate under harsh working environment and are often prone to failures
The results indicate that the hypersphere trained by the kernel function bandwidth and calculated by the proposed method is of moderate size, i.e., there are no over-fitting and under-fitting problems
The following conclusions are made: 1) The incipient fault detection (IFD) model, which is built based on the IL1TF and the support vector data description (SVDD) technology, can be regarded as a "black box"
Summary
As one of the most commonly employed parts in rotating machinery, rolling bearings operate under harsh working environment and are often prone to failures. Lu et al [15] proposed a deepstructured framework to detect the incipient fault, which was based on deep neural network (DNN) and long shortterm memory (LSTM) These methods can achieve IFD by constructing bearing performance degradation models. With the purpose of solving the problems such as incipient faults detection difficulties in engineering practice, and overcoming the deficiencies of the above provided investigations, an online bearing IFD method based on improved 1 trend filter (IL1TF) and SVDD is proposed. 3) The online IFD method based on IL1TF and the SVDD proposed in this paper can quickly and accurately identify the incipient fault samples It can solve the problems of false alarm interference and significant data fluctuations following the alarm. True degradation trend of the bearing well is obtained
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have