Modeling of the safe region based on support vector data description for health assessment of wheelset bearings

Liu, Zhiliang; Kang, Jinlong; Zhao, Xuejun; Zuo, Ming J.; Qin, Yong; Jia, Limin

doi:10.1016/j.apm.2019.03.040

Cited by 16 publications

(11 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, it is necessary to analyze how influences the result. Just like the discussion in previous literature 43 , the value of kernel function we adopted in (8) shows a minor difference between two points when they are far apart in original feather space. In this case, a small value is inapplicable to anomaly prediction as it would further worsen the situation.…”

Section: The Influence Of the Key Parametersmentioning

confidence: 63%

Anomaly Prediction: Successive Analysis for Healthy State Based on the Expansion to Remaining Useful Life

Yin

Liu

2021

Preprint

Self Cite

View full text Add to dashboard Cite

The prediction of remaining useful life is the key to realize the condition-based maintenance. However, Existing studies on remaining useful life cannot effectively cope with the scenarios with extremely high requirement to stability. In this paper, the conception of anomaly prediction is introduced to solve the problem by expanding the remaining useful life to healthy stage to describe and predict the equipment condition at this stage. To obtain more reasonable health indicator, inverse number is firstly introduced into the feature selection. On this basis, a novel method is put forward, which consists of center optimization, boundary adjustment and regressive analysis, in order to realize the anomaly prediction and fills the gap of the absence to remaining life research at healthy stage. Two validations are subsequently implemented with the open dataset of XJTU-SY. The experimental results and comparison analysis demonstrate the feasibility and the superiority of proposed method, respectively. The dissection to the remaining useful life based anomaly prediction indicates that the related research is essential and of great significant to the condition-based maintenance.

show abstract

Section: The Influence Of the Key Parametersmentioning

confidence: 63%

Anomaly Prediction: Successive Analysis for Healthy State Based on the Expansion to Remaining Useful Life

Yin

Liu

2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…However, determination of ε value has emerged as a new problem. Liu et al [30] proposed the variance-square root mean-ratio (VSRMR) method based on Evangelista et al The formula of this method is shown in Eq. (26).…”

Section: Determination Of Svdd Model Parametersmentioning

confidence: 99%

“…Wang et al [29] proposed a method for optimizing σ using the boundary tightness. Liu et al [30] suggested a method for locating the maximum value of the variance-square root mean-ratio (VSRMR) for selecting σ. These methods have made significant contributions to further application of the SVDD.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

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

et al. 2021

View full text Add to dashboard Cite

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

show abstract

“…Performance evaluation has been proven to be an effective technique in tracking and identifying the operational situation of the mechanical equipment in time [3]. So far, several artificial intelligent methods, such as logistic regression [4] support vector data description [5], [6], self-organizing mapping [7], neural network [8], Gaussian mixture model [9], The associate editor coordinating the review of this manuscript and approving it for publication was Yu Liu . have been widely employed in performance evaluation for mechanical equipment.…”

Section: Introductionmentioning

confidence: 99%

A Fuzzy Performance Evaluation Model for a Gearbox System Using Hidden Markov Model

Huang

et al. 2020

IEEE Access

View full text Add to dashboard Cite

In order to track and grasp the operation situations of the gearboxes, the vertical vibration signals of three different gear fault states, normal, worn and broken teeth, are collected via a gearbox vibration experiment. An online diagnosis and performance evaluation model with hidden Markov model (HMM) and fuzzy comprehensive evaluation is proposed. To address the limitation of maximum membership principle in the case of equal membership or the membership is very close to each other, a closeness evaluation strategy is proposed by defining the likelihood ratio of HMM as a similarity and selecting an combined membership function of the semi-trapezoidal and intermediate-ridge distribution. Results show that the online diagnosis has achieved a good performance with the similarity strategy. Compared with the evaluation strategy of the maximum membership principle, the proposed gearbox performance model with the closeness evaluation strategy is more accurately distinguished from the evaluation results of the broken teeth state and the worn state, especially for the case of the equal membership.

show abstract

Modeling of the safe region based on support vector data description for health assessment of wheelset bearings

Cited by 16 publications

References 30 publications

Anomaly Prediction: Successive Analysis for Healthy State Based on the Expansion to Remaining Useful Life

Anomaly Prediction: Successive Analysis for Healthy State Based on the Expansion to Remaining Useful Life

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

A Fuzzy Performance Evaluation Model for a Gearbox System Using Hidden Markov Model

Contact Info

Product

Resources

About

Modeling of the safe region based on support vector data description for health assessment of wheelset bearings

Cited by 16 publications

References 30 publications

Anomaly Prediction: Successive Analysis for Healthy State Based on the Expansion to Remaining Useful Life

Anomaly Prediction: Successive Analysis for Healthy State Based on the Expansion to Remaining Useful Life

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

A Fuzzy Performance Evaluation Model for a Gearbox System Using Hidden Markov Model

Contact Info

Product

Resources

About

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