A Novel Health Indicator Based on Cointegration for Rolling Bearings’ Run-To-Failure Process

Li, Hong‐Ru; Li, Yaolong; Yu, He

doi:10.3390/s19092151

Cited by 11 publications

(4 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is important to highlight that this research adopts a detection and prediction level of 70% for Scenario 5. The main justification of this selection is based on recommendations from the literature [31] and experts in the field.…”

Section: Intelligent Maintenancementioning

confidence: 99%

Lifetime Benefit Analysis of Intelligent Maintenance: Simulation Modeling Approach and Industrial Case Study

Nordal

El‐Thalji

2021

Applied Sciences

View full text Add to dashboard Cite

The introduction of Industry 4.0 is expected to revolutionize current maintenance practices by reaching new levels of predictive (detection, diagnosis, and prognosis processes) and prescriptive maintenance analytics. In general, the new maintenance paradigms (predictive and prescriptive) are often difficult to justify because of their multiple inherent trade-offs and hidden systems causalities. The prediction models, in the literature, can be considered as a “black box” that is missing the links between input data, analysis, and final predictions, which makes the industrial adaptability to such models almost impossible. It is also missing enable modeling deterioration based on loading, or considering technical specifications related to detection, diagnosis, and prognosis, which are all decisive for intelligent maintenance purposes. The purpose and scientific contribution of this paper is to present a novel simulation model that enables estimating the lifetime benefits of an industrial asset when an intelligent maintenance management system is utilized as mixed maintenance strategies and the predictive maintenance (PdM) is leveraged into opportunistic intervals. The multi-method simulation modeling approach combining agent-based modeling with system dynamics is applied with a purposefully selected case study to conceptualize and validate the simulation model. Three maintenance strategies (preventive, corrective, and intelligent) and five different scenarios (case study data, manipulated case study data, offshore and onshore reliability data handbook (OREDA) database, physics-based data, and hybrid) are modeled and simulated for a time period of 20 years (175,200 h). Intelligent maintenance is defined as PdM leveraged in opportunistic maintenance intervals. The results clearly demonstrate the possible lifetime benefits of implementing an intelligent maintenance system into the case study as it enhanced the operational availability by 0.268% and reduced corrective maintenance workload by 459 h or 11%. The multi-method simulation model leverages and shows the effect of the physics-based data (deterioration curves), loading profiles, and detection and prediction levels. It is concluded that implementing intelligent maintenance without an effective predictive horizon of the associated PdM and effective frequency of opportunistic maintenance intervals, does not guarantee the gain of its lifetime benefits. Moreover, the case study maintenance data shall be collected in a complete (no missing data) and more accurate manner (use hours instead of date only) and used to continuously upgrade the failure rates and maintenance times.

show abstract

Section: Intelligent Maintenancementioning

confidence: 99%

Lifetime Benefit Analysis of Intelligent Maintenance: Simulation Modeling Approach and Industrial Case Study

Nordal

El‐Thalji

2021

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…The DCAE model can be trained using noisy data for learning. Such configuration is used for fault diagnosis of ball-bearing failure in rotational machines [12], railways, and diesel engine operations [13].…”

Section: Review Of the Contributions In This Special Issuementioning

confidence: 99%

Special Issue “Advanced Signal Processing in Intelligent Systems for Health Monitoring”

Abbod

Shieh

2019

Sensors

View full text Add to dashboard Cite

Recently, significant developments have been achieved in the field of artificial intelligence, in particular the introduction of deep learning technology that has improved the learning and prediction accuracy to unpresented levels, especially when dealing with big data and high-resolution images. Significant developments have occurred in the area of medical signal processing, measurement techniques, and health monitoring, such as vital biological signs for biomedical systems and noise and vibration of mechanical systems, which are carried out by instruments that generate large data sets. These big data sets, ultimately driven by high population growth, would require Artificial Intelligence techniques to analyse and model. In this Special Issue, papers are presented on the latest signal processing and deep learning techniques used for health monitoring of biomedical and mechanical systems.

show abstract

“…Xia et al [12] applied MD-CUSUM to reduce the dimension of features to obtain the monotonical HI. Li et al [13] obtained a novel HI by the co-integration method. There are also some recent methods of RUL estimation and degradation trend prognostics that have achieved good results though raw data.…”

Section: Introductionmentioning

confidence: 99%

Remaining useful life prediction of bearings based on multiple-feature fusion health indicator and weighted temporal convolution network

Wang¹,

Zhang²,

Guo³

et al. 2022

Meas. Sci. Technol.

View full text Add to dashboard Cite

In the Prognostic and Health Management, predicting the remaining useful life (RUL) of bearings is of great significance. The traditional RUL prediction methods have two disadvantages: 1) Health indicator (HI) is mainly constructed through expert experience and signal processing technology, which is lack of monotonicity and generalization; 2) In the RUL prediction, the time correlation of information is emphasized, but the influence of the variation amplitude and severity of vibration signal is ignored. Therefore, a method of bearings RUL prediction based on multiple features fusion HI (MFF-HI) and weighted temporal convolution network (WTCN) is proposed in this paper. MFF-HI is constructed by multiple features fusion depth network (MFFDN) with MISH activation function extracting and fusing the degradation information of bearing time-domain features; WTCN is established based on TCN and a new loss function Time-Mean Square Error (T-MSE) to reduce the error of network pickup in the bearing degradation stage. The NASA IMS dataset, IEEE PHM 2012 dataset and XJTU-SY dataset are used to verify the superiority of the method. The results show that this method can validly predict the RUL of bearings and has higher prediction accuracy.

show abstract

A Novel Health Indicator Based on Cointegration for Rolling Bearings’ Run-To-Failure Process

Cited by 11 publications

References 35 publications

Lifetime Benefit Analysis of Intelligent Maintenance: Simulation Modeling Approach and Industrial Case Study

Lifetime Benefit Analysis of Intelligent Maintenance: Simulation Modeling Approach and Industrial Case Study

Special Issue “Advanced Signal Processing in Intelligent Systems for Health Monitoring”

Remaining useful life prediction of bearings based on multiple-feature fusion health indicator and weighted temporal convolution network

Contact Info

Product

Resources

About