A systematic review of machine learning algorithms for prognostics and health management of rolling element bearings: fundamentals, concepts and applications

Singh, Jaskaran; Azamfar, Moslem; Li, Fēi; Lee, Jay

doi:10.1088/1361-6501/ab8df9

Cited by 62 publications

(29 citation statements)

References 386 publications

(152 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Mohamed et al [20] they put a twolevel converter control scheme based on predictive model control and feedforward neural network, which can reduce the THD and improve system stability. Singh et al [21] took rolling bearings as the research object and summarized the machine learning algorithms for failure prediction and health management. Guo et al [22] discussed recent research progress and applications of data-driven, physical-based and hybrid prediction methods in predictive model approaches for engineering systems.…”

Section: Introductionmentioning

confidence: 99%

Harmonic Characteristics Data-Driven THD Prediction Method for LEDs Using MEA-GRNN and Improved-AdaBoost Algorithm

2021

View full text Add to dashboard Cite

Light-emitting Diode (LED) lamps have been widely used due to versatility and energy efficiency. However, LEDs are nonlinear loads, the massive usage will inject harmonics into the lighting system, which has influenced the power quality. Total Harmonic Distortion (THD) is an important parameter to evaluate the power quality, but the prediction of THD for LEDs is a challenging task. This paper addresses this issue by designing harmonic characteristics detection experiment and using artificial intelligence algorithm. Firstly, LED lamps with different driving circuits were tested, the relevant data of each harmonic were sampled and analyzed. Then, a THD prediction method based on an improved AdaBoost algorithm is proposed. In this method, a Generalized Regression Neural Network (GRNN) model is established, and its parameters are optimized by Mind Evolution Algorithm (MEA) to improve the search ability of GRNN. On this basis, the AdaBoost algorithm is utilized to integrate multiple MEA-GRNN individuals to form a strong predictor, which improves the generalization ability of the model. To avoid the integration failure caused by improper selection of threshold value, a sigmoid adaptive factor is added to improve the accuracy of AdaBoost algorithm. Finally, the Ada-MEA-GRNN model is trained and simulated with the LED harmonic data collected by the experiment. The simulation results show that the prediction accuracy of the proposed method is better than BP and GRNN, which can reach 95.48%. Meanwhile, even if the input dimension is reduced, the error is still small. INDEX TERMS LED lamps, THD prediction, ensemble learning, mind evolution algorithm (MEA), generalized regression neural network (GRNN).

show abstract

Section: Introductionmentioning

confidence: 99%

Harmonic Characteristics Data-Driven THD Prediction Method for LEDs Using MEA-GRNN and Improved-AdaBoost Algorithm

2021

View full text Add to dashboard Cite

show abstract

“…Traditional machine learning methods include artificial neural network (ANN), support vector machine (SVM), hidden Markov method, etc. 4–7 With shallow model structure, it is difficult for these traditional machine learning methods to extract deep fault features processing long time series, nonlinear, massive vibration signal data. As an important branch of machine learning, deep learning has potential application for deep feature ex-traction of large sample and multi-dimensional vibration signals.…”

Section: Introductionmentioning

confidence: 99%

Fault diagnosis of rolling bearing based on multimodal data fusion and deep belief network

Wang

Che

2021

Proceedings of the Institution of Mechanical Engineers, Part C:

View full text Add to dashboard Cite

Aiming at raw vibration signal of rolling bearing with long time series, a fault diagnosis model based on multimodal data fusion and deep belief network is proposed in this paper. First, multimodal data composed of artificial features and model features can be obtained by time-frequency domain analysis and unsupervised learning based on restricted Boltzmann machine (RBM). Second, canonical correlation analysis method is used to extract the typical feature pairs from the multimodal data to realize the feature-level multimodal data fusion. Third, deep belief network is applied to extract deep feature mapping between typical feature pairs and fault types. After greedy layer-wise pre-training and fine-tuning, it is available to achieve the trained model for fault diagnosis of rolling bearing. Typical rolling bearing datasets are used to testify the effectiveness of the proposed method. It is verified that the robustness and accuracy of the proposed method are superior to common methods.

show abstract

“…Keeping in mind that among the typical monitoring techniques, vibration monitoring can detect defect at an earlier stage compared to lubrication and temperature monitoring, this would still be used as the main indicator for a defect being present. However, the integration between the SCADA and condition monitoring systems data sources is often non-existent [23]. Therefore, a lot of hands-on work is needed in each individual case to prepare the data which greatly hampers a machine learning solution to be widely introduced in wind turbine monitoring.…”

Section: Introductionmentioning

confidence: 99%

Increasing Wind Turbine Drivetrain Bearing Vibration Monitoring Detectability Using an Artificial Neural Network Implementation

2021

View full text Add to dashboard Cite

The highest costs due to premature failures in wind turbine drivetrains are related to defects in the gearbox, with bearing failures being overrepresented. Vibration monitoring has been identified as the primary tool to detect and diagnose these types of failures. However, late or no signs of the failures are still being reported. Artificial neural networks (ANNs) has been shown to favourably be used as a classifier of bearing failures to increase the detection and diagnosis performance, which requires labelled data when training for all types of considered failures. However, less work has been done with an ANN used to create descriptive functions of the vibration and turbine operation data relationship and thereby negating inherent variance in the vibration data and increasing the detectability when a defect appears. Therefore, this study utilizes the relationship between the rotational speed recorded during a vibration measurement and the calculated condition indicator values of specific bearing failures in three wind turbine gearbox failures. An ANN establishes a function between the rotational speed and condition indicator values with healthy training data collected before the failure occurred. Thereafter, whole datasets leading up to the changing of the gearboxes is used to predict the condition indicator values without the failure influence. The difference between the predicted and true values show an increased sensitivity of the detection in two cases of gearbox output shaft bearing failures as well as indicating a planet bearing failure which with the previous data had gone undetected.

show abstract

A systematic review of machine learning algorithms for prognostics and health management of rolling element bearings: fundamentals, concepts and applications

Cited by 62 publications

References 386 publications

Harmonic Characteristics Data-Driven THD Prediction Method for LEDs Using MEA-GRNN and Improved-AdaBoost Algorithm

Harmonic Characteristics Data-Driven THD Prediction Method for LEDs Using MEA-GRNN and Improved-AdaBoost Algorithm

Fault diagnosis of rolling bearing based on multimodal data fusion and deep belief network

Increasing Wind Turbine Drivetrain Bearing Vibration Monitoring Detectability Using an Artificial Neural Network Implementation

Contact Info

Product

Resources

About