A recurrent neural network approach for remaining useful life prediction utilizing a novel trend features construction method

Zhao, Sen; Zhang, Yong; Wang, Shang; Zhou, Beitong; Cheng, Cheng

doi:10.1016/j.measurement.2019.06.004

Cited by 114 publications

(32 citation statements)

References 34 publications

(39 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A framework for estimating the RUL of mechanical systems is proposed, which is composed of the multi-layer perceptron and multilayer perceptron and evolutionary algorithm for optimizing parameters [27]. Besides, there are many other machine learning algorithms, such as neural networks [28]- [30], capsule neural networks [31], dynamic Bayesian networks [32] and so on.…”

Section: New Faultmentioning

confidence: 99%

An Improved PF Remaining Useful Life Prediction Method Based on Quantum Genetics and LSTM

Yang

Sun

2019

IEEE Access

View full text Add to dashboard Cite

Remaining useful life (RUL) is the premise and basis of the equipment health management plan. As accurate as possible life prediction is of great significance to reliability and economy of equipment maintenance. In this paper, a data-driven improved particle filter (PF) RUL prediction method is proposed. A health indicator extraction method based on multi-feature fusion is introduced for the RUL prediction, which can visually show the degradation trend of the healthy state of the equipment. The degradation model and observation model of equipment health indicators are established, and the PF algorithm is used to track parameters of the model. A quantum genetic algorithm is employed to improve the problem of particle degradation in PF. On the basis of filter tracking, long short term memory (LSTM) network is used to predict the trend of model coefficients, which further improves the accuracy of RUL prediction. The experiment using the C-MAPSS data set shows the proposed method has a better prediction accuracy than other methods. INDEX TERMS Remaining useful life, particle filter, quantum genetic algorithm, long short term memory.

show abstract

Section: New Faultmentioning

confidence: 99%

An Improved PF Remaining Useful Life Prediction Method Based on Quantum Genetics and LSTM

Yang

Sun

2019

IEEE Access

View full text Add to dashboard Cite

show abstract

“…However, these approaches show poor distinguishability for insufficient fault features for nonlinear and non-stationary signals. Another powerful signal processing method for non-linear and non-stationary signals, named empirical mode decomposition (EMD), ensemble empirical mode decomposition (EEMD) [18], and complete ensemble empirical mode decomposition (CEEMD) [19], has been widely used to solve fault diagnosis of rotating machinery and circuit systems. Additionally, compared with wavelet transform where the basic functions are fixed, the EMD-based method decomposes signals according to time-scale characteristics of data without setting any basis function in advance, which has stronger local stationary.…”

Section: Introductionmentioning

confidence: 99%

Open-Circuit Fault Diagnosis of Three-Phase PWM Rectifier Using Beetle Antennae Search Algorithm Optimized Deep Belief Network

Zhang

2020

Electronics

View full text Add to dashboard Cite

Effective open-circuit fault diagnosis for a two-level three-phase pulse-width modulating (PWM) rectifier can reduce the failure rate and prevent unscheduled shutdown. Nevertheless, traditional signal-based feature extraction methods show poor distinguishability for insufficient fault features. Shallow learning diagnosis models are prone to fall into local extremum, slow convergence speed, and overfitting. In this paper, a novel fault diagnosis strategy based on modified ensemble empirical mode decomposition (MEEMD) and the beetle antennae search (BAS) algorithm optimized deep belief network (DBN) is proposed to cope with these problems. Initially, MEEMD is applied to extract useful fault features from each intrinsic mode function (IMF) component. Meanwhile, to remove features with redundancy and interference, fault features are selected by calculating the importance of each feature based on the extremely randomized trees (ERT) algorithm, and the dimension of fault feature vectors is reduced by principal component analysis. Additionally, the DBN stacked with two layers of a restricted Boltzmann machine (RBM) is selected as the classifier, and the BAS algorithm is used as the optimizer to determine the optimal number of units in the hidden layers of the DBN. The proposed method combined with feature extraction, feature selection, optimization, and fault classification algorithms significantly improves the diagnosis accuracy.

show abstract

“…It was then used to construct a one-dimensional HI to reflect the degradation process of the device. Because learning features from a fixed window size may result in changes in local features and therefore may affect the prediction results, Zhao et al [8] proposed a RUL prediction method based on the trend feature of the total time series representing degradation. An empirical mode decomposition (EMD) method was used to decompose and reconstruct the signals to obtain trend features.…”

Section: Introductionmentioning

confidence: 99%

“…Then the model with the best predicted result is selected as the target model. The references [8]- [10] applied the grid search method to select the optimal structure of the model. The random search method is to randomly select several sets of parameters within the range for model development.…”

Section: Introductionmentioning

confidence: 99%

A Bayesian Optimization AdaBN-DCNN Method With Self-Optimized Structure and Hyperparameters for Domain Adaptation Remaining Useful Life Prediction

2020

IEEE Access

View full text Add to dashboard Cite

The prediction of remaining useful life (RUL) of mechanical equipment provides a timely understanding of the equipment degradation and is critical for predictive maintenance of the equipment. In recent years, the applications of deep learning (DL) methods to predict equipment RUL have attracted much attention. There are two major challenges when applying the DL methods for RUL prediction: (1) It is difficult to select the prediction model structure and hyperparameters such as network depth, learning rate, batch size, and etc. (2) The developed prediction model is domain dependent, i.e., it can only give good prediction performance in one data domain (one particular type of working conditions and fault modes). In order to meet the challenges, a novel RUL prediction method developed using a deep convolutional neural network (DCNN) combined with Bayesian optimization and adaptive batch normalization (AdaBN) is presented in this paper. The proposed RUL prediction model is validated by the turbofan engine degradation simulation dataset provided by NASA. The prediction results show that the proposed prediction model provides better prediction results than model structures obtained by random search and grid search. The results also show that the domain adaptation capability of the prediction model has been improved. INDEX TERMS Remaining useful life prediction, Bayesian optimization, adaptive batch normalization, domain adaptation.

show abstract

A recurrent neural network approach for remaining useful life prediction utilizing a novel trend features construction method

Cited by 114 publications

References 34 publications

An Improved PF Remaining Useful Life Prediction Method Based on Quantum Genetics and LSTM

An Improved PF Remaining Useful Life Prediction Method Based on Quantum Genetics and LSTM

Open-Circuit Fault Diagnosis of Three-Phase PWM Rectifier Using Beetle Antennae Search Algorithm Optimized Deep Belief Network

A Bayesian Optimization AdaBN-DCNN Method With Self-Optimized Structure and Hyperparameters for Domain Adaptation Remaining Useful Life Prediction

Contact Info

Product

Resources

About