Deep Convolutional Neural Network Based Regression Approach for Estimation of Remaining Useful Life

Babu, G. S. D.; Zhao, Peilin; Liu, Yuanyuan

doi:10.1007/978-3-319-32025-0_14

Cited by 558 publications

(167 citation statements)

References 17 publications

Supporting

Mentioning

124

Contrasting

Order By: Relevance

“…Indeed, NASA’s C-MAPSS turbofans time-to-failure data set have been extensively analyzed 15 with RUL estimation as the main focus. Sateesh Babu et al 31 proposed the first CNN-based model for RUL prediction where convolution and pooling filters were applied along the temporal dimension of data. Also, Li et al 32 proposed a DCNN where a time-window approach is employed as preprocessing of the turbofans C-MAPSS data set and convolution filters were applied along the temporal dimension.…”

Section: Introductionmentioning

confidence: 99%

“…In fact, a common practice for RUL estimation with CNNs is to apply the convolution and pooling filters along the temporal dimension of sensoring input data. 31,32 This might lead to loss of information and subsequent poor fault diagnosis and prognosis performance as some important features could be lost, such as periodicity and low-valued amplitudes in signals (e.g. vibration and acoustic emission monitoring).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A novel deep capsule neural network for remaining useful life estimation

Palazuelos

Droguett

Pascual

2019

Proceedings of the Institution of Mechanical Engineers, Part O:

View full text Add to dashboard Cite

With the availability of cheaper multi-sensor systems, one has access to massive and multi-dimensional sensor data for fault diagnostics and prognostics. However, from a time, engineering and computational perspective, it is often cost prohibitive to manually extract useful features and to label all the data. To address these challenges, deep learning techniques have been used in the recent years. Within these, convolutional neural networks have shown remarkable performance in fault diagnostics and prognostics. However, this model present limitations from a prognostics and health management perspective: to improve its feature extraction generalization capabilities and reduce computation time, ill-based pooling operations are employed, which require sub-sampling of the data, thus loosing potentially valuable information regarding an asset’s degradation process. Capsule neural networks have been recently proposed to address these problems with strong results in computer vision–related classification tasks. This has motivated us to extend capsule neural networks for fault prognostics and, in particular, remaining useful life estimation. The proposed model, architecture and algorithm are tested and compared to other state-of-the art deep learning models on the benchmark Commercial Modular Aero Propulsion System Simulation turbofans data set. The results indicate that the proposed capsule neural networks are a promising approach for remaining useful life prognostics from multi-dimensional sensor data.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A novel deep capsule neural network for remaining useful life estimation

Palazuelos

Droguett

Pascual

2019

Proceedings of the Institution of Mechanical Engineers, Part O:

View full text Add to dashboard Cite

show abstract

“…In order to use these features uniformly as input to the predictive model, it is necessary to normalize the feature values. In this paper, the z-score normalization processing was applied to process each feature separately, as shown in Equation 39: (39) where x n is the n-th feature vector, µ x is the average of feature vector x n , σ x is the standard deviation of feature vector x n , and y n is the normalized data. The first engine of the FD001 training set contains 192 cycles.…”

Section: Validation Of the Proposed Methods A C-mapss Dataset Desmentioning

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

“…An overview of the different applications of CNN regarding machine health monitoring is given in [11]. CNN have emerged into a broad variety of fields, such as predictive maintenance [12,13] and medical [14,15], or mechanical diagnosis [16][17][18][19][20][21][22][23]. The latter has been dominated by model-driven approaches for decades and more recently, data-driven approaches based on feature engineering.…”

Section: Introductionmentioning

confidence: 99%

Diagnosing Automotive Damper Defects Using Convolutional Neural Networks and Electronic Stability Control Sensor Signals

Zehelein

Hemmert-Pottmann

Lienkamp

2020

JSAN

View full text Add to dashboard Cite

Chassis system components such as dampers have a significant impact on vehicle stability, driving safety, and driving comfort. Therefore, monitoring and diagnosing the defects of these components is necessary. Currently, this task is based on the driver’s perception of component defects in series production vehicles, even though model-based approaches in the literature exist. As we observe an increased availability of data in modern vehicles and advances in the field of deep learning, this paper deals with the analysis of the performance of Convolutional Neural Networks (CNN) for the diagnosis of automotive damper defects. To ensure a broad applicability of the generated diagnosis system, only signals of a classic Electronic Stability Control (ESC) system, such as wheel speeds, longitudinal and lateral vehicle acceleration, and yaw rate, were used. A structured analysis of data pre-processing and CNN configuration parameters were investigated in terms of the defect detection result. The results show that simple Fast Fourier Transformation (FFT) pre-processing and configuration parameters resulting in small networks are sufficient for a high defect detection rate.

show abstract

Deep Convolutional Neural Network Based Regression Approach for Estimation of Remaining Useful Life

Cited by 558 publications

References 17 publications

A novel deep capsule neural network for remaining useful life estimation

A novel deep capsule neural network for remaining useful life estimation

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

Diagnosing Automotive Damper Defects Using Convolutional Neural Networks and Electronic Stability Control Sensor Signals

Contact Info

Product

Resources

About