Remaining useful life predictions for turbofan engine degradation using semi-supervised deep architecture

Ellefsen, André Listou; Bjørlykhaug, Emil; Æsøy, Vilmar; Ushakov, Sergey; Zhang, Houxiang

doi:10.1016/j.ress.2018.11.027

Cited by 342 publications

(94 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Wu et al [33] used the vanilla LSTM network to obtain good RUL prediction in the case of complicated operations, model degradation and strong noise. Ellefsen et al [34] established a semi-supervised model for RUL prediction to provide high RUL prediction accuracy, even with reduced amounts of labeled training data. Zhang et al [35] combined transfer learning and Bidirectional long short term memory (BLSTM) network for RUL prediction.…”

Section: Introductionmentioning

confidence: 99%

A Directed Acyclic Graph Network Combined With CNN and LSTM for Remaining Useful Life Prediction

2019

IEEE Access

171

View full text Add to dashboard Cite

Accurate and timely prediction of remaining useful life (RUL) of a machine enables the machine to have an appropriate operation and maintenance decision. Data-driven RUL prediction methods are more attractive to researchers because they can be deployed quicker and cheaper compared to other approaches. The existing deep neural network (DNN) models proposed for the applications of RUL prediction are mostly single-path and top-down propagation. In order to improve the prognostic accuracy of the network, this paper proposes a directed acyclic graph (DAG) network that combines long short term memory (LSTM) and a convolutional neural network (CNN) to predict the RUL. Different from the existing prediction models combined with CNN and LSTM, the method proposed in this paper combines CNN and LSTM organically instead of just using CNN for feature extraction. Moreover, when a single timestamp is used as an input, padding the signals in the same training batch would affect the prediction ability of the developed model. To overcome this drawback, the proposed method generates a short-term sequence by sliding the time window (TW) with one step size. In addition, based on the degradation mechanism, the piece-wise RUL function is used instead of the traditional linear function. In the experimental test, the turbofan engine degradation simulation dataset provided by NASA is used to validate the proposed RUL prediction model. By comparing with the existing methods using the same dataset, it can be concluded that the prediction method proposed in this paper has better prediction capability. INDEX TERMS Remaining useful life prediction, long-short-term memory network, convolutional neural networks, turbofan engine.

show abstract

Section: Introductionmentioning

confidence: 99%

A Directed Acyclic Graph Network Combined With CNN and LSTM for Remaining Useful Life Prediction

2019

IEEE Access

171

View full text Add to dashboard Cite

show abstract

“…The results reveal that the LSTM outperforms all other methods investigated for both RUL estimation and fault occurrence predictions. Ellefsen et al [25] first uses the restricted Boltzmann machine (RBM) to pretrain the model in an unsupervised manner to automatically generate new degradation-related features from the raw data. Subsequently, the newly generated features are used as input for LSTM to predict the RUL.…”

Section: Current Deep Learning Solutionsmentioning

confidence: 99%

Deep Learning with Dynamically Weighted Loss Function for Sensor-Based Prognostics and Health Management

Rengasamy

Jafari

Rothwell

et al. 2020

Sensors

View full text Add to dashboard Cite

Deep learning has been employed to prognostic and health management of automotive and aerospace with promising results. Literature in this area has revealed that most contributions regarding deep learning is largely focused on the model’s architecture. However, contributions regarding improvement of different aspects in deep learning, such as custom loss function for prognostic and health management are scarce. There is therefore an opportunity to improve upon the effectiveness of deep learning for the system’s prognostics and diagnostics without modifying the models’ architecture. To address this gap, the use of two different dynamically weighted loss functions, a newly proposed weighting mechanism and a focal loss function for prognostics and diagnostics task are investigated. A dynamically weighted loss function is expected to modify the learning process by augmenting the loss function with a weight value corresponding to the learning error of each data instance. The objective is to force deep learning models to focus on those instances where larger learning errors occur in order to improve their performance. The two loss functions used are evaluated using four popular deep learning architectures, namely, deep feedforward neural network, one-dimensional convolutional neural network, bidirectional gated recurrent unit and bidirectional long short-term memory on the commercial modular aero-propulsion system simulation data from NASA and air pressure system failure data for Scania trucks. Experimental results show that dynamically-weighted loss functions helps us achieve significant improvement for remaining useful life prediction and fault detection rate over non-weighted loss function predictions.

show abstract

“…Yan utilized SSL in order for the early diagnosis and detection of the air handling unit and showed high accuracy of defect diagnosis [16]. Ellefsen also applied SSL to predict the effective life for engine performance lowering of turbo fan [17]. Sen used SSL to validate data in order to recognize the authenticity of sensor data which is used to judge a defect in the pipe process [18].…”

Section: Semi-supervised Learningmentioning

confidence: 99%

Quality Prediction and Yield Improvement in Process Manufacturing Based on Data Analytics

2020

View full text Add to dashboard Cite

Quality management is important for maximizing yield in continuous-flow manufacturing. However, it is more difficult to manage quality in continuous-flow manufacturing than in discrete manufacturing because partial defects can significantly affect the quality of an entire lot of final product. In this paper, a comprehensive framework that consists of three steps is proposed to predict defects and improve yield by using semi-supervised learning, time-series analysis, and classification model. In Step 1, semi-supervised learning using both labeled and unlabeled data is applied to generate quality values. In addition, feature values are predicted in time-series analysis in Step 2. Finally, in Step 3, we predict quality values based on the data obtained in Step 1 and Step 2 and calculate yield values with the use of the predicted value. Compared to a conventional production plan, the suggested plan increases yield by up to 8.7%. The production plan proposed in this study is expected to contribute to not only the continuous manufacturing process but the discrete manufacturing process. In addition, it can be used in early diagnosis of equipment failure.

show abstract

Remaining useful life predictions for turbofan engine degradation using semi-supervised deep architecture

Cited by 342 publications

References 13 publications

A Directed Acyclic Graph Network Combined With CNN and LSTM for Remaining Useful Life Prediction

A Directed Acyclic Graph Network Combined With CNN and LSTM for Remaining Useful Life Prediction

Deep Learning with Dynamically Weighted Loss Function for Sensor-Based Prognostics and Health Management

Quality Prediction and Yield Improvement in Process Manufacturing Based on Data Analytics

Contact Info

Product

Resources

About