Predicting accurate remaining useful life (RUL) of components plays a crucial role in making optimal decision for maintenance management. As sensor technology develops, multiple sensors are used to collect information for monitoring the condition of components. Deep learning architectures, such as convolutional neural network (CNN) and long short term memory (LSTM), can be considered as a successful end-to-end framework to predict RUL from the multivariate time series collected by those sensors. For that, we employ an architecture combining the parallel branch of CNN in series with LSTM which is referred to as multi-head CNN-LSTM. Furthermore, we propose a combination of the network with time series prediction error analysis (PEA). The prediction errors on the entire time series are estimated by recursive least squares (RLS) and single exponential smoothing (SES) respectively. We analyze each of the two sequences of prediction errors with the exponentially weighted moving average (EWMA) and combine them with the Fisher's method. Finally, the output of the PEA is fed into the multi-head CNN-LSTM network as the additional input. We evaluate the performance of our method on the widely used C-MAPSS dataset. The experimental results suggest that using the PEA improves the performance of the deep learning-based RUL prediction model. Compared to other methods in recent literature, the proposed method achieves the state-of-the-art result on one sub-dataset and very competitive results on the others. In addition, it also shows promising results in the consecutive RUL prediction following the degradation process of components.
Multi-Head CNN-LSTM with Prediction Error Analysis for Remaining Useful Life Prediction / Mo, Hyunho; Lucca, Federico; Malacarne, Jonni; Iacca, Giovanni. - 2020-:(2020), pp. 164-171. (Intervento presentato al convegno 27th Conference of Open Innovations Association FRUCT, FRUCT 2020 tenutosi a Trento nel 7th-9th September 2020) [10.23919/FRUCT49677.2020.9211058].
Multi-Head CNN-LSTM with Prediction Error Analysis for Remaining Useful Life Prediction
Mo, Hyunho;Iacca, Giovanni
2020-01-01
Abstract
Predicting accurate remaining useful life (RUL) of components plays a crucial role in making optimal decision for maintenance management. As sensor technology develops, multiple sensors are used to collect information for monitoring the condition of components. Deep learning architectures, such as convolutional neural network (CNN) and long short term memory (LSTM), can be considered as a successful end-to-end framework to predict RUL from the multivariate time series collected by those sensors. For that, we employ an architecture combining the parallel branch of CNN in series with LSTM which is referred to as multi-head CNN-LSTM. Furthermore, we propose a combination of the network with time series prediction error analysis (PEA). The prediction errors on the entire time series are estimated by recursive least squares (RLS) and single exponential smoothing (SES) respectively. We analyze each of the two sequences of prediction errors with the exponentially weighted moving average (EWMA) and combine them with the Fisher's method. Finally, the output of the PEA is fed into the multi-head CNN-LSTM network as the additional input. We evaluate the performance of our method on the widely used C-MAPSS dataset. The experimental results suggest that using the PEA improves the performance of the deep learning-based RUL prediction model. Compared to other methods in recent literature, the proposed method achieves the state-of-the-art result on one sub-dataset and very competitive results on the others. In addition, it also shows promising results in the consecutive RUL prediction following the degradation process of components.File | Dimensione | Formato | |
---|---|---|---|
Multi-Head CNN-LSTM with Prediction Error Analysis for Remaining Useful Life Prediction.pdf
Solo gestori archivio
Tipologia:
Versione editoriale (Publisher’s layout)
Licenza:
Tutti i diritti riservati (All rights reserved)
Dimensione
910.69 kB
Formato
Adobe PDF
|
910.69 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione