“…This is an interesting observation since the inclusion of additional sensors is typically expected to improve or at least maintain the prediction accuracy of deep learning models. However, Gupta et al (2021) have also reported a similar observation that incorporating sensors from all the components did not improve the predictions of TOAB from one component. Therefore, only IC models, i.e., CC-LSTM models with only 10 sensors corresponding to each component were considered for evaluating the ECC-LSTM approach.…”
Breakdowns and unplanned shutdowns in industrial processes and equipment can lead to significant loss of availability and revenue. It is imperative to perform optimal maintenance of such systems when signs of abnormal behavior are detected and before they propagate and lead to catastrophic failure. This is particularly challenging in systems with interconnected multiple components as it is difficult to isolate the effect of one component on the operation of other components in the system. In this work, an ensemble approach based on Cascaded Convolutional neural network and Long Short-term Memory (CC-LSTM) network models is proposed for detecting and predicting the time of onset of faults in interconnected multicomponent systems. The performance of the ensemble CC-LSTM model was demonstrated on an industrial 4-component system and was found to improve the accuracy of onset time predictions by ~15% compared to individual CC-LSTM models and ~25-40% compared to commonly used deep learning techniques such as dense neural networks, convolutional neural networks and LSTMs. The CC-LSTM and the ensemble models also had the lowest missed detection rates and zero false positive rates making them ideal for real-time monitoring and fault detection in multicomponent systems.
“…This is an interesting observation since the inclusion of additional sensors is typically expected to improve or at least maintain the prediction accuracy of deep learning models. However, Gupta et al (2021) have also reported a similar observation that incorporating sensors from all the components did not improve the predictions of TOAB from one component. Therefore, only IC models, i.e., CC-LSTM models with only 10 sensors corresponding to each component were considered for evaluating the ECC-LSTM approach.…”
Breakdowns and unplanned shutdowns in industrial processes and equipment can lead to significant loss of availability and revenue. It is imperative to perform optimal maintenance of such systems when signs of abnormal behavior are detected and before they propagate and lead to catastrophic failure. This is particularly challenging in systems with interconnected multiple components as it is difficult to isolate the effect of one component on the operation of other components in the system. In this work, an ensemble approach based on Cascaded Convolutional neural network and Long Short-term Memory (CC-LSTM) network models is proposed for detecting and predicting the time of onset of faults in interconnected multicomponent systems. The performance of the ensemble CC-LSTM model was demonstrated on an industrial 4-component system and was found to improve the accuracy of onset time predictions by ~15% compared to individual CC-LSTM models and ~25-40% compared to commonly used deep learning techniques such as dense neural networks, convolutional neural networks and LSTMs. The CC-LSTM and the ensemble models also had the lowest missed detection rates and zero false positive rates making them ideal for real-time monitoring and fault detection in multicomponent systems.
Accurate remaining useful life (RUL) prediction is one of the most challenging problems in the prognostics of turbofan engines. Recently, RUL prediction methods for turbofan engines mainly involve data-driven models. Preprocessing the sensor data is essential for the performance of the prognostic models. Most studies on turbofan engines use piecewise linear (PwL) labeling, which starts with a constant initial RUL value in normal/healthy operating time. In this study, we designed a prognostic procedure that includes difference-based feature construction, change-point-detection-based PwL labeling, and a 1D-CNN-LSTM (one-dimensional–convolutional neural network–long short-term memory) hybrid neural network model for RUL prediction. The procedure was evaluated on the subset FD001 of the C-MAPSS dataset. The proposed procedure was compared with machine learning and deep learning models with and without the new difference feature. Also, the results were compared with the studies that used similar labeling approaches. Our analysis of the numerical results underscores the clear superiority of the proposed 1D-CNN-LSTM model with the difference feature in RUL prediction, with a score of 437.2 and an RMSE value of 16.1. This result illustrates the superior predictive capability of the 1D-CNN-LSTM model, which outperformed traditional machine learning methods and one of the earliest deep learning methods. These findings emphasize the superior predictive capability of the 1D-CNN-LSTM model and underline the potential of the feature engineering process for more accurate and robust RUL prediction in the context of turbofan engine prognostics.
“…Among the NN models, RNN models have also demonstrated potential for change point detection. For example, an RNN model was proposed to predict the change points from data of sensors attached to industrial equipment parts [13]. Moreover, an RNN model outperformed the previous method in detecting the change points in multiple time-series load data of an electric power company for power outage analysis [14].…”
Eye-fixation-related potential (EFRP)—an event-related potential that is time-locked to the saccade offset (SO)—can be measured without synchronizing with time when external stimuli occur. Such an advantage in measurement enables the mean amplitude of the EFRP to be used to estimate the cognitive workload, which is known to change the amplitude, under real-world conditions. However, to observe EFRPs reliably, the SO timing must be correctly and consistently determined in milliseconds owing to the high temporal resolution of the electroencephalogram (EEG). As the electrooculogram (EOG) is commonly measured simultaneously with the EEG and the SO timing is reflected as a steep change in the waveforms, attempts have been made to determine the SO timing from EOG signals visually (the VD method). However, the SO timing detected by the VD method may be inconsistent across trials. We propose a gated recurrent unit—a recurrent neural network model—to detect the SO timing from EOGs consistently and automatically. We used EOG data from a task that mimics visual inspections, in which participants periodically traversed their eyes from left to right, for the model training. As a result, the amplitudes of the EFRPs based on the proposed method were significantly larger than those based on the VD method and the previous automatic method. This suggests that the proposed method can prevent the decrease in EFRP amplitudes owing to the inconsistent determination of the SO timing and increase the applicability of cognitive workload estimation using the EFRP in real-world environments.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.