A Real-Time Fault Early Warning Method for a High-Speed EMU Axle Box Bearing

An axle box bearing is one of the most important components of high-speed EMUs (electric multiple units), which runs at a very fast speed, suffers a heavy load, and operates under various complex working conditions. Once a bearing fault occurs, it not only has an enormous impact on the railway system, but also poses a threat to personal safety. Therefore, there is significant value in studying a real-time fault early warning of a high-speed EMU axle box bearing. However, to our best knowledge, there are three obvious defects in the existing fault early warning methods used for high-speed EMU axle box bearings: (1) these methods based on vibration are extremely mature, but there are no vibration sensors installed in high-speed EMU axle box because it will greatly increase the manufacturing cost; (2) a TADS (trackside acoustic device system) can effectively detect early failures, but only a portion of railways are equipped with such a facility; and (3) an EMU-ODS (electric multiple unit onboard detection system) has reported numerous untimely warnings, along with warnings of frequent occurrence being missed. Whereupon, a method is proposed to realize the fault early warning of an axle box bearing without installing a vibration sensor on the high-speed EMU in service, namely a MLSTM-iForest (multilayer long short-term memory–isolation forest). First, the time-series data of the temperature-related variables of the axle box bearing is used as the input of MLSTM to predict the axle box bearing temperature in the future. Then, the deviation index of the predicted axle box bearing temperature is calculated. Finally, the deviation index is input into an iForest algorithm for unsupervised classification to realize the fault early warning of an axle box bearing. Experimental results on high-speed EMU operation data sets demonstrated the availability and feasibility of the presented method toward achieving early fault warnings of a high-speed EMU axle box bearing.

Download Full-text

Electricity Theft Detection in Smart Grid Systems: A CNN-LSTM Based Approach

Energies ◽

10.3390/en12173310 ◽

2019 ◽

Vol 12 (17) ◽

pp. 3310 ◽

Cited By ~ 15

Author(s):

Md. Nazmul Hasan ◽

Rafia Nishat Toma ◽

Abdullah-Al Nahid ◽

M M Manjurul Islam ◽

Jong-Myon Kim

Keyword(s):

Smart Grid ◽

Power Consumption ◽

Smart Grids ◽

Time Series Data ◽

Short Term Memory ◽

Detection System ◽

Series Data ◽

Grid Data ◽

Minority Class ◽

Electricity Theft

Among an electricity provider’s non-technical losses, electricity theft has the most severe and dangerous effects. Fraudulent electricity consumption decreases the supply quality, increases generation load, causes legitimate consumers to pay excessive electricity bills, and affects the overall economy. The adaptation of smart grids can significantly reduce this loss through data analysis techniques. The smart grid infrastructure generates a massive amount of data, including the power consumption of individual users. Utilizing this data, machine learning and deep learning techniques can accurately identify electricity theft users. In this paper, an electricity theft detection system is proposed based on a combination of a convolutional neural network (CNN) and a long short-term memory (LSTM) architecture. CNN is a widely used technique that automates feature extraction and the classification process. Since the power consumption signature is time-series data, we were led to build a CNN-based LSTM (CNN-LSTM) model for smart grid data classification. In this work, a novel data pre-processing algorithm was also implemented to compute the missing instances in the dataset, based on the local values relative to the missing data point. Furthermore, in this dataset, the count of electricity theft users was relatively low, which could have made the model inefficient at identifying theft users. This class imbalance scenario was addressed through synthetic data generation. Finally, the results obtained indicate the proposed scheme can classify both the majority class (normal users) and the minority class (electricity theft users) with good accuracy.

Download Full-text

Real Time Magnitude Classification of Earthquake Waveforms using Deep Learning

10.5194/egusphere-egu21-15941 ◽

2021 ◽

Author(s):

Megha Chakraborty ◽

Georg Rümpker ◽

Horst Stöcker ◽

Wei Li ◽

Johannes Faber ◽

...

Keyword(s):

Deep Learning ◽

Real Time ◽

Time Series Data ◽

Short Term Memory ◽

Warning System ◽

Machine Learning Algorithms ◽

Series Data ◽

Maximum Information ◽

Earthquake Early Warning System ◽

Seismic Waveforms

<p>This study attempts to use Deep Learning architectures to design an efficient real time magnitude classifier for seismic events. Various combinations of Convolutional Neural Networks (CNNs) and Bi- & Uni-directional Long-Short Term Memory (LSTMs) and Gated Recurrent Unit (GRUs) are tried and tested to obtain the best performing model with optimum hyperparameters. In order to extract maximum information from the seismic waveforms, this study uses not only the time series data but also its corresponding Fourier Transform (spectrogram) as input. Furthermore, the Deep Learning architecture is combined with other machine learning algorithms to generate the final magnitude classifications. This study is likely to help seismologists in improving the Earthquake Early Warning System to avoid issuing false warnings, which not only alarms people unnecessarily but can also result in huge financial losses due to stoppage of industrial machinery etc.</p>

Download Full-text

Forecasting Groundwater Table in a Flood Prone Coastal City with Long Short-term Memory and Recurrent Neural Networks

Water ◽

10.3390/w11051098 ◽

2019 ◽

Vol 11 (5) ◽

pp. 1098 ◽

Cited By ~ 15

Author(s):

Benjamin D. Bowes ◽

Jeffrey M. Sadler ◽

Mohamed M. Morsy ◽

Madhur Behl ◽

Jonathan L. Goodall

Keyword(s):

Time Series ◽

Real Time ◽

Sea Level ◽

Continuous Time ◽

Time Series Data ◽

Short Term Memory ◽

Groundwater Table ◽

Series Data ◽

Storm Events ◽

Coastal City

Many coastal cities are facing frequent flooding from storm events that are made worse by sea level rise and climate change. The groundwater table level in these low relief coastal cities is an important, but often overlooked, factor in the recurrent flooding these locations face. Infiltration of stormwater and water intrusion due to tidal forcing can cause already shallow groundwater tables to quickly rise toward the land surface. This decreases available storage which increases runoff, stormwater system loads, and flooding. Groundwater table forecasts, which could help inform the modeling and management of coastal flooding, are generally unavailable. This study explores two machine learning models, Long Short-term Memory (LSTM) networks and Recurrent Neural Networks (RNN), to model and forecast groundwater table response to storm events in the flood prone coastal city of Norfolk, Virginia. To determine the effect of training data type on model accuracy, two types of datasets (i) the continuous time series and (ii) a dataset of only storm events, created from observed groundwater table, rainfall, and sea level data from 2010–2018 are used to train and test the models. Additionally, a real-time groundwater table forecasting scenario was carried out to compare the models’ abilities to predict groundwater table levels given forecast rainfall and sea level as input data. When modeling the groundwater table with observed data, LSTM networks were found to have more predictive skill than RNNs (root mean squared error (RMSE) of 0.09 m versus 0.14 m, respectively). The real-time forecast scenario showed that models trained only on storm event data outperformed models trained on the continuous time series data (RMSE of 0.07 m versus 0.66 m, respectively) and that LSTM outperformed RNN models. Because models trained with the continuous time series data had much higher RMSE values, they were not suitable for predicting the groundwater table in the real-time scenario when using forecast input data. These results demonstrate the first use of LSTM networks to create hourly forecasts of groundwater table in a coastal city and show they are well suited for creating operational forecasts in real-time. As groundwater table levels increase due to sea level rise, forecasts of groundwater table will become an increasingly valuable part of coastal flood modeling and management.

Download Full-text

Deep Learning-Based GNSS Network-Based Real-Time Kinematic Improvement for Autonomous Ground Vehicle Navigation

Journal of Sensors ◽

10.1155/2019/3737265 ◽

2019 ◽

Vol 2019 ◽

pp. 1-8 ◽

Cited By ~ 6

Author(s):

Hee-Un Kim ◽

Tae-Suk Bae

Keyword(s):

Deep Learning ◽

Real Time ◽

Time Series Data ◽

Short Term Memory ◽

Advanced Technology ◽

Series Data ◽

Measurement Unit ◽

Vehicle Navigation ◽

Real Time Kinematic ◽

Autonomous Vehicle Navigation

Much navigation over the last several decades has been aided by the global navigation satellite system (GNSS). In addition, with the advent of the multi-GNSS era, more and more satellites are available for navigation purposes. However, the navigation is generally carried out by point positioning based on the pseudoranges. The real-time kinematic (RTK) and the advanced technology, namely, the network RTK (NRTK), were introduced for better positioning and navigation. Further improved navigation was also investigated by combining other sensors such as the inertial measurement unit (IMU). On the other hand, a deep learning technique has been recently evolving in many fields, including automatic navigation of the vehicles. This is because deep learning combines various sensors without complicated analytical modeling of each individual sensor. In this study, we structured the multilayer recurrent neural networks (RNN) to improve the accuracy and the stability of the GNSS absolute solutions for the autonomous vehicle navigation. Specifically, the long short-term memory (LSTM) is an especially useful algorithm for time series data such as navigation with moderate speed of platforms. From an experiment conducted in a testing area, the LSTM algorithm developed the positioning accuracy by about 40% compared to GNSS-only navigation without any external bias information. Once the bias is taken care of, the accuracy will significantly be improved up to 8 times better than the GNSS absolute positioning results. The bias terms of the solution need to be estimated within the model by optimizing the layers as well as the nodes each layer, which should be done in further research.

Download Full-text

Adaptive, Unlabeled and Real-time Approximate-Learning Platform (AURA) for Personalized Epileptic Seizure Forecasting

10.1101/2021.09.30.21264287 ◽

2021 ◽

Author(s):

Yikai Yang ◽

Nhan Duy Truong ◽

Jason K. Eshraghian ◽

Armin Nikpour ◽

Omid Kavehei

Keyword(s):

Real Time ◽

Event Detection ◽

Prediction Models ◽

Time Series Data ◽

Detection System ◽

Series Data ◽

Patient Specific ◽

Forecasting Model ◽

Detection Model ◽

Learning Platform

A high performance event detection system is all you need for some predictive studies. Here, we present AURA: an Adaptive forecasting model trained with Unlabeled, Real-time data using internally generated Approximate labels on-the-fly. By harnessing the correlated nature of time-series data, a pair of detection and prediction models are coupled together such that the detection model generates labels automatically, which are then used to train the prediction model. AURA relies on several simple principles and assumptions: (i) the performance of an event prediction/forecasting model in the target application remains below the performance of an event detection model, (ii) detected events are treated as weak labels and deemed reliable enough for online training of a predictive model, and (iii) system performance and/or system responsive feedback characteristics can be tuned for a subject-under-test. For example, in medical patient monitoring, this enables personalization of the forecasting model. Seizure prediction is identified as an ideal test case of AURA, as preictal brainwaves are patient-specific and tailoring models to individual patients can significantly improve forecasting performance. AURA is used to generate an individual forecasting model for 5 patients, showing an average relative improvement in sensitivity by 33.33% and reduction in false alarms by 13.62%.

Download Full-text

Development of A Pharmaceutical Early Warning Scoring Model for Cardiac Arrest Using Deep Learning: Retrospective Cohort Study (Preprint)

10.2196/preprints.21720 ◽

2020 ◽

Author(s):

Hsiao-Ko Chang ◽

Hui-Chih Wang ◽

Chih-Fen Huang ◽

Feipei Lai

Keyword(s):

Time Series ◽

Cardiac Arrest ◽

Early Warning ◽

Time Series Data ◽

Predictive Accuracy ◽

Vital Signs ◽

Sliding Window ◽

Early Warning Score ◽

Series Data ◽

Dynamic Time

BACKGROUND In most of Taiwan’s medical institutions, congestion is a serious problem for emergency departments. Due to a lack of beds, patients spend more time in emergency retention zones, which make it difficult to detect cardiac arrest (CA). OBJECTIVE We seek to develop a pharmaceutical early warning model to predict cardiac arrest in emergency departments via drug classification and medical expert suggestion. METHODS We propose a new early warning score model for detecting cardiac arrest via pharmaceutical classification and by using a sliding window; we apply learning-based algorithms to time-series data for a Pharmaceutical Early Warning Scoring Model (PEWSM). By treating pharmaceutical features as a dynamic time-series factor for cardiopulmonary resuscitation (CPR) patients, we increase sensitivity, reduce false alarm rates and mortality, and increase the model’s accuracy. To evaluate the proposed model we use the area under the receiver operating characteristic curve (AUROC). RESULTS Four important findings are as follows: (1) We identify the most important drug predictors: bits, and replenishers and regulators of water and electrolytes. The best AUROC of bits is 85%; that of replenishers and regulators of water and electrolytes is 86%. These two features are the most influential of the drug features in the task. (2) We verify feature selection, in which accounting for drugs improve the accuracy: In Task 1, the best AUROC of vital signs is 77%, and that of all features is 86%. In Task 2, the best AUROC of all features is 85%, which demonstrates that thus accounting for the drugs significantly affects prediction. (3) We use a better model: For traditional machine learning, this study adds a new AI technology: the long short-term memory (LSTM) model with the best time-series accuracy, comparable to the traditional random forest (RF) model; the two AUROC measures are 85%. (4) We determine whether the event can be predicted beforehand: The best classifier is still an RF model, in which the observational starting time is 4 hours before the CPR event. Although the accuracy is impaired, the predictive accuracy still reaches 70%. Therefore, we believe that CPR events can be predicted four hours before the event. CONCLUSIONS This paper uses a sliding window to account for dynamic time-series data consisting of the patient’s vital signs and drug injections. In a comparison with NEWS, we improve predictive accuracy via feature selection, which includes drugs as features. In addition, LSTM yields better performance with time-series data. The proposed PEWSM, which offers 4-hour predictions, is better than the National Early Warning Score (NEWS) in the literature. This also confirms that the doctor’s heuristic rules are consistent with the results found by machine learning algorithms.

Download Full-text

Development of A Drug Early Warning System Model for Cardiac Arrest Using Deep Learning: Retrospective Cohort Study (Preprint)

10.2196/preprints.26783 ◽

2020 ◽

Author(s):

Hsiao-Ko Chang ◽

Hui-Chih Wang ◽

Chih-Fen Huang ◽

Feipei Lai

Keyword(s):

Machine Learning ◽

Time Series ◽

Cardiac Arrest ◽

Early Warning ◽

Time Series Data ◽

Predictive Accuracy ◽

Vital Signs ◽

Warning System ◽

Series Data ◽

Dynamic Time

BACKGROUND In most of Taiwan’s medical institutions, congestion is a serious problem for emergency departments. Due to a lack of beds, patients spend more time in emergency retention zones, which make it difficult to detect cardiac arrest (CA). OBJECTIVE We seek to develop a Drug Early Warning System Model (DEWSM), it included drug injections and vital signs as this research important features. We use it to predict cardiac arrest in emergency departments via drug classification and medical expert suggestion. METHODS We propose this new model for detecting cardiac arrest via drug classification and by using a sliding window; we apply learning-based algorithms to time-series data for a DEWSM. By treating drug features as a dynamic time-series factor for cardiopulmonary resuscitation (CPR) patients, we increase sensitivity, reduce false alarm rates and mortality, and increase the model’s accuracy. To evaluate the proposed model, we use the area under the receiver operating characteristic curve (AUROC). RESULTS Four important findings are as follows: (1) We identify the most important drug predictors: bits (intravenous therapy), and replenishers and regulators of water and electrolytes (fluid and electrolyte supplement). The best AUROC of bits is 85%, it means the medical expert suggest the drug features: bits, it will affect the vital signs, and then the evaluate this model correctly classified patients with CPR reach 85%; that of replenishers and regulators of water and electrolytes is 86%. These two features are the most influential of the drug features in the task. (2) We verify feature selection, in which accounting for drugs improve the accuracy: In Task 1, the best AUROC of vital signs is 77%, and that of all features is 86%. In Task 2, the best AUROC of all features is 85%, which demonstrates that thus accounting for the drugs significantly affects prediction. (3) We use a better model: For traditional machine learning, this study adds a new AI technology: the long short-term memory (LSTM) model with the best time-series accuracy, comparable to the traditional random forest (RF) model; the two AUROC measures are 85%. It can be seen that the use of new AI technology will achieve better results, currently comparable to the accuracy of traditional common RF, and the LSTM model can be adjusted in the future to obtain better results. (4) We determine whether the event can be predicted beforehand: The best classifier is still an RF model, in which the observational starting time is 4 hours before the CPR event. Although the accuracy is impaired, the predictive accuracy still reaches 70%. Therefore, we believe that CPR events can be predicted four hours before the event. CONCLUSIONS This paper uses a sliding window to account for dynamic time-series data consisting of the patient’s vital signs and drug injections. The National Early Warning Score (NEWS) only focuses on the score of vital signs, and does not include factors related to drug injections. In this study, the experimental results of adding the drug injections are better than only vital signs. In a comparison with NEWS, we improve predictive accuracy via feature selection, which includes drugs as features. In addition, we use traditional machine learning methods and deep learning (using LSTM method as the main processing time series data) as the basis for comparison of this research. The proposed DEWSM, which offers 4-hour predictions, is better than the NEWS in the literature. This also confirms that the doctor’s heuristic rules are consistent with the results found by machine learning algorithms.

Download Full-text

Time–frequency time–space LSTM for robust classification of physiological signals

Scientific Reports ◽

10.1038/s41598-021-86432-7 ◽

2021 ◽

Vol 11 (1) ◽

Author(s):

Tuan D. Pham

Keyword(s):

Time Series ◽

Network Architecture ◽

Time Series Data ◽

Short Term Memory ◽

Physiological Signals ◽

Series Data ◽

Time Frequency ◽

Time Space ◽

Deep Recurrent Neural Network

AbstractAutomated analysis of physiological time series is utilized for many clinical applications in medicine and life sciences. Long short-term memory (LSTM) is a deep recurrent neural network architecture used for classification of time-series data. Here time–frequency and time–space properties of time series are introduced as a robust tool for LSTM processing of long sequential data in physiology. Based on classification results obtained from two databases of sensor-induced physiological signals, the proposed approach has the potential for (1) achieving very high classification accuracy, (2) saving tremendous time for data learning, and (3) being cost-effective and user-comfortable for clinical trials by reducing multiple wearable sensors for data recording.

Download Full-text

Rapid Foreign Object Detection System on Seaweed Using VNIR Hyperspectral Imaging

Sensors ◽

10.3390/s21165279 ◽

2021 ◽

Vol 21 (16) ◽

pp. 5279

Author(s):

Dong-Hoon Kwak ◽

Guk-Jin Son ◽

Mi-Kyung Park ◽

Young-Duk Kim

Keyword(s):

Object Detection ◽

Real Time ◽

Hyperspectral Imaging ◽

High Speed ◽

Detection System ◽

Imaging Techniques ◽

Foreign Object ◽

Subtraction Method ◽

Foreign Objects ◽

Time Operation

The consumption of seaweed is increasing year by year worldwide. Therefore, the foreign object inspection of seaweed is becoming increasingly important. Seaweed is mixed with various materials such as laver and sargassum fusiforme. So it has various colors even in the same seaweed. In addition, the surface is uneven and greasy, causing diffuse reflections frequently. For these reasons, it is difficult to detect foreign objects in seaweed, so the accuracy of conventional foreign object detectors used in real manufacturing sites is less than 80%. Supporting real-time inspection should also be considered when inspecting foreign objects. Since seaweed requires mass production, rapid inspection is essential. However, hyperspectral imaging techniques are generally not suitable for high-speed inspection. In this study, we overcome this limitation by using dimensionality reduction and using simplified operations. For accuracy improvement, the proposed algorithm is carried out in 2 stages. Firstly, the subtraction method is used to clearly distinguish seaweed and conveyor belts, and also detect some relatively easy to detect foreign objects. Secondly, a standardization inspection is performed based on the result of the subtraction method. During this process, the proposed scheme adopts simplified and burdenless calculations such as subtraction, division, and one-by-one matching, which achieves both accuracy and low latency performance. In the experiment to evaluate the performance, 60 normal seaweeds and 60 seaweeds containing foreign objects were used, and the accuracy of the proposed algorithm is 95%. Finally, by implementing the proposed algorithm as a foreign object detection platform, it was confirmed that real-time operation in rapid inspection was possible, and the possibility of deployment in real manufacturing sites was confirmed.

Download Full-text