Anomaly Detection of the Brake Operating Unit on Metro Vehicles Using a One-Class LSTM Autoencoder

Detecting anomalies in the Brake Operating Unit (BOU) braking system of metro trains is very important for trains’ reliability and safety. However, current periodic maintenance and inspection cannot detect anomalies at an early stage. In addition, constructing a stable and accurate anomaly detection system is a very challenging task. Hence, in this work, we propose a method for detecting anomalies of BOU on metro vehicles using a one-class long short-term memory (LSTM) autoencoder. First, we extracted brake cylinder (BC) pressure data from the BOU data since one of the anomaly cases of metro trains is that BC pressure relief time is delayed by 4 s. After that, extracted BC pressure data is split into subsequences which are fed into our proposed one-class LSTM autoencoder which consists of two LSTM blocks (encoder and decoder). The one-class LSTM autoencoder is trained using training data which only consists of normal subsequences. To detect anomalies from test data that contain abnormal subsequences, the mean absolute error (MAE) for each subsequence is calculated. When the error is larger than a predefined threshold which was set to the maximum value of MAE in the training (normal) dataset, we can declare that example an anomaly. We conducted the experiments with the BOU data of metro trains in Korea. Experimental results show that our proposed method can detect anomalies of the BOU data well.

Download Full-text

Data-Driven Anomaly Detection in High-Voltage Transformer Bushings with LSTM Auto-Encoder

Sensors ◽

10.3390/s21217426 ◽

2021 ◽

Vol 21 (21) ◽

pp. 7426

Author(s):

Imene Mitiche ◽

Tony McGrail ◽

Philip Boreham ◽

Alan Nesbitt ◽

Gordon Morison

Keyword(s):

Anomaly Detection ◽

High Voltage ◽

Short Term Memory ◽

Absolute Error ◽

Real World Data ◽

Voltage Transformer ◽

Power Outage ◽

Phase Measurements ◽

High Voltage Transformer ◽

Supply Industry

The reliability and health of bushings in high-voltage (HV) power transformers is essential in the power supply industry, as any unexpected failure can cause power outage leading to heavy financial losses. The challenge is to identify the point at which insulation deterioration puts the bushing at an unacceptable risk of failure. By monitoring relevant measurements we can trace any change that occurs and may indicate an anomaly in the equipment’s condition. In this work we propose a machine-learning-based method for real-time anomaly detection in current magnitude and phase angle from three bushing taps. The proposed method is fast, self-supervised and flexible. It consists of a Long Short-Term Memory Auto-Encoder (LSTMAE) network which learns the normal current and phase measurements of the bushing and detects any point when these measurements change based on the Mean Absolute Error (MAE) metric evaluation. This approach was successfully evaluated using real-world data measured from HV transformer bushings where anomalous events have been identified.

Download Full-text

Deep-Learning-Based Approach to Anomaly Detection Techniques for Large Acoustic Data in Machine Operation

Sensors ◽

10.3390/s21165446 ◽

2021 ◽

Vol 21 (16) ◽

pp. 5446

Author(s):

Hyojung Ahn ◽

Inchoon Yeo

Keyword(s):

Deep Learning ◽

Anomaly Detection ◽

Short Term Memory ◽

Detection System ◽

Noisy Environments ◽

Acoustic Data ◽

Detection Techniques ◽

Machine Operation ◽

Detection Technology ◽

Environment Noise

As the workforce shrinks, the demand for automatic, labor-saving, anomaly detection technology that can perform maintenance on advanced equipment such as vehicles has been increasing. In a vehicular environment, noise in the cabin, which directly affects users, is considered an important factor in lowering the emotional satisfaction of the driver and/or passengers in the vehicles. In this study, we provide an efficient method that can collect acoustic data, measured using a large number of microphones, in order to detect abnormal operations inside the machine via deep learning in a quick and highly accurate manner. Unlike most current approaches based on Long Short-Term Memory (LSTM) or autoencoders, we propose an anomaly detection (AD) algorithm that can overcome the limitations of noisy measurement and detection system anomalies via noise signals measured inside the mechanical system. These features are utilized to train a variety of anomaly detection models for demonstration in noisy environments with five different errors in machine operation, achieving an accuracy of approximately 90% or more.

Download Full-text

Fall detection system based on BiLSTM neural network

Transactions on Machine Learning and Artificial Intelligence ◽

10.14738/tmlai.75.6318 ◽

2019 ◽

Vol 7 (5) ◽

pp. 01-12

Author(s):

Biao YE ◽

Lasheng Yu

Keyword(s):

Neural Network ◽

Short Term Memory ◽

Detection System ◽

Input Sequence ◽

Fall Detection ◽

Detection Algorithm ◽

Training Data ◽

Sensor Data ◽

Data Set ◽

Sequence Modeling

The purpose of this article is to analyze the characteristics of human fall behavior to design a fall detection system. The existing fall detection algorithms have problems such as poor adaptability, single function and difficulty in processing large data and strong randomness. Therefore, a long-term and short-term memory recurrent neural network is used to improve the effect of falling behavior detection by exploring the internal correlation between sensor data. Firstly, the serialization representation method of sensor data, training data and detection input data is designed. The BiLSTM network has the characteristics of strong ability to sequence modeling and it is used to reduce the dimension of the data required by the fall detection model. then, the BiLSTM training algorithm for fall detection and the BiLSTM-based fall detection algorithm convert the fall detection into the classification problem of the input sequence; finally, the BiLSTM-based fall detection system was implemented on the TensorFlow platform. The detection and analysis of system were carried out using a bionic experiment data set which mimics a fall. The experimental results verify that the system can effectively improve the accuracy of fall detection to 90.47%. At the same time, it can effectively detect the behavior of Near-falling, and help to take corresponding protective measures.

Download Full-text

Danger-Pose Detection System Using Commodity Wi-Fi for Bathroom Monitoring

Sensors ◽

10.3390/s19040884 ◽

2019 ◽

Vol 19 (4) ◽

pp. 884 ◽

Cited By ~ 9

Author(s):

Zizheng Zhang ◽

Shigemi Ishida ◽

Shigeaki Tagashira ◽

Akira Fukuda

Keyword(s):

Machine Learning ◽

Anomaly Detection ◽

Detection Method ◽

Detection System ◽

Wearable Sensors ◽

Low Frequency ◽

Training Data ◽

Support Vector ◽

Dynamic Features ◽

Pose Detection

A bathroom has higher probability of accidents than other rooms due to a slippery floor and temperature change. Because of high privacy and humidity, we face difficulties in monitoring inside a bathroom using traditional healthcare methods based on cameras and wearable sensors. In this paper, we present a danger-pose detection system using commodity Wi-Fi devices, which can be applied to bathroom monitoring, preserving privacy. A machine learning-based detection method usually requires data collected in target situations, which is difficult in detection-of-danger situations. We therefore employ a machine learning-based anomaly-detection method that requires a small amount of data in anomaly conditions, minimizing the required training data collected in dangerous conditions. We first derive the amplitude and phase shift from Wi-Fi channel state information (CSI) to extract low-frequency components that are related to human activities. We then separately extract static and dynamic features from the CSI changes in time. Finally, the static and dynamic features are fed into a one-class support vector machine (SVM), which is used as an anomaly-detection method, to classify whether a user is not in bathtub, bathing safely, or in dangerous conditions. We conducted experimental evaluations and demonstrated that our danger-pose detection system achieved a high detection performance in a non-line-of-sight (NLOS) scenario.

Download Full-text

A Novel Deep Learning Approach for Anomaly Detection of Time Series Data

Scientific Programming ◽

10.1155/2021/6636270 ◽

2021 ◽

Vol 2021 ◽

pp. 1-11

Author(s):

Zhiwei Ji ◽

Jiaheng Gong ◽

Jiarui Feng

Keyword(s):

Time Series ◽

Anomaly Detection ◽

Time Series Data ◽

Short Term Memory ◽

Training Data ◽

Series Data ◽

Real Time System ◽

Statistical Strategy ◽

Anomaly Classification ◽

Real World Datasets

Anomalies in time series, also called “discord,” are the abnormal subsequences. The occurrence of anomalies in time series may indicate that some faults or disease will occur soon. Therefore, development of novel computational approaches for anomaly detection (discord search) in time series is of great significance for state monitoring and early warning of real-time system. Previous studies show that many algorithms were successfully developed and were used for anomaly classification, e.g., health monitoring, traffic detection, and intrusion detection. However, the anomaly detection of time series was not well studied. In this paper, we proposed a long short-term memory- (LSTM-) based anomaly detection method (LSTMAD) for discord search from univariate time series data. LSTMAD learns the structural features from normal (nonanomalous) training data and then performs anomaly detection via a statistical strategy based on the prediction error for observed data. In our experimental evaluation using public ECG datasets and real-world datasets, LSTMAD detects anomalies more accurately than other existing approaches in comparison.

Download Full-text

Real-Time Prediction of Temperature Distribution in Additive Manufacturing Processes Based on Machine Learning

Volume 2A: Advanced Manufacturing ◽

10.1115/imece2020-24107 ◽

2020 ◽

Author(s):

Sharareh Bayat ◽

Mohammad Mohseni ◽

Delaram Behnami ◽

Purang Abolmaesumi

Keyword(s):

Machine Learning ◽

Additive Manufacturing ◽

Real Time ◽

Manufacturing Process ◽

Network Architecture ◽

Short Term Memory ◽

Absolute Error ◽

Training Data ◽

Computational Time ◽

Element Analysis

Abstract Simulation tools improve various aspects of the additive manufacturing process, however, they come with an undesirable computational time for real-world applications. Finite element analysis (FEA) that solves partial differential equations (PDE) presents promising capabilities in simple additive manufactured components as an expository problem. Yet, PDE-based solutions take significantly long CPU time due to a large number of timesteps required to simulate an additively manufactured part. With modern machine learning (ML) capabilities, a new shift towards integration of FEA and ML has been introduced, where ML algorithms emulate the behavior of the time-consuming PDE-solver for real-time analysis of PDE in a given application. In this paper, we present a deep learning (DL) model that can substitute the thermal analysis of the additive manufacturing process. The training data is obtained by sampling the established physical model’s behavior over different temperatures, cooling rates, and part’s geometries. The network architecture is composed of a Long Short-Term Memory (LSTM) to model the temporal sequence of deposition temperatures derived by PDEs. The reported R2 value on validations data is 97%, while the Mean Absolute Error (MAE) is 0.04. This paper compares the performance between the PDE and DL forecast for the thermal results. We show DL models are promising for simulation of the additive manufacturing process, and can be reliable alternatives for computationally-expensive FEM tools.

Download Full-text

Using Acoustic Information to Diagnose the Health of A Printer

Electronic Imaging ◽

10.2352/issn.2470-1173.2020.15.color-352 ◽

2020 ◽

Vol 2020 (15) ◽

pp. 353-1-353-6

Author(s):

Chin-Ning Chen ◽

Katy Ferguson ◽

Anton Wiranata ◽

Mark Shaw ◽

Wan-Eih Huang ◽

...

Keyword(s):

Anomaly Detection ◽

Data Augmentation ◽

Detection System ◽

Model Performance ◽

Principal Component ◽

Training Data ◽

Acoustic Data ◽

Acoustic Information ◽

Anomaly Detection System ◽

Previous Step

We present a sound-based anomaly detection system to diagnose printer health. Also, we improve the model performance by using acoustic data augmentation. We first use the detector to extract the important acoustic information from the input printer sound. Second, we use principal component analysis to do feature extraction. Third, we feed the extracted features from the previous step into the two different anomaly detection models to evaluate the model performances. Finally, we go through the same system pipeline with different augmented training data to see whether or not acoustic data augmentation can improve the model performance.

Download Full-text

MADGAN: unsupervised medical anomaly detection GAN using multiple adjacent brain MRI slice reconstruction

BMC Bioinformatics ◽

10.1186/s12859-020-03936-1 ◽

2021 ◽

Vol 22 (S2) ◽

Author(s):

Changhee Han ◽

Leonardo Rundo ◽

Kohei Murao ◽

Tomoyuki Noguchi ◽

Yuki Shimahara ◽

...

Keyword(s):

Brain Metastases ◽

Anomaly Detection ◽

Late Stage ◽

Large Scale ◽

Early Stage ◽

Brain Mri ◽

Disease Diagnosis ◽

Training Data ◽

Mri Scans ◽

Adjacent Brain

Abstract Background Unsupervised learning can discover various unseen abnormalities, relying on large-scale unannotated medical images of healthy subjects. Towards this, unsupervised methods reconstruct a 2D/3D single medical image to detect outliers either in the learned feature space or from high reconstruction loss. However, without considering continuity between multiple adjacent slices, they cannot directly discriminate diseases composed of the accumulation of subtle anatomical anomalies, such as Alzheimer’s disease (AD). Moreover, no study has shown how unsupervised anomaly detection is associated with either disease stages, various (i.e., more than two types of) diseases, or multi-sequence magnetic resonance imaging (MRI) scans. Results We propose unsupervised medical anomaly detection generative adversarial network (MADGAN), a novel two-step method using GAN-based multiple adjacent brain MRI slice reconstruction to detect brain anomalies at different stages on multi-sequence structural MRI: (Reconstruction) Wasserstein loss with Gradient Penalty + 100 $$\ell _1$$ ℓ 1 loss—trained on 3 healthy brain axial MRI slices to reconstruct the next 3 ones—reconstructs unseen healthy/abnormal scans; (Diagnosis) Average $$\ell _2$$ ℓ 2 loss per scan discriminates them, comparing the ground truth/reconstructed slices. For training, we use two different datasets composed of 1133 healthy T1-weighted (T1) and 135 healthy contrast-enhanced T1 (T1c) brain MRI scans for detecting AD and brain metastases/various diseases, respectively. Our self-attention MADGAN can detect AD on T1 scans at a very early stage, mild cognitive impairment (MCI), with area under the curve (AUC) 0.727, and AD at a late stage with AUC 0.894, while detecting brain metastases on T1c scans with AUC 0.921. Conclusions Similar to physicians’ way of performing a diagnosis, using massive healthy training data, our first multiple MRI slice reconstruction approach, MADGAN, can reliably predict the next 3 slices from the previous 3 ones only for unseen healthy images. As the first unsupervised various disease diagnosis, MADGAN can reliably detect the accumulation of subtle anatomical anomalies and hyper-intense enhancing lesions, such as (especially late-stage) AD and brain metastases on multi-sequence MRI scans.

Download Full-text

Anomaly Detection Approach Based on Deep Neural Network and Dropout

Baghdad Science Journal ◽

10.21123/bsj.2020.17.2(si).0701 ◽

2020 ◽

Vol 17 (2(SI)) ◽

pp. 0701

Author(s):

Zaid Hussien et al.

Keyword(s):

Neural Network ◽

Feature Selection ◽

Intrusion Detection ◽

Anomaly Detection ◽

Deep Neural Network ◽

Detection System ◽

Early Stage ◽

High Detection Rate ◽

Network Intrusion ◽

Proposed Model

Regarding to the computer system security, the intrusion detection systems are fundamental components for discriminating attacks at the early stage. They monitor and analyze network traffics, looking for abnormal behaviors or attack signatures to detect intrusions in early time. However, many challenges arise while developing flexible and efficient network intrusion detection system (NIDS) for unforeseen attacks with high detection rate. In this paper, deep neural network (DNN) approach was proposed for anomaly detection NIDS. Dropout is the regularized technique used with DNN model to reduce the overfitting. The experimental results applied on NSL_KDD dataset. SoftMax output layer has been used with cross entropy loss function to enforce the proposed model in multiple classification, including five labels, one is normal and four others are attacks (Dos, R2L, U2L and Probe). Accuracy metric was used to evaluate the model performance. The proposed model accuracy achieved to 99.45%. Commonly the recognition time is reduced in the NIDS by using feature selection technique. The proposed DNN classifier implemented with feature selection algorithm, and obtained on accuracy reached to 99.27%.

Download Full-text

Detecting Deception from Gaze and Speech Using a Multimodal Attention LSTM-Based Framework

Applied Sciences ◽

10.3390/app11146393 ◽

2021 ◽

Vol 11 (14) ◽

pp. 6393

Author(s):

Ascensión Gallardo-Antolín ◽

Juan M. Montero

Keyword(s):

Short Term Memory ◽

Deception Detection ◽

Detection System ◽

Support Vector ◽

The Gaze ◽

Combination Strategies ◽

Speech Feature ◽

Speech Features ◽

The One ◽

Modal Systems

The automatic detection of deceptive behaviors has recently attracted the attention of the research community due to the variety of areas where it can play a crucial role, such as security or criminology. This work is focused on the development of an automatic deception detection system based on gaze and speech features. The first contribution of our research on this topic is the use of attention Long Short-Term Memory (LSTM) networks for single-modal systems with frame-level features as input. In the second contribution, we propose a multimodal system that combines the gaze and speech modalities into the LSTM architecture using two different combination strategies: Late Fusion and Attention-Pooling Fusion. The proposed models are evaluated over the Bag-of-Lies dataset, a multimodal database recorded in real conditions. On the one hand, results show that attentional LSTM networks are able to adequately model the gaze and speech feature sequences, outperforming a reference Support Vector Machine (SVM)-based system with compact features. On the other hand, both combination strategies produce better results than the single-modal systems and the multimodal reference system, suggesting that gaze and speech modalities carry complementary information for the task of deception detection that can be effectively exploited by using LSTMs.

Download Full-text