Deep learning through LSTM classification and regression for transmission line fault detection, diagnosis and location in large-scale multi-machine power systems

Belagoune, Soufiane; Bali, Noureddine; Bakdi, Azzeddine; Baadji, Bousaadia; Atif, Karim

doi:10.1016/j.measurement.2021.109330

Cited by 166 publications

(65 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Through the operation of the control gates, the LSTM is hence able to minimise errors by retaining relevant information and forgetting irrelevant information as needed. Similar to the ANN and deep learning ML algorithms in general, the LSTM requires high computational power to train and develop predictive models 52 , 53 . The high memory-bandwidth needed given the presence of linear layers in each cell may reduce the hardware efficiency of this algorithm.…”

Section: Methodsmentioning

confidence: 99%

Predicting suspended sediment load in Peninsular Malaysia using support vector machine and deep learning algorithms

Essam

Huang

Birima

et al. 2022

Sci Rep

View full text Add to dashboard Cite

High loads of suspended sediments in rivers are known to cause detrimental effects to potable water sources, river water quality, irrigation activities, and dam or reservoir operations. For this reason, the study of suspended sediment load (SSL) prediction is important for monitoring and damage mitigation purposes. The present study tests and develops machine learning (ML) models, based on the support vector machine (SVM), artificial neural network (ANN) and long short-term memory (LSTM) algorithms, to predict SSL based on 11 different river data sets comprising of streamflow (SF) and SSL data obtained from the Malaysian Department of Irrigation and Drainage. The main objective of the present study is to propose a single model that is capable of accurately predicting SSLs for any river data set within Peninsular Malaysia. The ANN3 model, based on the ANN algorithm and input scenario 3 (inputs consisting of current-day SF, previous-day SF, and previous-day SSL), is determined as the best model in the present study as it produced the best predictive performance for 5 out of 11 of the tested data sets and obtained the highest average RM with a score of 2.64 when compared to the other tested models, indicating that it has the highest reliability to produce relatively high-accuracy SSL predictions for different data sets. Therefore, the ANN3 model is proposed as a universal model for the prediction of SSL within Peninsular Malaysia.

show abstract

Section: Methodsmentioning

confidence: 99%

Predicting suspended sediment load in Peninsular Malaysia using support vector machine and deep learning algorithms

Essam

Huang

Birima

et al. 2022

Sci Rep

View full text Add to dashboard Cite

show abstract

“…By continuously manipulating information cyclically, it can connect the neuron to itself across time, thereby solving the problem of low predictive performance for sequence learning [30]. RNN can learn and extract features directly from inputs in time series domain to complete related learning [3].…”

Section: 2mentioning

confidence: 99%

Intelligent fault diagnosis method of spacecraft control system based on sequence data-image mapping

Liang¹,

Liu²,

Liu³

et al. 2024

MFC

View full text Add to dashboard Cite

<p style='text-indent:20px;'>Satellite networking, as the future development direction of aero-space, requires high-precision autonomous fault diagnosis capability for a single satellite. In this paper, aiming at the characteristics of closed-loop fault propagation and high data dimensionality of spacecraft control system, neural network algorithms are conducted to study the fault diagnosis of spacecraft high-dimensional coupled data. Based on the ground test data of a certain spacecraft, this paper converts the high-dimensional sequence data into grayscale images, and then uses Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) to diagnose them respectively. The effectiveness of the methods in this paper is illustrated by comparing and validating them with three non-image-based machine learning algorithms, namely, K-NearestNeighbor, Bayesian classifier, and K-NearestNeighbor based on Principal Component Analysis.</p>

show abstract

“…As a variance of RNN in particular, Long short-term memory (LSTM), originally applied in NLP tasks, also yielded promising results for time series classification [5,34,36]. The cell unit and three gates (input gate, output gate and forget gate) in the LSTM unit allow this architecture to remember values over arbitrary time intervals and regulate the flow of information [27].…”

Section: Lstm-based Modelmentioning

confidence: 99%

Machine Learning Applied for Spectra Classification

Sun

Brockhauser

Hegedűs

2021

Computational Science and Its Applications – ICCSA 2021

View full text Add to dashboard Cite

Spectroscopy experiment techniques are widely used and produce a huge amount of data especially in facilities with very high repetition rates. In High Energy Density (HED) experiments with high-density materials, changes in pressure will cause changes in the spectral peak. Immediate feedback on the actual status (e.g. time-resolved status of the sample) would be essential to quickly judge how to proceed with the experiment. The two major spectral changes we aim to capture are either the change of intensity distribution (e.g., drop or appearance) of peaks at certain locations, or the shift of those on the spectrum. In this work, we apply recent popular machine learning/deep learning models to HED experimental spectra data classification. The models we presented range from supervised deep neural networks (state-of-the-art LSTM-based model and Transformer-based model) to unsupervised spectral clustering algorithm. These are the common architectures for time series processing. The PCA method is used as data preprocessing for dimensionality reduction. Three different ML algorithms are evaluated and compared for the classification task. The results show that all three methods can achieve 100% classification confidence. Among them, the spectra clustering method consumes the least calculation time (0.069 s), and the transformer-based method uses the most training time (0.204 s).

show abstract

Deep learning through LSTM classification and regression for transmission line fault detection, diagnosis and location in large-scale multi-machine power systems

Cited by 166 publications

References 39 publications

Predicting suspended sediment load in Peninsular Malaysia using support vector machine and deep learning algorithms

Predicting suspended sediment load in Peninsular Malaysia using support vector machine and deep learning algorithms

Intelligent fault diagnosis method of spacecraft control system based on sequence data-image mapping

Machine Learning Applied for Spectra Classification

Contact Info

Product

Resources

About