Power Quality Disturbance Classification Based on Compressed Sensing and Deep Convolution Neural Networks

Wang, Jidong; Xu, Zhilin; Che, Yanbo

doi:10.1109/access.2019.2922367

Cited by 63 publications

(22 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These algorithms’ structure contains hierarchical architecture and different non‐linear layers in order to extract generic features from big trained data [21]. The deep convolution neural network is utilised as an end‐to‐end classifier to decrease the consuming time in the pre‐processing operations [22]. The multi‐fusion convolutional neural network as a deep learning algorithm is introduced in [23] to automatically analyse complex PQ disturbances.…”

Section: Introductionmentioning

confidence: 99%

Categorisation of power quality problems using long short‐term memory networks

Abdelsalam

Hassanin

Hasanien

2021

IET Generation Trans & Dist

View full text Add to dashboard Cite

Recognition of power quality (PQ) troubles is a critical task in the electrical power system. All previous works solve this problem using two‐step methodology: Feature extraction and classification steps with each step using its own techniques, and this consumes a computation time. The purpose of this study is to utilise a novel artificial intelligence (AI) technique for recognition of PQ events. The proposed AI technique is the long short‐term memory (LSTM) network, which detects and classifies the PQ events in one step. This technique extracts amplitude, disturbance duration and total harmonic distortion from the captured waveform, and the LSTM uses its own rules to classify the PQ events. Many simple PQ events such as interruption, sag, flicker, swell and surge or complex PQ events such as sag plus harmonics and swell plus harmonics are generated using MATLAB programming environment to evaluate the performance of LSTM. Also, real‐time measurements are collected from an industrial substation and are used to ensure the effectiveness of the proposed LSTM technique. A comparison with other techniques is conducted and the results verify the good performance of LSTM in classifying the PQ problems.

show abstract

Section: Introductionmentioning

confidence: 99%

Categorisation of power quality problems using long short‐term memory networks

Abdelsalam

Hassanin

Hasanien

2021

IET Generation Trans & Dist

View full text Add to dashboard Cite

show abstract

“…Recently, the artificial intelligence techniques are used to classify PQ problems. These techniques include artificial neural network (ANN) [32,33], probabilistic neural network (PNN) [34], support vector machine (SVM) [35], extreme learning machine (ELM) [36], K-nearest neighbor [37], decision tree (DT) [38], deep convolutional network [39] and long short-term memory networks [40]. These techniques have some shortages such as disability to classify complex PQ events and the need to be retrained in case of appearing a new PQ event [41].…”

Section: Introductionmentioning

confidence: 99%

A Generalized Approach for Power Quality Disturbances Recognition Based on Kalman Filter

2021

View full text Add to dashboard Cite

This paper presents a new automatic detection and classification approach of power quality (PQ) problems using Kalman filter. Kalman filter is used as an estimator to calculate the fundamental frequency and harmonic components amplitudes of the voltage or current signals. Then the instantaneous total harmonic distortion (iTHD) and the energy are calculated. For each half cycle of the processed signal, five decision quantities are calculated based on iTHD and energy and these quantities are the three consecutive maximum values of iTHD, standard deviation and energy difference between distorted signal and its fundamental frequency component . Decision rules based on these decision quantities are applied to identify and classify the PQ events in this captured signal. The proposed approach is tested on single and combined PQ events that are generated using the MATLAB with the help of mathematical models that are conformity with standard IEEE-1159. The performance is assessed using more than 100 dataset of every PQ event and the results show that the accuracy is 100 and 98.8 for noiseless and high-level of noise, respectively. In addition, the proposed approach performance is validated through comparisons with other classification. Several practical PQ events are generated by lab experiments to validate the proposed approach. The simulation and experimental results show that the proposed approach is efficient and robust and can be implemented to design PQ monitoring device.

show abstract

“…Furthermore, there are many research works using CNN for power quality (PQ) analysis. Several works that use DL in PQ have been reported and they are interested in not detection but usually a classification of parameters or events in PQ [15–27]. In this study, a new method based on CNN is proposed for fast and accurate detection of phase and amplitude information of rapidly time‐varying harmonic components of voltages and currents of the power systems.…”

Section: Introductionmentioning

confidence: 99%