scholarly journals A Novel Pix2Pix Enabled Traveling Wave-Based Fault Location Method

Sensors ◽  
2021 ◽  
Vol 21 (5) ◽  
pp. 1633
Author(s):  
Jinxian Zhang ◽  
Qingwu Gong ◽  
Haojie Zhang ◽  
Yubo Wang ◽  
Yilin Wang
Keyword(s):  
Traveling Wave ◽  
Fault Location ◽  
Arrival Time ◽  
Low Frequency ◽  
Recognition Algorithm ◽  
Estimation Accuracy ◽  
Test Results ◽  
Accuracy Test ◽  
Image Translation ◽  
Component Image

This paper proposes a new Image-to-Image Translation (Pix2Pix) enabled deep learning method for traveling wave-based fault location. Unlike the previous methods that require a high sampling frequency of the PMU, the proposed method can translate the scale 1 detail component image provided by the low frequency PMU data to higher frequency ones via the Pix2Pix. This allows us to significantly improve the fault location accuracy. Test results via the YOLO v3 object recognition algorithm show that the images generated by pix2pix can be accurately identified. This enables to improve the estimation accuracy of the arrival time of the traveling wave head, leading to better fault location outcomes.

A Novel Fault-location Method for HVDC Transmission Lines Based on Concentric Relaxation Principle and Wavelet Packet

2020 ◽  
Vol 13 (5) ◽  
pp. 705-716
Author(s):  
Congshan Li ◽  
Ping He ◽  
Feng Wang ◽  
Cunxiang Yang ◽  
Yukun Tao ◽  
...  
Keyword(s):  
Transmission Lines ◽  
Traveling Wave ◽  
Fault Location ◽  
Wave Attenuation ◽  
Wavelet Packet ◽  
Low Frequency ◽  
Evaluation Process ◽  
Location Method ◽  
Hvdc Transmission ◽  
Instantaneous Energy

Background: A novel fault location method of HVDC transmission line based on a concentric relaxation principle is proposed in this paper. Methods: Due to the different position of fault, the instantaneous energy measured from rectifier and inverter are different, and the ratio k between them is the relationship to the fault location d. Through the analysis of amplitude-frequency characteristics, we found that the wave attenuation characteristic of low frequency in the traveling wave is stable, and the amplitude of energy is larger, so we get the instantaneous energy ratio by using the low-frequency data. By using the method of wavelet packet decomposition, the voltage traveling wave signal was decomposed. Results: Finally, calculate the value k. By using the data fitting, the relative function of k and d can be got, that is the fault location function. Conclusion: After an exhaustive evaluation process considering different fault locations, fault resistances, and noise on the unipolar DC transmission system, four-machine two-area AC/DC parallel system, and an actual complex grid, the method presented here showed a very accurate and robust behavior.

scholarly journals A Novel Traveling-Wave-Based Method Improved by Unsupervised Learning for Fault Location of Power Cables via Sheath Current Monitoring

Sensors ◽  
2019 ◽  
Vol 19 (9) ◽  
pp. 2083 ◽  
Author(s):  
Mingzhen Li ◽  
Jianming Liu ◽  
Tao Zhu ◽  
Wenjun Zhou ◽  
Chengke Zhou
Keyword(s):  
Unsupervised Learning ◽  
Traveling Wave ◽  
Fault Location ◽  
Arrival Time ◽  
Spatial Clustering ◽  
Power Cables ◽  
Sample Distribution ◽  
Maximum Slope ◽  
The Matrix ◽  
Current Monitoring

In order to improve the practice in maintenance of power cables, this paper proposes a novel traveling-wave-based fault location method improved by unsupervised learning. The improvement mainly lies in the identification of the arrival time of the traveling wave. The proposed approach consists of four steps: (1) The traveling wave associated with the sheath currents of the cables are grouped in a matrix; (2) the use of dimensionality reduction by t-SNE (t-distributed Stochastic Neighbor Embedding) to reconstruct the matrix features in a low dimension; (3) application of the DBSCAN (density-based spatial clustering of applications with noise) clustering to cluster the sample points by the closeness of the sample distribution; (4) the arrival time of the traveling wave can be identified by searching for the maximum slope point of the non-noise cluster with the fewest samples. Simulations and calculations have been carried out for both HV (high voltage) and MV (medium voltage) cables. Results indicate that the arrival time of the traveling wave can be identified for both HV cables and MV cables with/without noise, and the method is suitable with few random time errors of the recorded data. A lab-based experiment was carried out to validate the proposed method and helped to prove the effectiveness of the clustering and the fault location.

scholarly journals An Improved Traveling-Wave-Based Fault Location Method with Compensating the Dispersion Effect of Traveling Wave in Wavelet Domain

2017 ◽  
Vol 2017 ◽  
pp. 1-11 ◽  
Author(s):  
Huibin Jia
Keyword(s):  
Transmission Lines ◽  
Traveling Wave ◽  
Fault Location ◽  
Arrival Time ◽  
Wide Band ◽  
Correction Function ◽  
Wavelet Domain ◽  
Dispersion Effect ◽  
Location Method ◽  
Location Methods

The fault generated transient traveling waves are wide band signals which cover the whole frequency range. When the frequency characteristic of line parameters is considered, different frequency components of traveling wave will have different attenuation values and wave velocities, which is defined as the dispersion effect of traveling wave. Because of the dispersion effect, the rise or fall time of the wavefront becomes longer, which decreases the singularity of traveling wave and makes it difficult to determine the arrival time and velocity of traveling wave. Furthermore, the dispersion effect seriously affects the accuracy and reliability of fault location. In this paper, a novel double-ended fault location method has been proposed with compensating the dispersion effect of traveling wave in wavelet domain. From the propagation theory of traveling wave, a correction function is established within a certain limit band to compensate the dispersion effect of traveling wave. Based on the determined arrival time and velocity of traveling wave, the fault distance can be calculated precisely by utilizing the proposed method. The simulation experiments have been carried out in ATP/EMTP software, and simulation results demonstrate that, compared with the traditional traveling-wave fault location methods, the proposed method can significantly improve the accuracy of fault location. Moreover, the proposed method is insensitive to different fault conditions, and it is adaptive to both transposed and untransposed transmission lines well.

scholarly journals A Novel Fault Location Method for Power Cables Based on an Unsupervised Learning Algorithm

Energies ◽  
2021 ◽  
Vol 14 (4) ◽  
pp. 1164
Author(s):  
Mingzhen Li ◽  
Jialong Bu ◽  
Yupeng Song ◽  
Zhongyi Pu ◽  
Yuli Wang ◽  
...  
Keyword(s):  
Unsupervised Learning ◽  
Traveling Waves ◽  
Traveling Wave ◽  
Computer Aided Design ◽  
Fault Location ◽  
Arrival Time ◽  
Short Circuit ◽  
Power Cable ◽  
Density Level ◽  
Location Method

In order to locate the short-circuit fault in power cable systems accurately and in a timely manner, a novel fault location method based on traveling waves is proposed, which has been improved by unsupervised learning algorithms. There are three main steps of the method: (1) build a matrix of the traveling waves associated with the sheath currents of the cables; (2) cluster the data in the matrix according to its density level and the stability, using Hierarchical Density-Based Spatial Clustering of Applications with Noise (HDBSCAN); (3) search for the characteristic cluster point(s) of the two branch clusters with the smallest density level to identify the arrival time of the traveling wave. The main improvement is that high-dimensional data can be directly used for the clustering, making the method more effective and accurate. A Power System Computer Aided Design (PSCAD) simulation has been carried out for typical power cable circuits. The results indicate that the hierarchical structure of the condensed cluster tree corresponds exactly to the location relationship between the fault point and the monitoring point. The proposed method can be used for the identification of the arrival time of the traveling wave.

scholarly journals ELECTRICAL NETWORK OF FAULT REGIME MODEL CONSTRUCT FEATURES IN A SINGLE-END TRAVELING WAVE FAULT LOCATION METHOD

2021 ◽  
pp. 133-139
Author(s):  
Aleksey O. Fedorov ◽  
Vladimir S. Petrov ◽  
Vitaliy A. Hristoforov
Keyword(s):  
Traveling Wave ◽  
Fault Location ◽  
Arrival Time ◽  
Short Circuit ◽  
Characteristic Impedance ◽  
Line Length ◽  
Power Line ◽  
Electrical Network ◽  
Arrival Times ◽  
Location Method

In the single-end traveling wave (TW) fault location methods, for determining TW front, the arrival time of which is determined by the place of the short circuit (SC) on the power line, electrical network of fault regime models are constructed. From the electrical network of fault regime models, only one is selected that allows, by the first TW front magnitude and its arrival time, to obtain estimations of the TWs fronts magnitudes and their arrival times which are closest to the corresponding quantities determined from locator measurements. Based on the selected electrical network of fault regime model the used TWs are identified and the fault place is determined. Known implementations of the single-end traveling wave fault location method use simplified electrical network of fault regime model: the influence of the fault type and its resistance, as well as the parameters of the electrical network elements, are not taken into account on the TWs fronts magnitude. These disadvantages can cause both an increased error in determining the fault location and even failure in the operation of the locator. In this article, the theory of constructing electrical network of fault regime model is presented: the influence of fault location and its type on the TWs fronts magnitude are considered. Particular attention is paid to the study of the issue of the TWs generation as a result of the cross-transmission effect. It is shown that in order to correctly determine the used TW front in the electrical network of fault regime model, in addition to the power line length and its characteristic impedance, it is necessary to take into account the short circuit type and its resistance, and the possible TWs generation in one mode under the influence of TWs in another one.

scholarly journals Design of Traveling Wave Location Method for Transmission Line Fault Based on EMD-TEO

2020 ◽  
Vol 2 (3) ◽  
pp. 73
Author(s):  
Juan Xia
Keyword(s):  
Transmission Line ◽  
Instantaneous Frequency ◽  
Traveling Wave ◽  
Fault Location ◽  
Arrival Time ◽  
Location Accuracy ◽  
Location Method ◽  
Hilbert Huang Transform ◽  
Mode Decomposition ◽  
Positioning Method

<p>With the rapid development of the social economy and the continuous extension of Internet technology, China’s power grid has entered the ranks of large-scale, high-voltage, and intelligent. The main purpose of the fault location of the transmission line is to eliminate hidden trouble and restore the fault line in time to ensure the safe and stable operation of the power system. With the advent of the smart grid, higher requirements are put forward for fault location accuracy, while the traditional wavelet transform and Hilbert-Huang transform have larger defects.</p><p>Therefore, this paper extensively analyses the generation and characteristics of fault traveling waves in transmission line fault, which proves that the traveling wave location method has higher location accuracy than the fault analysis method. Among them, the two-terminal traveling wave positioning method only uses the arrival time of the initial traveling wave, avoiding the principled defects and locating the dead zone of the single-terminal traveling wave positioning method, so the two-terminal traveling wave positioning method is generally used. The key of the two-terminal traveling wave location method is that it can accurately detect the arrival time of the initial traveling wave head. Although the Hilbert-Huang Transform (HHT) method can be used to detect the arrival time of the initial traveling wave head, the problem of inaccurate detection or failure of the wave head may arise when the instantaneous frequency of the IMF component decomposed by the Hilbert-Huang transform is used because of the mode aliasing in the empirical mode decomposition algorithm. Based on the above analysis, an empirical mode decomposition (EMD) combined with the Teager energy operator(TEO) is proposed for the traveling wave fault location of transmission lines. A large number of simulations prove that the EMD-TEO method in this paper can solve the problem of inaccuracy or failure of the HHT method using instantaneous frequency to detect the arrival time of wave head, and has higher fault location accuracy.</p>

scholarly journals A graphical heuristic for reduction and partitioning of large datasets for scalable supervised training

2019 ◽  
Vol 6 (1) ◽  
Author(s):  
Sumedh Yadav ◽  
Mathis Bode
Keyword(s):  
Prediction Accuracy ◽  
Clustering Algorithm ◽  
Nearest Neighbor ◽  
Large Datasets ◽  
Test Results ◽  
Accuracy Test ◽  
Reasonable Proportion ◽  
Speed Up ◽  
Run Time ◽  
Information Graph

Abstract A scalable graphical method is presented for selecting and partitioning datasets for the training phase of a classification task. For the heuristic, a clustering algorithm is required to get its computation cost in a reasonable proportion to the task itself. This step is succeeded by construction of an information graph of the underlying classification patterns using approximate nearest neighbor methods. The presented method consists of two approaches, one for reducing a given training set, and another for partitioning the selected/reduced set. The heuristic targets large datasets, since the primary goal is a significant reduction in training computation run-time without compromising prediction accuracy. Test results show that both approaches significantly speed-up the training task when compared against that of state-of-the-art shrinking heuristics available in LIBSVM. Furthermore, the approaches closely follow or even outperform in prediction accuracy. A network design is also presented for a partitioning based distributed training formulation. Added speed-up in training run-time is observed when compared to that of serial implementation of the approaches.

Sign in / Sign up

Export Citation Format

Share Document