ANFIS Approach for Locating Faults in Underground Cables (original) (raw)
Related papers
Fault location in underground cables using ANFIS nets and discretewavelet transform
This paper presents an accurate algorithm for locating faults in a medium voltage underground power cable using a combination of Adaptive Network-Based Fuzzy Inference System (ANFIS) and discrete wavelet transform (DWT). The proposed method uses five ANFIS networks and consists of 2 stages, including fault type classification and exact fault location. In the first part, an ANFIS isused to determine the fault type, applying four inputs, i.e., the maximum detailed energy of three phase and zero sequence currents.Other four ANFIS networks are utilized to pinpoint the faults (one for each fault type). Four inputs, i.e., the maximum detailed energy of three phase and zero sequence currents, are used to train the neuro-fuzzy inference systems in order to accurately locate the faults on the cable. The proposed method is evaluated under different fault conditions such as different fault locations, different fault inception angles and different fault resistances.
Development of a Wavelet-ANFIS based fault location system for underground power cables
In the past decade, electricity demand has increased rapidly in metropolitan areas. All over the world, large scale underground power cable installations networks are replacing overhead transmission lines due to environmental concerns in densely populated areas. Underground cable systems are manufactured to have long life with reliability. However, the useful life span of these cables is not infinite. The increase in failure rates and system breakdowns on older underground power cables are now adversely impacting system reliability and many losses involved; therefore it is readily apparent that necessary action has to be taken to manage the consequences of this trend. In this paper, a method that combines wavelets and Neuro-fuzzy technique for fault location and identification is proposed. A 10km, 34.5KV, 50Hz power transmission line model was developed and different faults and locations simulated, and then certain selected features of the wavelet transformed signals were extracted to develop an ANFIS for fault location. Comparison of the ANFIS output values and the actual values show that the percentage error was established to be less than 1%. Thus, it can be concluded that the wavelet-ANFIS technique is accurate enough to be used in identifying and locating underground power line faults.
This paper describes a technique to detect, classify and locate faults on an underground cable system based on the principles of continuous wavelet transform (CWT). Due to the fault in the power system a high frequency current and voltage generates and propagate along the power. These generated signals contain a lot of information and can be used for fault detection and location. The high frequency components generated are extracted using the wavelet technique and analysis of the extracted signals is carried. The MATLAB simulink version 7.6 is used to model the underground cable network and faults at various locations are simulated. The resulting waveforms are subjected through a wavelet transform to extract the required signals for analysis. The results show that the wavelet transform is very effective to extract the transient components from the fault signals and detection and location of faults can be done accurately. In this paper three phase 11KV; 100km long cable is considered for the analysis purpose.
International Journal on Electrical Engineering and Informatics
This article presents an advanced continuous wavelet transform (CWT) based approach for fault detection and localization in distribution systems using the artificial neural network (ANN). In this study, CWT extracts distinct features from the transient signals captured from the bus. The derived features are utilized to train and test appropriate ANN architecture in different stages to detect and localize the faults. The proposed scheme provides an optimum method for classification as well as localization of the various kinds of fault with different source short circuit (SSC) level in different locations. The whole detection and localization process consists of several stages. In the first stage, it detects faulty feeder. The faulty line is identified in the second stage. Finally, in the third stage, fault type and fault location are being calculated from the relaying point. The performance of the proposed CWT-ANN based approach is quite promising as compared to traditionally used algorithms. However, a correlation-based feature selection technique is also implemented to reduce training time and improve accuracy. This algorithm is tested in 11 kV radial Indian distribution network but can be applied in other distribution networks also.
Identification of Fault in a Transmission Line by using Wavelet and Location by Fuzzy
International Journal for Research in Applied Science and Engineering Technology
Fault Location estimation is very important issue in power system engineering in order to quickly clear faults and restore power supply as soon as possible with minimum interruptions. In case if any fault occurs in the system, then it may damage the whole system if it is not rectified quickly. So in order to rectify it we have several methods. We use real time wavelet-Fuzzy combined approach for digital relaying. Wavelet Transform is one the efficient tools for analysing non stationary signals such as transients and have been widely applied to solve numerous problems in power systems. This paper presents Fuzzy logic is employed to incorporate expert evaluation through FIS so as to extract important features through wavelet MRA coefficients for obtaining coherent conclusions regarding fault location. Computer simulation using MATLAB have been conducted.
IEEJ Transactions on Electrical and Electronic Engineering, 2014
Locating the faulty section of a high-impedance fault (HIF) is quite challenging for the underground distribution network of a power system. The complexity of the distribution network, such as branches, nonhomogenous lines, and HIF, contributes to the difficulties in locating the faulty section. In this paper, the shortest distance (SD) technique and a database approach have been proposed to determine the faulty section. A multiresolution analysis based on discrete wavelet transforms is chosen to extract the unique features from voltage signals during the HIF event. The output coefficients from the decomposition process is stored in a database and used as the input data for the SD algorithm. The first, second, and third level of detailed coefficients of the post-disturbance voltage signal were utilized for the identification of the faulty section using the proposed method. A ranking analysis was created to provide a number of possibilities of faulty section. In this paper, a 38-node underground distribution network system in a national grid in Malaysia was modeled using the PSCAD software. The proposed method was able to successfully determine the faulty section
Fault Location on Transmission line using Wavelet Transform and Artificial Neural Network
IOSR Journal of Electrical and Electronics Engineering
Finding and designing new methods for determining type and exact location of faults in power system has been a major subject for power system protection. One of the main capabilities that can improve the efficiency of new protection relays in power systems is fault location. In this paper wavelet transform along with neural networks is used for determiningfault location in Transmission lines. In the present work, theauthors have developed an algorithm for locating eleven types offaults over a 100% of line length. The extraction of features of voltage signals and current signals by wavelet transform and subjecting it to artificial neural network, the fault location is calculated. Thealgorithm has been developed keeping in view the pragmaticshurdle of the multiple estimation which have been successfullytackled.
European Journal of Engineering Research and Science
Unlike low impedance faults, which involve relatively large magnitude of fault currents and are easily detected by conventional over-current protection devices, high impedance faults pose a serious challenge to protection engineers because they can remain on the system without the protective relays being able to detect them. This paper presents an improved method for detection and location of high impedance fault using ANFIS model. The study was conducted on the 33 kV Uyo-Ikot Ekpene power distribution line. The case study power distribution system was modeled using MATLAB software. HIFs were introduced at various locations along the distribution line. The data obtained from the MATLAB/Simulink simulated fault using discrete wavelet transform (DWT) were used to train the ANFIS for the location of HIF points along the distribution system as well as for prediction of the distance of the fault location to the nearest injection substation. The results show that ANFIS model gives 52.5 pe...
Wavelet based detection and location of faults in 400kv, 50km Underground Power Cables
In this paper, a method for detection and location of fault in a power line is presented. This method can be applied for both transmission and distribution power lines of underground cables. The proposed method is capable to determine type of fault the fault location upon its occurrence, based on the data available from the measuring equipment. The algorithm presented uses the fault steady-state data (voltages and currents) and the system parameters, to calculate the fault location. The influence of the fault resistance that depends on the design characteristics of the power line is disregarded. The proposed method has been modeled in mat lab simulink of version 7.8.
2020
Underground cables are being faced with a wide variety of faults due to underground conditions, wear and tear, etc. Diagnosing fault source is difficult and the entire cable should be taken out from the ground to check and fix faults. This paper presents the details of faults and the Wavelet transform-based technique for fault detection, classification in the Underground transmission line. Due to the fault in the power system, a highfrequency current and voltage generate and propagate along with the power. These generated signals contain a lot of information and can be used for fault detection and location. The high-frequency components generated are extracted using the wavelet technique and analysis of the extracted signals is carried. The MATLAB Simulink is used to model the underground cable network and faults at various locations are simulated. The resulting waveforms are subjected through a wavelet transform to extract the required signals for analysis The presented model is user friendly and can be used as a common platform for both control and power system engineers. The accuracy of the wavelet transform scheme is 100%.