Electrical power transmission systems suffer from unexpected failures due to various random causes. Un-predicted faults that occur in power systems are required to prevent from propagation to other area in the protective system. The functions of the protective systems are to detect, then classify and finally determine the location of the faulty. This paper presents some techniques that helps to find, determine and diagnosing faults in transmission line. Artificial neural networks, impedance measurement based methods, fuzzy expert method, wavelet transform and so on have been used to achieve fault identification and classification.This paper will review the type of fault that possibly occurs in an electric power system, the type of fault detection and location technique that are available together with the protection device that can be utilized in the power system to protect the equipment from electric fault.
Fault Detection, Protection and Location on Transmission
Line - A Review
ABSTRACT - Electrical power transmission systems suffer from unexpected failures due to various random causes. Un-predicted faults that occur in power systems are required to prevent from propagation to other area in the protective system. The functions of the protective systems are to detect, then classify and finally determine the location of the faulty. This paper presents some techniques that helps to find, determine and diagnosing faults in transmission line. Artificial neural networks, impedance measurement based methods, fuzzy expert method, wavelet transform and so on have been used to achieve fault identification and classification.This paper will review the type of fault that possibly occurs in an electric power system, the type of fault detection and location technique that are available together with the protection device that can be utilized in the power system to protect the equipment from electric fault.
Keywords- Artificial neural networks, Fault identification and classification, Fuzzy expert method, Impedance measurement based methods, Transmission systems, Wavelet transform, Protection device
I. INTODUCTION
An overhead transmission line is one of the main components in every electric power system. The transmission line is exposed to the environment and the possibility of experiencing faults. Those are single line-ground, line-line, double line-ground and three phase faults.to detect those faults many authors developed different techniques..M. Sanaye-Pasand. et al 1 was presented use of neuro computing technology and implementation and improve the performance of conventional algorithms. Nan Zhang 2 was presented Fuzzy K-NN decision rule combined with Adaptive Resonance Theory (ART) based neural network (NN) algorithm for fault detection and classification on transmission lines. A. Jain .et al 3 based on application of artificial neural network (ANN) An adaptive protection scheme is proposed to detect and classify the faults in the double circuit transmission line with double end infeed. A. Yadav. et al 4 used artificial neural networks for protection of doubly fed transmission lines, accurate fault distance and direction estimation by using only one terminal data. S. S. D.
Sahel.et al 5 to simulate distance protection relay FNN (feed-forward neural network) trained by PSO algorithm is proposed which is tested under different fault conditions such as different fault locations, different fault inception angles and different fault resistances.S. Kesharwani. et al 6 detects fault by using artificial neural network but they are considers only single line ground fault. Azriyenn.et al 7 was reduced the outcome of fault and determining the accurate zone setting of distance relay based on the data of the training set from ANFIS which is integration of Fuzzy If-then-rules into Neural Network construction using appropriate learning. V Gomathy.et al 8 presented an effective methodology for fault detection and classification with the optimization techniques such as EPSO with RVM, EPSO with Fuzzy, CSO with RVM and CSO with Fuzzy. D.
Kumar. et al 9 ANFIS function was used for fault analysis and diagnosis in transmission line by Digital image processing wavelet shrinkage function. Ming-Yuan Cho.etal 10 a particle swarm optimization (PSO) based support vector machine (SVM) classifier has been proposed to solve electrical faults in radial distribution systems. B, Bhattacharya. et al 11 proposed machine learning and deep learning techniques for forecasted the maximum voltage deviation using pre-fault data, classified the type of faults and found out the location of the fault. Priyanka. et al 12 used voltages and three phase currents as inputs artificial neural networks for the detection and classification of faults on a three phase transmission lines system. Neke Jude.et al 13 used Artificial Intelligence techiniques such as Expert System Techniques (XPS), Artificial Neural Networks (ANN), and Fuzzy Logic Systems (FLS) for fault identification and detection in transmission line.V. Malviya.et al 14 proposed fault detection technique in a transmission line using artificial neural network used for single phase to ground faults, double phase faults and double phase to ground faults.S. Heo.et al 15 showed that the data augmentation in artificial neural network can be a key to increase the fault detection accuracy further, and it also turned out to be beneficial for the fault classification case. KAluder S.et al 16 propose a fuzzy expert offline system for fault diagnosis and include non-electrical data.17 identifyied electricity theft using various sensors and detected issues are communicated through SMS to the concerned persons.This paper summarizes different authors wok related to fault detection, fault identification, and distance location in transmission line.
II. OBJECTIVE OF THE REVIEW
The aim of this paper is to put all studies together under single heading for comparison & better understanding of fault type identification, fault location and detection techniques in transmission line and identifying the gap on this area.
III. MOTIVATION OF THE REVIEW
Due to various random causes there may be unexpected failure on power transmission line since fault is unavoidable. The motivation behind this paper is Fault detection and diagnosis has been an active research topic still and the problem is not solved yet.
IV. OVERVIEW OF TRANSMISSION LINE FAULTS
Transmission line fault could not be avoided in an electrical power system, some protection devices are needed to protect the expensive equipment in electric power systems before the fault occurrence. But after the fault occurs once on the system we must determine fault locations and types for maintaining the system. Because of those reason, we will design smart devices and techniques in our system 28.
A. Causes transmission line faults
The most common causes of faults in overhead lines are 78:
- Aircraft and cars hitting lines and structures
- Birds and animals
- Contaminated insulators
- Ice and snow loading
- Lighting
- Partial discharges (corona) not controlled
- Punctured or broken insulators
- Trees
- Wind and so on.
B. Types of transmission line Faults
Faults can be classified by series and shunt faults. Series faults represent open conductor and take place when unbalanced series impedance conditions of the lines are present. These faults disturb the symmetry in one or two phases and are therefore unbalanced faults. Series faults are characterized by increase of voltage and frequency and fall in current in the faulted phases. And shunt faults (short circuit faults) are mainly classified as symmetrical and unsymmetrical faults as shown in Fig. 1: an asymmetric or unbalanced fault which does not affect each of the three phases. And symmetric or balanced are affects each of the three phases equally, 5% of transmission faults are symmetric29. Different fault types percentage occurrence is shown in Fig 2:
Abbildung in dieser Leseprobe nicht enthalten
Fig 1: Classification of fault types
Abbildung in dieser Leseprobe nicht enthalten
Fig 2. Percentage occurrence of different fault types
V. FAULT DETECTION TECHNIQUES OF TRANSMISSION LINE
Fault detection techniques of transmission system is classified in to:
1. Fault classification techniques - Methods that determine the fault type
2. Fault Location Techniques - Methods that calculate the distance of the fault
Both techniques play a vital role in development of protection mechanisms for a given power system model. There is a sub category listed in Fig 3. and some of them are described below:
- Support vector machine (SVM): the main idea of SVM classifiers is to find an optimal hyper plane that maximizes the margin between two groups of examples. The advantages of SVM made it a powerful tool for fault classification in transmission lines and distribution systems21.
- PMU(phasor measurement unit) Technique: A fault location method for two-terminal multi section composite transmission lines, which combine overhead lines with underground power cables, using synchronized phasor measurements acquired by global positioning system (GPS) based phasor measurement units (PMUs) or digital relays with embedded PMU or by fault-on relay data synchronization algorithms24.
- Artificial intelligent method: is a method to detect and categorize the faults on the transmission lines proved competitive with respect to optimization. This method employs the phase voltage and phase currents (with respect to their pre-fault values) as its input 25.
- Wavelet Technique: A protection scheme based on transient directional principle was anticipated for transmission line protection. The wavelet transform was used to extract the traveling fronts of the current transients deriving from the faults 26.
Abbildung in dieser Leseprobe nicht enthalten
Fig 3. Fault Detection Techniques of Transmission line and related works
VI. FAULT PROTECTION DEVICES IN TRANSMISSION SYSTEM
There are different types of protection device used in transmission system as shown in Fig 4. The purpose of those protective device is to identify the abnormal signals representing faults on a power transmission system.
Abbildung in dieser Leseprobe nicht enthalten
Fig 4. Protection devices of transmission line and related works
VII. OBSERVATIONS
- Artificial neural network is suitable for high resistance fault and varying fault inception angle, remote end faults are not affect the method4. Its scope is wide enough and can be explored more12.
- Feed-forward neural network better to indicate whether the fault is located inside or outside the protection zones and gives a good estimation of fault location 5.
- The fault can be detected in transmission line based on the classification and optimization techniques 8.
- Come up with some health metrics which can, from the close monitoring of grid states, determine whether the network is going to a vulnerable state or not is challenging 11.
- Back Propagation networks have been chosen for all the three steps in the fault location process namely fault detection, classification and fault location 13.
- Some papers have proposed a novel approach that is combination of SVM and wavelet techniques.This is used to detect and classify the types of the faults 21 .
- A new method for faulty region detection and classification for thyristor controlled series compensator (TCSC) and unified power flow controller (UPFC) line using decision tree(DT) 22.
VIII. CONCLUSION AND RECOMMENDATION
A. Conclusion
The fault of transmission line controls by using current and voltage sensing units, microcontroller, GSM (global system for mobile communication) module & different protective equipment’s. From the general machine learning methods, Artificial neural networks are a reliable and effective method for an electrical power system transmission line fault classification and detection especially in view of the increasing dynamic connectivity of the modern electrical power transmission systems. For detecting and classifying transmission line faults using cross-correlation and k-Nearest Neighbor (k-NN). This method computes the cross correlation between pure and faulty current signals. Extracted features are used as input the k-NN algorithm which then computes distance of a given sample to all other samples in the set and class of the sample with least distance is predicted.
B. Recommendations
- For future woks recommended to improve the size of the fault dictionary.
- The effects of data augmentation need to be investigated further.
[...]
- Citar trabajo
- Seada Hussen Adem (Autor), 2020, Fault Detection, Protection and Location on Transmission Line. A Review, Múnich, GRIN Verlag, https://www.grin.com/document/936637