Best ANN structures for fault location in single-and double-circuit transmission lines

The great development in computing power has allowed the implementation of artificial neural networks (ANNs) in the most diverse fields of technology. This paper shows how diverse ANN structures can be applied to the processes of fault classification and fault location in overhead two-terminal transmission lines, with single and double circuit. The existence of a large group of valid ANN structures guarantees the applicability of ANNs in the fault classification and location processes. The selection of the best ANN structures for each process has been carried out by means of a software tool called SARENEUR.     

A neural network-based scheme for fault diagnosis of power transformers

This research presents an artificial neural network (ANN)-based scheme for fault diagnosis of power transformers. The scheme is designed to detect the fault, estimate the faulted side, classify the fault type and identify the faulted phase.The proposed fault diagnosis scheme (FDS) consists of three hierarchical levels. In the first level, a pre-processing of input data is performed. In the second level, there is an ANN which is designed to detect the fault and determine the faulted side if any. In the third level, there are two sides diagnosis systems. Each system is dedicated to one side and consists of one ANN in series with four paralleled ANNs (for fault type classification).The proposed FDS is trained and tested using local measurements of three-phase primary voltage and primary and secondary currents. These samples are generated using EMTP simulation of the High Dam 15.75/500 kV transformer substation in Upper Egypt. All the possible fault types were simulated. The fault locations and fault incipience time were varied within each fault type. Testing results proved that the performance of the proposed ANN-based FDS is satisfactory.     

Transient stability assessment using artificial neural networks

This paper presents a study of the feasibility of using artificial neural networks (ANNs) in transient stability assessment for power systems. In the study ANNs have been developed to synthesize the complex mapping that carries the power system operating variables and fault locations into the Critical Clearing Times. The training of the ANNs was achieved through the method of backpropagation. The critical fault clearing time values were obtained by the Extended Equal Area Criterion method and used for training. In this work, an attempt was made to avoid the restrictions on load and topology variations. The parameters of the ANNs consist of the generation and loading levels. None of these inputs require any computation. This feature is desirable for on-line transient stability assessment purposes.Training of the ANNs was achieved using a combined production learning phase. The training patterns were not limited to a given collection of samples. This scheme eliminates the problem that an ANN may be influenced by the regions of attraction of a specific category.     

A new fault location method for distribution networks using sparse measurements

In response to the growing demand to improve reliability and quality of power supply, distributed monitoring devices are gradually being implemented in distribution networks. On the other hand, utilities are demanding more accurate and reliable fault location systems to reduce the economic impact of power outages. This paper presents a novel method that takes full advantage of all available measurements to provide an accurate fault location. The developed method first uses an iterative state estimation based algorithm to find the nearest node to the fault location. It then examines all lines connected to the selected node and locates the fault. The performance of the proposed method is studied by simulation tests on a real 13.8 kV, 134-node distribution system under different fault scenarios. The results verify the accuracy of the algorithm and its robustness even under uncertain measured data. The method robustly handles measurement errors, and is applicable to any distribution network with laterals, load taps and heterogeneous lines.     

A classification approach for power distribution systems fault cause identification

Power distribution systems play an important role in modern society. When distribution system outages occur, fast and proper restorations are crucial to improve the quality of services and customer satisfaction. Proper usages of outage root cause identification tools are often essential for effective outage restorations. This paper reports on the investigation and results of two popular classification methods: logistic regression (LR) and artificial neural network (ANN) applied on power distribution fault cause identification. LR is seldom used in power distribution fault diagnosis, while ANN has been extensively used in power system reliability researches. This paper discusses the practical application problems, including data insufficiency, imbalanced data constitution, and threshold setting that are often faced in power distribution fault cause identification problems. Two major distribution fault types, tree and animal contact, are used to illustrate the characteristics and effectiveness of the investigated techniques.     

Single-ended time domain fault location based on transient signal measurements of transmission lines

Precise fault location plays an important role in the reliability of modern power systems. With the increasing penetration of renewable energy sources, the power system experiences a decrease in system inertia and alterations in steady-state characteristics following a fault occurrence. Most existing single-ended phasor domain methods assume a certain impedance of the remote-end system or consistent current phases at both ends. These problems present challenges to the applicability of conventional phasor-domain location methods. This paper presents a novel single-ended time domain fault location method for single-phase-to-ground faults, one which fully considers the distributed parameters of the line model. The fitting of transient signals in the time domain is realized to extract the instantaneous amplitude and phase. Then, to eliminate the error caused by assumptions of lumped series resistance in the Bergeron model, an improved numerical derivation is presented for the distributed parameter line model. The instantaneous symmetrical components are extracted for decoupling and inverse transformation of three-phase recording data. Based on the above, the equation of instantaneous phase constraint is established to effectively identify the fault location. The proposed location method reduces the negative effects of fault resistance and the uncertainty of remote end parameters when relying on one-terminal data for localization. Additionally, the proposed fault analysis methods have the ability to adapt to transient processes in power systems. Through comparisons with existing methods in three different systems, the fault position is correctly identified within an error of 1%. Also, the results are not affected by sampling rates, data windows, fault inception angles, and load conditions.     

GA trained parallel hidden layered ANN based differential protection of three phase power transformer

Transformer protection is an established area of research to find the fastest and efficient differential relay algorithm that isolates the transformer from remaining system causing least damage. Algorithm should also avoid mal-operation when differentiating between the operating conditions. Various differential algorithms were proposed in the past, allowing a scope for further research. In this paper, ANN is used as a pattern classifier which discriminates among normal, magnetizing inrush, over-excitation and internal fault currents in a power transformer. The proposed scheme has been realized through different ANN architectures including a new customized parallel-hidden layered design, which originates to be more accurate in differentiating between the normal wave and faulty wave despite the shape similarity. A combination of two ANNs in Master–Slave mode has also been discussed. Back Propagation (BP) and Genetic Algorithm (GA) are used to train the multi-layered feed forward neural network and their simulated results are compared. The neural network trained by GA gives more accurate results (in terms of mean square error) than by BP Algorithm. Simulated data are used as an input to the ANN to verify the accuracy of the algorithm. Thus, GA trained Master–Slave ANN based differential protection scheme provides faster, accurate, more secured and dependable relay for power transformers.     

Development of an expert system for analyzing faults in power systems based on waveform recognition by artificial neural networks

In restoration of a power system, appropriate steps need to be taken in early stages to avoid damage to the public caused by insufficient power supply. The first step in the restoration is the analysis of the fault that has produced the power failure. Today, a large amount of data will be available at the occurrence of a fault due to the use of advanced communication systems using digital relays and optical fibers. Such systems are intended to obtain data from the relays and circuit breakers (CBs) under operation and the voltage and current during operation and the prefault and postfault periods. This paper presents a hybrid fault analyzing expert system using data based on protective devices and phasor measurement. This system is a combination of an artificial neural network (ANN) and an expert system (ES). The ANN is used to simplify the analysis of current and voltage waveforms, which is considered difficult due to the presence of a fault resistance and the line impedance. Since a fault waveforms contains various ambiguous factors depending on the nature of the fault resistance, load current and other system parameters, the use of ANN is effective making such determinations.     

基于最优光伏阵列重构的电网调频策略研究

为实现光伏电站功率输出最大化以及功率输出与调频(FR)信号之间的功率偏差最小化,提出了一种新型光伏电站最优阵列重构(OAR)模型.为快速获取最优Pareto前沿,采用了一种寻优性能高效的多目标黑猩猩优化器(MOBO).采用了一种多准则妥协解排序法(VIKOR)的决策方法,从所获取的Pareto前沿中确定最佳折衷解.为验...     

基于故障分量的配电网单端量测距方法

为了消除负荷电流对配电网单相接地故障测距精度的影响,提出了基于故障分量的单端量测距方法。该方法根据横向故障电流的特征构造了测距函数,利用线路单端电压、电流的故障分量来计算故障距离。大量的ATP-EMTP仿真结果表明该方法不受负荷电流的影响,可以有效地搜索到配电网多分支线路的故障点或故障范围。     

A differential relay for power transformers using intelligent tools

This paper presents alternative approaches using artificial neural networks (ANNs) for the protection of power transformers. A complete protection scheme was implemented. An ANN subroutine was used to discriminate internal faults from other situations, replacing the traditional Fourier method for harmonic restraint. In addition, a routine for reconstruction of saturated current signals based on recurrent ANNs is also proposed. The proposed methods were extensively tested and then compared to the traditional differential protection algorithm, showing promising results. The application of the ANN tools is a new and important stage in the differential relay analysis methodology for power transformer protection.     

Optimal design of adaptive under frequency load shedding using artificial neural networks in isolated power system

The frequency and voltage stability is a basic principle in the power system operation. Different short circuits, load growth, generation shortage, and other faults which disturb the voltage and frequency stability are serious threats to the system security. The frequency and voltage instability causes dispersal of a power system into sub-systems, and leads to blackout as well as heavy damages of the system equipments. Optimum load shedding during contingency situations is one of the most important issues in power system security analysis. This paper presents a fast and optimal adaptive load shedding method, for isolated power system using Artificial Neural Networks (ANNs). By creating an appropriate data-base of contingencies for training the neural network, the proposed method is able to perform correct load shedding in various loading scenario. In this regard, the total power generation, the total loads in power system, the existing spinning reserved capacity value in the network and frequency reduction rate were selected as the ANN inputs. This method has been tested on the New-England power system. The simulation results show that the proposed algorithm is very fast, robust and optimal values of load shedding in different loading scenarios, related to conventional method.     

GPS travelling wave fault locator systems: investigation into theanomalous measurements related to lightning strikes

A new fault location system based on the travelling wave principle and capable of locating faults on power lines to within ±one tower span (300 m) has been successfully developed and applied to BC Hydro's extensive 500 kV network. Unlike earlier schemes which are based on impedance measurements, its accuracy is not affected by load conditions, high grounding resistance and most notably series capacitor banks. This system measures the time of arrival of a fault-generated travelling wave at the line terminals using the precise timing signals from the Global Positioning System (GPS). Operating experience with the fault locator on lightning related faults indicated highly accurate results were obtained for the majority of the cases. In a few of the lightning-caused disturbances, the system gave anomalous measurements. This paper describes the operation of the system, summarizes the operating experience and explains the observed anomalous measurements     

Accurate fault locator for EHV transmission lines based on radial basis function neural networks

This paper describes the design and implementation of an artificial neural networks-based fault locator for extra high voltage (EHV) transmission lines. This locator utilizes faulted voltage and current waveforms at one end of the line only. The radial basis function (RBF) networks are trained with data under a variety of fault conditions and used for fault type classification and fault location on the transmission line. The results obtained from testing of RBF networks with simulated fault data and recorded data from a 400 kV system clearly show that this technique is highly robust and very accurate. The technique takes into account all the practical limitations associated with a real system. Thereby making it possible to effectively implement an artificial intelligence (AI) based fault locator on a real system.     

基于神经网络建模的零序直流原理测距研究

针对电力电缆铺设隐蔽,发生故障时不易准确定位故障点的问题,研究了零序直流保护原理和人工神经网络技术在小电流接地系统单相接地故障测距中的应用。利用直流信号不受电网电容影响的特性,建立检测源与接地网电路模型。根据电路模型,建立其数学模型,由于模型中的一些参数复杂,则考虑采用神经网络对其未知量进行辨识,并将辨识好的数学模型应用于故障测距中。根据实验数据,对两种模型进行分析,结果表明利用神经网络辨识参数的方法能够准确、可靠地实现故障测距。     

On-line evaluation of capacity and energy losses in powertransmission systems by using artificial neural networks

An adaptive loss evaluation algorithm for power transmission systems is proposed in this paper. The algorithm is based on training of artificial neural networks (ANNs) using backpropagation. Due to the capability of parallel information processing of the ANNs, the proposed method is fast and yet accurate. Active and reactive powers of generators and loads, as well as the magnitudes of voltages at voltage-controlled buses are chosen as inputs to the ANN. System losses are chosen as the outputs. Training data are obtained by load flow studies, assuming that the state variables of the power system to be studied take the values uniformly distributed in the ranges of their lower and upper limits. Load flow studies for different system topologies are carried out and the results are compiled to form the training set. Numerical results are presented in the paper to demonstrate the effectiveness of the proposed algorithm in terms of accuracy and speed. It is concluded that the trained ANN can be utilized for both off-line simulation studies and on-line calculation of demand and energy losses. High performance has been achieved through complex mappings, modeled by the ANN, between system losses and system topologies, operating conditions and load variations     

Neural network load forecasting with weather ensemble predictions

In recent years, a large amount of literature has evolved on the use of artificial neural networks (ANNs) for electric load forecasting. ANNs are particularly appealing because of their ability to model an unspecified nonlinear relationship between load and weather variables. Weather forecasts are a key input when the ANN is used for forecasting. This paper investigates the use of weather ensemble predictions in the application of ANNs to load forecasting for lead times from one to ten days ahead. A weather ensemble prediction consists of multiple scenarios for a weather variable. We use these scenarios to produce multiple scenarios for load. The results show that the average of the load scenarios is a more accurate load forecast than that produced using traditional weather forecasts. We use the load scenarios to estimate the uncertainty in the ANN load forecast. This compares favorably with estimates based solely on historical load forecast errors.     

Digital fault location for high resistance grounded transmissionlines

This paper demonstrates that the presence of multiple load taps cannot be neglected for single-phase-to-ground fault location. A new method has been developed taking this into consideration, that can be applied to correct the location error due to intermediate power sources. Then fault location methods for parallel double-circuit two-terminal transmission lines are discussed. Finally, a new fault location method is proposed for high-resistance grounded double-circuit transmission lines with three terminals     

建筑配电系统故障电弧的仿真与检测

选择合适的电弧模型,对建筑配电系统发生的故障电弧进行了仿真.基于小波的时一频分析特点和人工神经网络(ANN)的学习能力,提出了一种分辨故障电弧和正常负荷电流的方法.该方法通过小波变换对信号进行多分辨率分析,提取信号的特征矢量,利用人工神经网络对输入特征矢量进行故障识别.仿真实验的结果表明,该方法具有良好的故障识别件能.     

Artificial neural networks in power system restoration

Power system restoration (PSR) has been a subject of study for many years. Many techniques were proposed to solve the limitations of the predetermined restoration guidelines and procedures used by a majority of system operators to restore a system following the occurrence of a wide area disturbance. This paper discusses limitations encountered in some currently used PSR techniques and a proposed improvement based on artificial neural networks (ANNs). The proposed scheme is tested on a 162-bus transmission system and compared with a breadth-search restoration scheme. The results indicate that the use of ANN in power system restoration is a feasible option that should be considered for real-time applications.