Identification of Type of a Fault in Distribution System Using Shallow Neural Network with Distributed Generation

A distributed generation system(DG)has several benefits over a traditional centralized power system.However,the protection area in the case of the distributed generator requires special attention as it encounters stability loss,failure re-closure,fluctuations in voltage,etc.And thereby,it demands immediate attention in identifying the location&type of a fault without delay especially when occurred in a small,distributed generation system,as it would adversely affect the overall system and its operation.In the past,several methods were proposed for classification and localisation of a fault in a distributed generation system.Many of those methods were accurate in identifying location,but the accuracy in identifying the type of fault was not up to the acceptable mark.The proposed work here uses a shallow artificial neural network(sANN)model for identifying a particular type of fault that could happen in a specific distribution network when used in conjunction with distributed generators.Firstly,a distribution network consisting of two similar distributed generators(DG1 and DG2),one grid,and a 100 Km distribution line is modeled.Thereafter,different voltages and currents corresponding to various faults(line to line,line to ground)at different locations are tabulated,resulting in a matrix of 500×18 inputs.Secondly,the sANN is formulated for identifying the types of faults in the system in which the above-obtained data is used to train,validate,and test the neural network.The overall result shows an unprecedented almost zero percent error in identifying the type of the faults.

Review of Artificial Intelligence for Oil and Gas Exploration: Convolutional Neural Network Approaches and the U-Net 3D Model

Deep learning, especially through convolutional neural networks (CNN) such as the U-Net 3D model, has revolutionized fault identification from seismic data, representing a significant leap over traditional methods. Our review traces the evolution of CNN, emphasizing the adaptation and capabilities of the U-Net 3D model in automating seismic fault delineation with unprecedented accuracy. We find: 1) The transition from basic neural networks to sophisticated CNN has enabled remarkable advancements in image recognition, which are directly applicable to analyzing seismic data. The U-Net 3D model, with its innovative architecture, exemplifies this progress by providing a method for detailed and accurate fault detection with reduced manual interpretation bias. 2) The U-Net 3D model has demonstrated its superiority over traditional fault identification methods in several key areas: it has enhanced interpretation accuracy, increased operational efficiency, and reduced the subjectivity of manual methods. 3) Despite these achievements, challenges such as the need for effective data preprocessing, acquisition of high-quality annotated datasets, and achieving model generalization across different geological conditions remain. Future research should therefore focus on developing more complex network architectures and innovative training strategies to refine fault identification performance further. Our findings confirm the transformative potential of deep learning, particularly CNN like the U-Net 3D model, in geosciences, advocating for its broader integration to revolutionize geological exploration and seismic analysis.

Intelligent identification method and application of seismic faults based on a balanced classification network

This study combined fault identification with a deep learning algorithm and applied a convolutional neural network(CNN)design based on an improved balanced crossentropy(BCE)loss function to address the low accuracy in the intelligent identification of seismic faults and the slow training speed of convolutional neural networks caused by unbalanced training sample sets.The network structure and optimal hyperparameters were determined by extracting feature maps layer by layer and by analyzing the results of seismic feature extraction.The BCE loss function was used to add the parameter which is the ratio of nonfaults to the total sample sets,thereby changing the loss function to find the reference of the minimum weight parameter and adjusting the ratio of fault to nonfault data.The method overcame the unbalanced number of sample sets and improved the iteration speed.After a brief training,the accuracy could reach more than 95%,and gradient descent was evident.The proposed method was applied to fault identification in an oilfield area.The trained model can predict faults clearly,and the prediction results are basically consistent with an actual case,verifying the effectiveness and adaptability of the method.

Wear Fault Diagnosis of Machinery Based on Neural Networks and Gray Relationships

In this paper, the regular characteristic of -wear particles related to fault type of machines based on condition monitoring of reciprocal machinery is discussed. The typical -wear particles spectrum is established according to the equipment structure , friction and wear rule and the characteristic of 'wear particles; The identification technology of wear particles is proposed based on neural networks and a gray relationship ; an intelligent wear particles identification system is designed. The diagnosis example shows that this system can promote the accuracy and the speed of wear particles identification.

Fault Identification and Health Monitoring of Gas Turbine Engines Using Hybrid Machine Learning-based Strategies

Ahealth monitoring scheme is developed in this work by using hybrid machine learning strategies to iden-tify the fault severity and assess the health status of the aircraft gas turbine engine that is subject to component degrada-tions that are caused by fouling and erosion.The proposed hybrid framework involves integrating both supervised recur-rent neural networks and unsupervised self-organizing maps methodologies,where the former is developed to extract ef-fective features that can be associated with the engine health condition and the latter is constructed for fault severity modeling and tracking of each considered degradation mode.Advantages of our proposed methodology are that it ac-complishes fault identification and health monitoring objectives by only discovering inherent health information that are available in the system I/O data at each operating point.The effectiveness of our approach is validated and justified with engine data under various degradation modes in compressors and turbines.

APPLICATION OF NEURAL NETWORK FOR THE TRANSFORMER PROTECTIVE RELAYS

A neural network method used to identify the different operating states of transformers has been proposed and established.It is superior to the traditional transformer protective principles and can correctly identify,within half cycle from the fault inception,the internal faults,magnetizing inrush current state,external faults and switching on the internal faults of a no load transformer.In addition,this method has broad availability and high fault tolerant ability.A lot of simulations have demonstrated its superiority.

融合Residual Network-50残差块与卷积注意力模块的地震断层自动识别

传统的断层识别是由地质解释人员以人工标记的方式进行检测,不仅耗时长、效率低,且识别结果存在一定的人为误差。为解决以上问题,提高断层识别的精度,提出了一种基于深度学习的断层识别方法,利用注意力机制聚焦目标特征的能力,在U-Net网络的解码层引入了卷积注意力模块(Convolutional Block Attention Module,CBAM),在编码层引入了ResNet-50残差块,建立基于卷积神经网络(Convolutional Neural Networks,CNN)的断层识别方法(Res-CBAM-UNet)。将合成地震数据与相应的断层标签进行数据增强操作,新生成的训练数据集作为输入对网络模型进行训练,以提高模型的泛化能力。随后将该模型与CBAM-UNet、ResNet34-UNet和ResNet50-UNet网络进行对比分析,利用实际工区地震数据进行测试。结果表明,设计的Res-CBAM-UNet网络对断层具有较好的识别效果,且识别出的断层连续性好,计算效率高。

基于SVD-CWT和CNN的水轮发电机转子故障识别

水轮发电机转子振动故障识别是水电站运维的重难点问题,为此提出一种基于转子振动信号的故障识别方法。首先针对发电机转子的非平稳和非线性振动信号,采用奇异值分解(SVD)并结合能量差分谱理论进行降噪预处理;对预处理数据使用连续小波变换(CWT)转换为时频图并形成图像数据集;然后将该图像数据集作为卷积神经网络(CNN)输入,通过CNN多层池化及卷积形成分布式故障特征表达,最终实现发电机转子故障模式识别和分类。经实验验证,该方法准确率达到99.5%以上,能有效识别出发电机转子的故障类型。

Welch功率谱与卷积神经网络结合的滚动轴承故障诊断

针对滚动轴承故障诊断在小训练样本下和强噪声下无法取得高精度识别的问题,提出一种基于Welch功率谱结合卷积神经网络进行诊断的方法。该方法以原始时域振动信号作为输入,用Welch功率谱转换数据形态同时对高强度噪声进行抑制,再用得到的功率谱训练卷积神经网络,最后将训练好的模型用于轴承的故障诊断。与WDCNN[1]等方法进行对比,实验发现在混合负载下,该方法平均识别率正确达到99%,其它方法达到这个精度至少需要20倍以上的训练样本量,明显优于WDCNN等方法。抗噪实验结果表明噪声对信号的干扰越强烈,该方法的抗噪表现越好,其抗噪性能要显著优于WDCNN等方法。

基于声振融合的二次EWT-CNN刀具磨损监测

为了实现加工过程中对刀具磨损状态的监测,提出一种基于协同过滤融合的方法。首先,对工作刀具振动信号和声音信号进行特征相关性分析后进行数据层融合;然后,将得到的声振融合信号进行二次经验小波变换(EWT)后去噪重构;最后,将重构信号进行信号增强并送入CNN实现特征提取及刀具故障识别。通过对不同故障类型的麻花钻头进行故障识别实验,在声音、振动以及声振融合信号和不同信号去噪重构方法的对比下,该方法对不同故障类型的钻头作出了98.96%的高识别率。验证了所提方法在刀具故障识别方面的优越性。

基于图半监督与多任务学习的配电网故障区段与类型统一辨识

为解决深度学习类配电网故障辨识方法在量测不足和标记率低时准确率不高的问题,提出了基于图半监督与多任务学习的故障区段与类型统一辨识方法。首先,设计了故障区段与类型统一辨识的图神经网络架构,在图嵌入层中融入网络拓扑和线路参数信息,以充分挖掘不同位置、类型的故障特征。其次,采用多任务注意力网络构建了故障区段定位和类型辨识两个任务,以提取故障的多重信息,实现不同任务间知识转移。再次,将图嵌入特征与无标签样本的编码压缩特征进行融合,得到新的多任务共享特征,以充分利用未标记数据,增强模型泛化能力。最后,通过算例测试表明,所提方法的故障辨识精度优于传统神经网络,且在实时量测少、标签率低及不同量测噪声条件下具有更好的鲁棒性。

基于BS-1DCNN的海缆振动信号识别

光纤振动信号是非线性的,传统的非线性振动信号识别方法通常需要信号分析和特征选择,既耗时又复杂。本文提出一种光纤振动信号识别新方法,可以直接提取特征,对原始信号进行分类,简化识别过程。本方法用支持向量机代替Softmax分类器,优化一维卷积神经网络(one-dimensional convolution neural network,1DCNN),以提高1DCNN结果在小样本条件下的稳定性。采用鸟群算法(bird swarm algorithm,BSA)对支持向量机(support vector machine,SVM)参数进行了优化,有效地提高识别精度。将本文提出的BS-1DCNN方法与1DCNN、VMD-GA-SVM、VMD-PSO-SVM、VMD-BSA-SVM共4种方法进行比较,结果表明,BS-1DCNN在识别准确率和测试时间方面性能表现良好。该算法能有效提高海缆振动信号识别率,且在不同样本比例下均能达到较好的识别效果。

基于CNN-BiGRU的高压直流输电线路故障识别

针对高压直流(high voltage direct current,HVDC)输电线路故障暂态行波具有时序性和强非线性的特点,导致高过渡电阻情况下故障识别率低的问题,提出基于卷积神经网络(convolutional neural networks,CNN)和双向循环门单元(bidirectional gate recurrent unit,BiGRU)的HVDC输电线路故障识别方法。首先,采用故障后整流侧的双极暂态电流行波作为特征向量,利用CNN提取全局特征,并从中剔除噪声和不稳定成分,完成对数据的降维处理。然后,采用BiGRU来捕获CNN提取到特征的前后时间信息,进一步提取数据中的时序特征,以实现HVDC输电线路故障识别。仿真结果表明:该方法可在不同故障地点以及不同过渡电阻下对单极接地、双极短路、雷击故障、雷击干扰共四种故障实现准确识别,可靠性高,具有较强的耐受过渡电阻能力,同时具备一定的抗噪性能。

基于长短时记忆神经网络的励磁涌流与故障电流识别方法

变压器空载合闸时产生励磁涌流导致差动保护误动作的问题至今仍未能被完全解决.针对该问题,提出一种利用长短时记忆(LSTM)神经网络识别励磁涌流与故障电流的方法.首先,在PSCAD软件平台上搭建变压器空载合闸及内部故障仿真模型,通过仿真产生大量三相电流瞬时采样数据作为训练神经网络的样本集;然后,利用Keras平台搭建LSTM神经网络模型并完成训练;最后,利用新的仿真数据和现场故障录波数据对训练好的LSTM神经网络进行测试.结果表明LSTM神经网络可以快速准确地区分各种情况下的励磁涌流和故障电流,从而证实该方法的有效性.

一种基于CNN与FFT‑ELM的输电线路故障识别与定位方法

及时、准确地检测输电线路故障类型与位置是提高电力系统可靠性最重要的问题之一,为此提出一种基于卷积神经网络(convolutional neural networks,CNN)与基于快速傅里叶变换(fast Fourier transform,FFT)的极限学习机(extreme learning machine,ELM)分类模型并行的输电线路故障识别及定位方法。首先,以故障电压时序图作为输入,构建CNN;然后,利用FFT将时域故障电压数据分解,提取各频段的电压峰值与相角作为故障特征样本;接着,以提取的故障特征样本集作为输入,构建ELM网络;最后,通过特征融合层将2个神经网络进行融合,输出故障类型和定位结果。实验结果表明,此方法对输电线路故障识别的准确率为99.95%、故障定位误差在500 m以内、平均误差为263.5 m,可靠性优于其他模型。

基于CNN-SVM的特高压三端混合直流线路故障区域识别方法

提出一种基于卷积神经网络–支持向量机(Convolutional neural network-support vector machine,CNN-SVM)的特高压三端混合直流线路故障区域识别方法。首先,对昆北侧、龙门侧的直流线路边界和柳北侧T区边界的频率特性进行分析,发现不同故障区域的故障特征存在一定差异。然后,使用经验小波变换提取故障特征,将其作为CNN-SVM的输入量,故障区域作为输出量,构建并训练CNN-SVM模型;将由测量点得到的故障特征量输入到训练完成的CNN-SVM模型中,进行故障区域识别。最后,搭建昆柳龙仿真模型,进行故障仿真实验验证。结果表明,该方法的故障区域识别率高,且可耐受300Ω的过渡电阻。

基于深度SVDD的发动机外涵静子叶片故障预警

外涵静子叶片是大涵道比涡扇发动机气路的核心部件之一,外涵静子脱出是一种较为严重的故障模式,此故障可能会导致飞机或其他发动机部件损伤,进而造成灾难性事故。对外涵静子叶片脱出故障的预警是一项重要的工作。但因其早期特征不明显,现有的方法较难对此类故障进行有效的预警。因此,针对该问题,基于监控数据提出一种深度特征提取的支持向量数据域描述(Support vector data description,SVDD)的故障预警方法,以实现对外涵静子叶片脱出故障的早期预警。首先,采用基于发动机气路性能辨识的建模方法,建立发动机特定性能参数的观测模型对气路参数进行深度特征提取,以真实状态量与模型观测量的差值作为航空发动机是否发生故障的特征;然后利用SVDD算法建立决策边界,实现故障数据的自动划分,决策边界生成的阈值可在故障发生之前的一定时间之内给出告警;最后,经过多次计算,结果表明,在故障早期直至故障发生的区间内,表征其健康状态的性能参数都与观测量有较大的偏移,表明了所选特征的有效性。使用数据增强方法生成故障仿真数据与真实数据进行对比验证,预警时间比故障真实发生时间预警模型平均提前3.14 h。

基于3D U-Net++卷积神经网络的断层识别方法及应用

断层解释是地震资料解释的基础与关键,准确合理的断层识别对油气开采有着至关重要的作用。随着油田对断层解释精度需求的日益提高,单纯通过基于人工的如相干体、曲率等属性的传统断层解释方法,其精度已无法满足要求。笔者在U-Net卷积神经网络模型的基础上进行改进,得出了一种自动断层识别方法,能够从任意三维地震图像中自动提取断层。文中该模型在足量样本集训练下,对两区块的实际地震数据进行自动断层识别,将识别结果进行分析对比。实验结果表明,该模型能够对任意三维地震数据进行自动断层识别,基于3D U-Net++网络模型的断层识别结果相比于传统U-Net网络识别结果准确性有明显提高,对潜山内部的小断层识别也表现出良好的效果,明显提高了常规、复杂断层识别的工作效率。

基于功图故障特征分析的液量计量方法研究

为解决现有原油产液量计量方法精度低、成本高、工艺流程复杂等问题,对一种基于功图故障种类与强弱特征和BP神经网络分析的组合计算油井产量方法进行研究。该方法通过马田系统操作从地面示功图中提取特征向量,并融合自适应算法,对示功图故障种类及强弱进行识别与区分;建立了产液量与不同故障类型及强弱之间的映射关系以此代替泵效,计算映射关系并建立BP神经网络以此来优化关系,得到基于故障种类与强弱的计量产量的模型。给出该计量模型的具体实现,并选择2020年的新疆公司某作业区50井的产量及示功图作为数据集进行计算验证。上述模型计算结果表明,计算的石油产量和实际的石油产量作对比,其平均相对误差为3.26%,平均绝对误差为0.5%,满足油田生产利益最大化。

HVD分解和GA-BP神经网络结合的井架钢结构损伤识别

针对井架钢结构冲击载荷振动信号非线性、非平稳性对损伤识别的干扰问题,提出了一种基于希尔伯特振动分解(Hilbert Vibration Decomposition,HVD)与遗传算法优化的神经网络(Genetic BP Neural Networks,GA-BP)相结合的智能故障诊断方法。首先,利用HVD分解的方法处理冲击载荷作用下的加速度非平稳振动信号;其次,由斯皮尔曼相关系数选取HVD分解后的最优(Intrinsic Mode Function,IMF)分量,以最优IMF分量能量变化率构造特征向量;最后,通过特征向量建立数据集进行神经网络训练,完成信号的特征学习和故障分类。利用ZJ70型井架钢结构模型进行冲击载荷作用下的单处损伤和多处损伤的不同工况实验验证,结果表明:对于单处损伤位置识别率达到90%,多处损伤位置识别率高达96%,利用HVD分解与GA-BP神经网络相结合的方法具有较好的稳定性,能够准确判断出井架钢结构损伤位置,具有一定的实际应用价值。