基于矢谱和L-M神经网络的旋转机械故障诊断研究

讨论了矢谱融合技术和Levenberg-Marquardt(L-M)神经网络的相关理论,提出了基于矢谱和L-M神经网络的旋转机械故障诊断方法,建立了基于矢谱的旋转机械常见故障诊断L-M神经网络模型.模拟实验结果表明:与基于单通道数据的诊断结果对比,将矢谱数据融合应用于旋转机械常见故障诊断,可有效提高故障诊断的准确率.     

基于全信息小波包和支持向量机的旋转机械故障诊断

针对传统旋转机械单通道故障诊断的信息不完整以及缺少故障样本等问题,提出了基于全信息小波包和支持向量机的旋转机械故障诊断方法.运用小波包频道能量分解技术提取了全信息能量特征向量,以此作为支持向量机多故障分类器的故障样本,经训练的分类器作为故障智能分类器可对设备工作状态进行自动识别和诊断.实验研究表明:基于全信息小波包和支持向量机的故障诊断方法能准确、有效地对旋转机械的工作状态和故障类型进行分类,显著提高了故障诊断的准确率.     

基于小波包能量特征向量神经网络的旋转机械故障诊断

为精确诊断旋转机械的故障,提出一种基于小波包特征向量的神经网络故障诊断方法。用转子台信号模拟旋转机械故障,并对采集到的信号进行3层小波包分解,构造小波包特征向量,并以此为故障样本对3层BP网络进行训练,实现智能化故障诊断。实验结果表明训练好的神经网络能够很好地诊断出转子台故障类型,为旋转机械的故障诊断提供了新方向。     

获取知识的一种新方法_粗糙集_Rough Set

旋转机械故障诊断的一个困难问题是诊断规则的获取。提出获取知识的一种方法－粗糙集（RS）,RS能自动地从旋转机械的大量信息中有效地获取诊断知识,并能减少误诊与漏诊现象。文中介绍了RS的原理与方法并给出应用实例。     

基于MATLAB的集成神经网络在给水泵故障诊断中的应用

针对大型火电站电动给水泵常见的振动故障,采用基于MATLAB的集成神经网络对给水泵的振动故障进行诊断。从单个神经网络开始,从信息融合的角度建立了集成神经网络故障诊断方法,探讨集成神经网络的实现策略和组建原则,并给出给水泵振动故障诊断的实例,证明该诊断方法提高了故障确诊率。     

谐小波模糊神经网络应用于旋转机械的故障诊断

根据旋转机械复杂的故障特点，提出了结合谐小波分析、模糊理论和神经网络形成的谐小波模糊神经网络方法，并将其应用于旋转机械的故障诊断，实现了模糊故障诊断。通过计算机实现了全部算法。仿真和试验的结果表明：谐小波模糊神经网络在处理多故障耦合的情况时优势明显，故障诊断正确率高，证明该方法行之有效，为旋转机械的故障诊断提供了理论支持和新方法。图2表3参7     

嵌入式旋转机械状态监测系统的设计与研究

旋转机械状态监测对旋转设备运行安全,降低设备维修费用,提高设备利用率有重大意义。介绍了一种基于ARM的嵌入式监测装置,通过该装置实现对旋转机械的在线监测。同时,为建立旋转机械的故障诊断和维护系统奠定了基础。     

基于多信息融合的汽轮发电机组故障诊断方法研究

目前汽轮发电机组配备了大量的传感器，如何将各种传感器信息充分利用起来提高诊断准确性是一个很实际的问题。提出了基于多信息融合的汽轮发电机组故障诊断方法，介绍了信息融合的基本概念，给出了基于主观Bayes方法和基于模型的多传感器融合的实例。     

粗集理论在水力机械故障诊断中的应用潜力分析

粗集理论是一种处理模糊性和不精确问题的新型数学工具,为分析和处理不完备信息提供了有力的分析手段。文中对近年来粗集理论在机械故障诊断应用方面作了介绍及评述,并将此方法推广到水力机械故障诊断方面,重点阐述了粗集理论与常见的数据挖掘、人工神经网络、支持向量机等软计算方法的融合,这将为解决水力机械故障诊断中的难题提供一种新的思路和方法。     

基于因子隐Markov模型的旋转机械故障诊断方法的研究

针对旋转机械升降速过程非平稳、重复再现性不佳的特点，隐Markov模型具有很强的针对性。因子隐Markov模型是一种多链隐Markov模型，它是隐Markov模型的一种扩展形式。作者将因子隐Markov模型引入到旋转机械升降速过程的故障诊断中，提出了基于因子隐Markov模型的旋转机械故障诊断方法，并且利用它成功地对旋转机械的故障进行了分类。实验结果表明：该方法是有效的。图4表2参8     

基于信息融合技术的发动机故障诊断的研究

研究了信息融合技术在电控发动机故障诊断中的应用。研究结果表明,基于神经网络的特征层信息融合诊断效果明显优于单一传感器,而且可实现信息压缩,进行实时处理与诊断;基于Dempster-Shafer证据推理的决策层信息融合,可对异质传感器信息进行非同步处理,对发动机故障分类准确性高、可靠性强,但融合精度不及特征层融合方法,预处理代价高。在实际应用中,应根据传感器类型、信号预处理方式、系统的复杂程度等合理选择信息融合方法。     

基于虚拟阻抗的复杂配电网故障区段定位新方法

针对含分布式电源的复杂配电网故障区段定位的问题,提出一种基于虚拟阻抗的故障定位新方法。首先以区段为单元识别畸变节点的故障信息,采用尝试赋值法校正畸变节点的故障信息;然后利用区段端节点过流信息初步判定故障位置,将故障区段端节点间的虚拟导纳值用0表示,非故障区段端节点间虚拟导纳用1表示,进而形成能反映节点间连通性的节点虚拟阻抗矩阵,根据故障电流的流通路径最终确定故障区段位置。具体算例验证表明,该方法具有故障定位准确、效率高的特点,可以满足复杂配电网的多点信息畸变校正和多重故障的故障区段定位。     

Fault diagnosis of proton exchange membrane fuel cell system of tram based on information fusion and deep learning

The reliability of fuel cell tram depends largely on the normal operation of on-board proton exchange membrane fuel cell (PEMFC) system. Therefore, timely and accurate fault diagnosis is necessary to further commercialize the fuel cell tram. And, a new fault diagnosis method BPNN-InceptionNet based on information fusion and deep learning is proposed in this paper. In this method, high-dimensional abstract features are extracted from the original measurement information by back propagation neural network (BPNN) and converted into feature maps for information fusion in feature level. Then the feature maps are transferred to a proposed Convolutional Neural Network (CNN) based on InceptionNet to realize fault classification. From the experiments, it is found that the kappa coefficient by BPNN-InceptionNet for the test set can reach 0.9884, which is better than that by BPNN, BPNN-VGG, and support vector machine (SVM) classifiers, meaning that the proposed method can achieve better diagnostic performance.     

基于模型融合的风电机组齿轮箱故障诊断

针对单视角特征及单一模型对齿轮箱复合故障的诊断准确率较低的问题,提出基于比例冲突分配规则的模型融合故障诊断方法.首先,对齿轮箱振动信号进行特征提取,针对复杂复合故障,从时域及时频域角度构造特征;然后,将多视角特征送入多个子模型中进行初步诊断,得到互补性强的诊断结果;最后,模型输出的分类概率由第6类比例冲突分配规则(PC...     

基于故障状态信息重要度分析的复合故障判别

在设备故障诊断领域,复合故障的诊断一直是一个难点,缺乏行之有效的诊断方法。通过对故障信息熵的研究,提出了基于故障信息重要度的故障分类规则,此方法不仅能有效识别单一故障,也能诊断复合故障,经用于发动机的故障诊断,取得了良好的效果。     

Fault location of distribution networks based on multi-source information

In order to promote the development of the Internet of Things (loT), there has been an increase in the coverage of the customer electric information acquisition system (CEIAS). The traditional fault location method for the distribution network only considers the information reported by the Feeder Terminal Unit (FTU) and the fault tolerance rate is low when the information is omitted or misreported. Therefore, this study considers the influence of the distributed generations (DGs) for the distribution network. This takes the CEIAS as a redundant information source and solves the model by applying a binary particle swarm optimization algorithm (BPSO). The improved Dempster/S-hafer evidence theory (D-S evidence theory) is used for evidence fusion to achieve the fault section location for the distribution network. An example is provided to verify that the proposed method can achieve single or multiple fault locations with a higher fault tolerance.     

Field experiment using transient energy method to locate a single-phase to ground fault

Distribution networks in China and several other countries are predominantly neutral inefficiently grounding systems (NIGSs), and more than 80% of the faults in distribution networks are single-phase-to-ground (SPG) faults. Because of the weak fault current and imperfect monitoring equipment configurations, methods used to determine the faulty line sections with SPG faults in NIGSs are ineffective. The development and application of distribution-level phasor measurement units (PMUs) provide further comprehensive fault information for fault diagnosis in a distribution network. When an SPG fault occurs, the transient energy of the faulted line section tends to be higher than the sum of the transient energies of other line sections. In this regard, transient energy-based fault location algorithms appear to be a promising resolution. In this study, a field test plan was designed and implemented for a 10 kV distribution network. The test results demonstrate the effectiveness of the transient energy-based SPG location method in practical distribution networks.     

Fault diagnosis of PEMFC systems based on an auxiliary transfer network

Data-driven fault diagnosis methods require huge amounts of expensive experimental data. Due to the irreversible damage of severe fault embedding experiments to proton exchange membrane fuel cell (PEMFC) systems, rare available data can be obtained. In view of this issue, a fault diagnosis method based on an auxiliary transfer network (ATN) is proposed. This method uses two parallel neural networks (main and auxiliary neural network) and a prediction fusion module to realize fault diagnosis. The auxiliary neural network is a fault diagnosis classifier pretrained based on both slight and severe fault simulative data, and its weights are transmitted into the ATN structure and frozen. After that, the main neural network is trained based on a large number of slight fault experimental data and a small number of severe fault experimental data. Through ATN, the main neural network learns the abstract features of severe faults under the guidance of auxiliary neural network, and realizes the transfer learning from simulation-based fault diagnosis classifier to experiment-based fault diagnosis classifier. Through testing, the accuracy and precision of ATN-based fault diagnosis classifier with LSTM as both main and auxiliary neural network reaches 0.993 and 1.0 respectively, which is higher than the common data-driven methods.     

基于多尺度深度卷积网络特征融合的滚动轴承故障诊断

针对传统滚动轴承故障诊断模型对工程先验知识依赖性强、提取特征不充分、分类器选取困难等问题,提出一种基于多尺度深度卷积网络特征融合的滚动轴承故障诊断模型.首先,建立集特征提取与模式识别于一体的卷积神经网络模型,利用小波变换将滚动轴承振动信号转换为二维图像作为输入样本集.然后,在网络结构中构建多尺度特征融合模块自适应提取故...