Fault Diagnosing System of Steam Generator for Nuclear Power Plant Based on Fuzzy Neural Networks

All kinds of reasons are analysed in theory and a fault repository combined with local expert experiences is establishedaccording to the structure and the operation characteristic of steam generator in this paper. At the same time, Kohonen algo-rithm is used for fault diagnoses system based on fuzzy neural networks. Fuzzy arithmetic is inducted into neural networks tosolve uncertain diagnosis induced by uncertain knowledge. According to its self-association in the course of default diagnosis. thesystem is provided with non-supervise, self-organizing, self-learning, and has strong cluster ability and fast cluster velocity.

Self-diagnosis method for faulty modules on wireless sensor node

In order to diagnose the working status of each module on sensor node and make sure the wireless sensor networks （WSN） work properly, the components of sensor node and their working characteristics are studied. An on-line fault self-diagnosis method for sensor node is proposed. First, a flexible fault sensing circuit is designed as a state detection module on sensor node. Second, a self- diagnosis algorithm is proposed based on the hardware design and the failure analysis on sensor node. Finally, in order to ensure the WSN reliability, the voltage changes of each module working statuses can be observed using the state detection module and the faulty module will be found out timely. The experimental results show that this self-diagnosis method is suitable to sensor nodes in WSN.

Evaluation of fault diagnosability for nonlinear uncertain systems with multiple faults occurring simultaneously

Up to present,the problem of the evaluation of fault diagnosability for nonlinear systems has been investigated by many researchers.However,no attempt has been done to evaluate the diagnosability of multiple faults occurring simultaneously for nonlinear systems.This paper proposes a method based on differential geometry theories to solve this problem.Then the evaluation of fault diagnosability for affine nonlinear systems with multiple faults occurring simultaneously is achieved.To deal with the effect of control laws on the evaluation results of fault diagnosability,a design scheme of the evaluation of fault diagnosability is proposed.Then the influence of uncertainties on the evaluation results of fault diagnosability for affine nonlinear systems with multiple faults occurring simultaneously is analyzed.The numerical simulation results are obtained to show the effectiveness of the proposed evaluation scheme of fault diagnosability.

Analysis of superheater's pipe wall overtemperature by fault tree diagnose

After research on a 2000t/h subcritical forced-circulation balanced ventilation were applied boiler and the structure and operation of its auxiliary system builds up this heat transfer model of a superheater's pipe wall and analyze the effect of primary factors on the overtemperature of the pipe wall. Fault tree structure was used to uncover the multiplayer logic between the overtemperature of the superheater's pipe wall and the faults.

Application of Single Channel Blind Separation Algorithm Based on EEMD-PCA-RobustICA in Bearing Fault Diagnosis

Aiming at the problem that ICA can only be confined to the condition that the number of observed signals is larger than the number of source signals;a single channel blind source separation method combining EEMD, PCA and RobustICA is proposed. Through the eemd decomposition of the single-channel mechanical vibration observation signal the multidimensional IMF components are obtained, and the principal component analysis (PCA) is performed on the matrix of these IMF components. The number of principal components is determined and a new matrix is generated to satisfy the overdetermined blind source separation conditions, the new matrix input RobustICA, to achieve the separation of the source signal. Finally, the isolated signals are respectively analyzed by the envelope spectrum, the fault frequency is extracted, and the fault type is judged according to the prior knowledge. The experiment was carried out by using the simulation signal and the mechanical signal. The results show that the algorithm is effective and can accurately diagnose the location of mechanical fault.

Study on Method of Fault Vibration Diagnose for Roller Overrunning Clutch

Fault diagnose of the roller overrunning clutch is a headache problem in engineering at home and abroad. This paper introduces a new method to solve the problem by using the wavelet transform to separate fault si gnal and further analyzing the impact frequency. The signal local singularities under the wavelet transform are studied. According to the propagation features of modulus maximums of the fault signal and the noise under the wavelet transfor m different on the scales, and by use of the signal wavelet decomposition-recon struction algorithm, the clutch shell vibration acceleration signal is decompose d, denoised, and reconstructed.The signal-to-noise of the monitored signal imp roved greatly.The fault characteristic signal on time domain is positioned.The f ault characteristic frequency is picked up. Experiment shows that it is quite effective.

FAULT DIAGNOSIS BASED ON INTEGRATION OF CLUSTER ANALYSIS, ROUGH SET METHOD AND FUZZY NEURAL NETWORK

In order to increase the efficiency and decrease the cost of machinerydiagnosis, a hybrid system of computational intelligence methods is presented. Firstly, thecontinuous attributes in diagnosis decision system are discretized with the self-organizing map(SOM) neural network. Then, dynamic reducts are computed based on rough set method, and the keyconditions for diagnosis are found according to the maximum cluster ratio. Lastly, according to theoptimal reduct, the adaptive neuro-fuzzy inference system (ANFIS) is designed for faultidentification. The diagnosis of a diesel verifies the feasibility of engineering applications.

Petri net model for diagnosis of permanent faults of a hydraulic system

Petri net model is applied to diagnose the permanent fault of hydraulic system within the framework of interpreted Petri net. The permanent fault is described as redundant structure of the model. A definition and a theorem are proposed to determine the diagnosability of the hydraulic system. The relations bwtween the diagnosability and other structure properties are also discussed. An example of actual hydraulic system is presented and its permanent fault can be diagnosed by the proposed method efficiently.

Restricted-Faults Identification in Folded Hypercubes under the PMC Diagnostic Model

System-level fault identification is a key subject for maintaining the reliability of multiprocessor interconnected systems. This task requires fast and accurate inferences based on big volume of data, and the problem of fault identification in an unstructured graph has been proved to be NP-hard （non-deterministic polynomial-time hard）. In this paper, we adopt the PMC diagnostic model （first proposed by Preparata, Metze, and Chien） as the foundation of point-to-point probing technology, and a system contains only restricted-faults if every of its fault-free units has at least one fault-free neighbor. Under this condition we propose an efficient method of identifying restricted-faults in the folded hypercube, which is a promising alternative to the popular hypercube topology.

THEORY ON THE DIAGNOSABILITY OF THE FAULT CLASSIFICATION APPROACH

This paper describes, by means of a Voronoi hypersphere, the nearest neighbor relations of all the feature submatrices in the fault classification space and analyses the deviation disturbance angles between fault feature submatrices and a k-dimension unitary matrix of the measured voltage change matrix. With the above two concepts, this paper discusses the diagnos-ability in the fault classification approach. The paper also classifies and defines the fault diagnosis problems. Finally, the paper derives the corresponding necessary and sufficient conditions for correct location of faults.

Self Fault-Tolerance of Protocols: A Case Study

The prerequisite for the existing protocols' correctness is that protocols can be normally operated under the normal conditions, rather than dealing with abnormal conditions. In other words, protocols with the fault-tolerance can not be provided when some fault occurs. This paper discusses the self fault-tolerance of protocols. It describes some concepts and methods for achieving self fault-tolerance of protocols. Meanwhile, it provides a case study, investigates a typical protocol that does not satisfy the self fault-tolerance, and gives a new redesign version of this existing protocol using the proposed approach.

Research on Early Fault Self-Recovery Monitoring of Aero-Engine Rotor System

In order to increase robustness of the AERS (Aero-engine Rotor System) and to solve the problem of lacking fault samples in fault diagnosis and the difficulty in identifying early weak fault, we proposed a new method that it not only can identify the early fault of AERS but also it can do self-recovery monitoring of fault. Our method is based on the analysis of the early fault features on AERS, and it combined the SVM (Support Vector Machine) with the stochastic resonance theory and the wavelet packet decomposition and fault self-recovery. First, we zoom the early fault feature signals by using the stochastic resonance theory. Second, we extract the feature vectors of early fault using the multi-resolution analysis of the wavelet packet. Third, we input the feature vectors to a fault classifier, which can be used to identify the early fault of AERS and carry out self-recovery monitoring of fault. In this paper, features of early fault on AERS, the zoom of early fault characteristics, the extraction method of early fault characteristics, the construction of multi-fault classifier and way of fault self-recovery monitoring are studied. Results show that our method can effectively identify the early fault of AERS, especially for identifying of fault with small samples, and it can carry on self-recovery monitoring of fault.

基于广域信息分析的智能配电网故障自愈技术研究

针对传统电网接地故障检测和自愈方法存在识别速度慢且精确度较低的问题,文中结合信号广域信息提出了一种配电网故障自愈方法。该方法构建了单相接地故障模型,并将零序电流作为分析的对象。将零序电流信号分解为时域、频域以及小波域分量来作为广域特征向量,并采用随机森林算法对其权重进行分类,利用LightGBM算法对分类后的广域特征向量加以训练,进而得到故障预测结果。在仿真测试中,所提算法能够抵抗过渡电阻与初始相位角变化对预测结果的影响,且其故障分析准确率的平均值为98.9%,在对比算法中为最优,表明该算法可为配电网故障自愈提供有效的技术支撑。

AVAILABILITY MODEL FOR SELF TEST AND REPAIR IN FAULT TOLERANT FPGA-BASED SYSTEMS

Dynamically reconfigurable Field Programmable Gate Array(dr-FPGA) based electronic systems on board mission-critical systems are highly susceptible to radiation induced hazards that may lead to faults in the logic or in the configuration memory. The aim of our research is to characterize self-test and repair processes in Fault Tolerant(FT) dr-FPGA systems in the presence of environmental faults and explore their interrelationships. We develop a Continuous Time Markov Chain(CTMC) model that captures the high level fail-repair processes on a dr-FPGA with periodic online Built-In Self-Test(BIST) and scrubbing to detect and repair faults with minimum latency. Simulation results reveal that given an average fault interval of 36 s, an optimum self-test interval of 48.3 s drives the system to spend 13% of its time in self-tests, remain in safe working states for 76% of its time and face risky fault-prone states for only 7% of its time. Further, we demonstrate that a well-tuned repair strategy boosts overall system availability, minimizes the occurrence of unsafe states, and accommodates a larger range of fault rates within which the system availability remains stable within 10% of its maximum level.

贫数据中基于模型自训练的空气处理设备故障诊断

针对空气处理设备(AHU)故障贫数据,基于深度置信网络(DBN)模型对4种特征选择算法进行对比研究,结果表明最大相关最小冗余算法的特征子集在诊断准确率及子集元素稳定性上表现最优。提出将DBN嵌入自训练框架的故障诊断模型,发现DBN自训练的诊断准确率较单纯DBN最高可提升19.5%。提出均匀抽样及按比例抽样2种自训练伪标签抽样策略,二者的诊断准确率均随抽样数减小而增大,在不同抽样数中的最大差异为3.42%;在所有贫数据样本中,均匀抽样策略始终优于按比例抽样,诊断准确率最大相差1.39%,表明在故障标签匮乏时,采用均匀抽样策略及较小的抽样数有利于提升DBN自训练的诊断性能。

A Self-Learning Diagnosis Algorithm Based on Data Clustering

The article describes an approach to building a self-learning diagnostic algorithm. The self-learning algorithm creates models of the object under consideration. The models are formed periodically through a certain time period. The model includes a set of functions that can describe whole object, or a part of the object, or a specified functionality of the object. Thus, information about fault location can be obtained. During operation of the object the algorithm collects data received from sensors. Then the algorithm creates samples related to steady state operation. Clustering of those samples is used for the functions definition. Values of the functions in the centers of clusters are stored in the computer’s memory. To illustrate the considered approach, its application to the diagnosis of turbomachines is described.

自监督学习结合对抗迁移的跨工况轴承故障诊断

轴承智能故障诊断应用中,由于实际工况复杂多变,极难获得足够的真实故障数据,且目标域和源域信号存在较大差异,导致深度模型的跨工况迁移识别也出现特征提取及分类困难、模型泛化性弱。考虑到目标域存在大量无标签数据,引入无监督思想,提出基于自监督学习结合对抗迁移的改进方法。首先根据信号本身特点创建辅助任务,对大量无标签数据学习,建立源域与目标域故障类别之间的内在联系;再通过对抗域适应和联合最大平均差异将源域知识迁移到目标域中,结合辅助任务优化两域差异,最终实现目标域准确的故障分类。用2个公开的轴承数据集上验证了所提方法的性能,实验结果表明,所提方法的故障诊断识别准确率在多数情况下均高于98%。

基于GWO-FCM的输油泵故障诊断模型自学习框架

随着输油泵场站无人化建设的发展,企业对输油泵故障诊断技术的要求也越来越高。目前,被广泛使用的利用机器学习算法进行输油泵故障诊断的方法都只能针对模型训练集中已包含的几类故障进行诊断,在企业的实际使用中,仍会出现其他不包含在训练集中的故障而不能被正确自动识别、诊断。针对上述问题,设计了一种输油泵故障诊断模型自学习框架,通过信号处理技术结合深度学习提取深层故障特征,提高工业现场数据的可分性;通过模糊C均值聚类结合相似度度量判别已知故障和未知故障,对出现的未知故障模式进行识别和记录;利用频繁出现的未知故障数据重训练模型,在原有诊断功能的基础上提高对未知故障的识别、诊断及学习能力。为验证方法的有效性,使用工业现场采集的输油泵数据进行实验,结果表明,现有诊断方法所提出的输油泵故障诊断模型自学习框架能够实现对未知故障的准确识别。

基于SOM-BP的全自动口罩机传动系统故障检测

针对口罩机在多工序生产中故障特征难以诊断的问题,提出了一种基于自组织映射(SOM)和误差反向传播网络(BP)的故障检测模型。首先针对4种减速机故障类型搭建SOM-BP复合型神经网络模型并完成检测分类,其次通过提取原振动信号的20组时域和频域参数作为SOM网络的输入样本进行初步聚类,并根据仿真结果确定最佳竞争层结构,最后将聚类后结果输入BP网络进行预测并完成分类,实现故障检测。研究结果表明,7×7竞争层结构下的SOM-BP复合型神经网络对于减速机的8种时域和频域参数的检测效果最优,分类准确率可达93.5%,173次迭代即可收敛,数据拟合度最高达0.99876,达到实际检测要求,验证了该方案的有效性和可行性。

初诊2型糖尿病肥胖患者自我管理行为的潜在剖面分析

目的 分析初诊2型糖尿病肥胖患者自我管理行为的潜在剖面类别特征及影响因素,为制订个体化干预措施提供依据。方法 便利选取2022年3月至2023年3月收治的新诊断2型糖尿病合并肥胖患者作为研究对象,采用一般资料调查表、糖尿病自我管理行为量表、糖尿病风险感知量表对其进行调查。使用潜在剖面分析识别患者自我管理行为潜在剖面类别,多项Logistic回归分析患者自我管理行为潜在剖面类别的影响因素。结果 最终纳入113例初诊2型糖尿病肥胖患者,其自我管理行为分为3个剖面类别:自我管理高效型26例占23.0%,自我管理有效型33例占29.2%,自我管理欠缺型54例占47.8%。相比较于“自我管理欠缺型”,糖尿病风险感知量表得分越高的患者归属于“自我管理高效型”和“自我管理有效型”的概率越大(OR=9.062、4.208,P<0.05);本科以上文化程度、家庭人均月收入>5 000元归属于“自我管理高效型”的概率大(OR=0.033、0.034,P<0.05)。结论 初诊2型糖尿病肥胖患者自我管理行为有3种潜在剖面类别,医护人员应根据其自我管理行为特征实施针对性干预措施。