基于概率因果模型与遗传算法的船舶核动力装置故障诊断的新方法

介绍一种将概率因果模型和遗传算法相结合的核动力装置二回路凝给水系统的故障诊断方法,它将概率因果模型的似然函数作为遗传算法的适应函数,从而将复杂系统的故障诊断转化为最优问题。仿真结果表明,该方法能够适应诊断过程中出现的不确定性,并实现多故障诊断,具有较高的诊断可靠性和实用性。     

基于遗传算法的核动力设备实时故障诊断系统

在神经网络故障诊断研究的基础上,将遗传算法引入神经网络优化,并在此基础上设计了核动力设备遗传优化神经网络实时故障诊断系统.测试结果表明,该故障诊断系统能迅速、准确、可靠地诊断出所设故障.     

基于遗传算法和知识库的核电厂故障诊断方法

将遗传算法与经典概率论相接合，利用知识库，针对核电厂故障诊断的特点，提出了一种故障诊断方法。本方法将核电厂部件状态与遗传算法中的群体相联系，利用专家知识的核电厂信号对群体进行约束，使该群体在诊断过程中不断发展变化，从而找出适合条件的个体，达到故障诊断的目的。在北京核电厂模拟培训中心９５０ＭＷ全尺寸模拟机上的实验表明，该方法对诊断过程中出现的虚假信号、专家知识不完备等问题有相当的适应性。     

BP神经网络改进算法在核电设备故障诊断中的应用

根据训练误差大小自适应调整神经元输入特性参数,并应用改进的遗传算法对神经网络的权值和隐含层数目进行优化,对传统的人工神经网络误差反传算法进行了改进,使训练算法的收敛速度大大提高.将人工神经网络技术和改进的BP网络训练算法应用于核电设备故障诊断,并以核电蒸汽发生器U形管破裂为例,建立了故障诊断模型.仿真结果表明,该算法的应用是可行的.     

核动力系统神经网络故障诊断专家系统研究

以核动力系统故障诊断专家系统所存储的产生式规则库为依据,构造专家系统人工神经网络模块,建立具有并行推理能力的神经网络推理机制和实例学习能力的知识获取机制。在此基础上,建立了针对所构建神经网络的推理解释机制。系统融合了传统专家系统和神经网络各自优势,能高效、准确地进行实例推理、诊断解释并能有效地从实例中获取知识。     

核电厂故障检测与报警系统的发展概况

文章分析了传统的核电厂故障诊断及报警系统的结构和特点，指出了其需要改进的方面。概要评述基于知识库的专家系统、基于故障树的诊断系统和基于神经网络的诊断系统的发展概况及其在核电厂报警系统中的应用前景。     

基于HMM&SVM的核动力设备机械故障诊断方法研究

核动力设备复杂且积累的资料与故障样本少,传统的诊断方法有待改进。隐马尔可夫模型与支持向量机是一种新的智能诊断技术。本文针对核动力设备机械故障诊断的特点,采用隐马尔可夫模型建模的方式进行故障的初步诊断,再利用支持向量机小样本的强推广能力进行进一步甄别。主泵故障模拟装置上的验证实验表明,HMM&SVM混合模型具有较高的故障识别率。     

基于故障树的复杂系统故障诊断软件设计研究

可靠性和安全性要求较高的复杂系统对故障诊断工作的要求也很高。根据系统在设计、运行阶段工作中建立有大量的故障树模型的特点,本文将故障树分析方法与基于知识的方法相结合,提出了一种基于故障树的复杂系统故障诊断方法,并且开发了原型系统,实现了对车载大气光学参数测量系统的快速、智能诊断。     

反应堆补水系统诊断贝叶斯网络的建立和应用

建立了补水系统贝叶斯故障诊断网络。在结合补水系统结构特点、运行规程和专家经验的基础上构建了初始诊断贝叶斯网络,运用基于微粒群优化的贝叶斯网络学习算法学习故障数据集,进一步构建完整网络,并进行推理分析。所建网络能有效分析和更新系统中各节点故障概率,为故障诊断提供辅助决策。     

混合遗传算法在核事故源项反演中的应用

针对国内外普遍关注的核事故源项反演与事故后果评价的问题,运用遗传算法-单纯形法结合烟团模型实现了对源强的反算和释放点位置的快速定位。遗传算法-单纯形法与遗传算法-模式搜索法、遗传算法和单纯形法等3种算法的比较结果表明：遗传算法-单纯形法结合了遗传算法和单纯形法两种算法的优势,也弥补了各自算法的缺陷,其反算值可与期望值较精确符合;扩散模式模块、GA模块和NM模块3者可简单直接的结合,结合所需编写的代码较少,通用性广;GA模块和NM模块的计算花费较少,适用于核电厂对源项的快速估计。     

Enhanced graph-based fault diagnostic system for nuclear power plants

Scheduled maintenance and condition-based online monitoring are among the focal points of recent research to enhance nuclear plant safety.One of the most effective ways to monitor plant conditions is by implementing a full-scope,plant-wide fault diagnostic system.However,most of the proposed diagnostic techniques are perceived as unreliable by operators because they lack an explanation module,their implementation is complex,and their decision/inference path is unclear.Graphical formalism has been considered for fault diagnosis because of its clear decision and inference modules,and its ability to display the complex causal relationships between plant variables and reveal the propagation path used for fault localization in complex systems.However,in a graphbased approach,decision-making is slow because of ruleexplosion.In this paper,we present an enhanced signed directed graph that utilizes qualitative trend evaluation and a granular computing algorithm to improve the decision speed and increase the resolution of the graphical method.We integrate the attribute reduction capability of granular computing with the causal/fault propagation reasoning capability of the signed directed graph and comprehensive rules in a decision table to diagnose faults in a nuclear power plant.Qualitative trend analysis is used to solve the problems of fault diagnostic threshold selection and signed directed graph node state determination.The similarity reasoning and detection ability of the granular computing algorithm ensure a compact decision table and improve the decision result.The performance of the proposed enhanced system was evaluated on selected faults of the Chinese Fuqing 2 nuclear reactor.The proposed method offers improved diagnostic speed and efficient data processing.In addition,the result shows a considerable reduction in false positives,indicating that the method provides a reliable diagnostic system to support further intervention by operators.     

Study and Design a Strategy to Control High-Voltage Power Supply Based on PSM Technology on J-TEXT Tokamak

To energize the gyrotrons of the electron cyclotron resonance heating system, high-voltage power supply (HVPS) at 100 kV/60 A was built on J-TEXT tokamak. The HVPS was based on pulse step modulation (PSM) technology. This power supply has many advantages, including low energy storage and a short protective time in the event of a short circuit fault. In this study, we analyze and compare several common control strategies for PSM power supply. To enhance dynamic performance, the improved queue-up PSM control strategy was proposed. The control algorithm was generated using a controller based on parallel card interface extensions for instrumentation (PXI) technology, which conforms to the control, data access, and communication standard of the ITER. This controller also operates the modules of switch power supply. Experimental results show that the control system is effective and that the PXI controller can rebalance the output voltage and frequency in 60 μs when the module malfunctions under the improved system. Furthermore, the algorithm for the PXI controller is easy to develop and does not require much resources.     

核电系统故障诊断专家系统研究

针对核电系统的故障特征建立了专家系统．专家系统中的知识采用启发式表示,知识可以表示成确定性和不确定性知识,对于确定性知识采用简单的命题逻辑的推理机制,对于不确定性知识引入了确定性因子kCF来表示知识的不确定性,并据此建立了推理机制．为了解决专家系统中知识获取的“瓶颈”问题、将Rough集理论引入到了核电系统故障诊断系统,并将基于可辨识矩阵的约简算法做了改进在对属性重要性进行测度时,考虑了同决策类内具有同属性值属性及属性在可辨识矩阵中出现的次数等因素用该诊断系统在仿真机上对不同的故障进行了诊断,并得到了正确的结果。     

基于自适应神经网络技术的钠泵诊断

在径向基函数网络的基础上，应用自适应算法使故障诊断方法具有自适应性：在对中国实验快堆(CEFR)钠泵的故障仿真实验中．该方法能够较好地识别出故障。为了提高识别的准确率，本文在自适应径向基函数网络的基础上对其自适应算法进行了一定的改进，使故障诊断的准确率有所提高。     

Monitoring and fault diagnosis of the steam generator system of a nuclear power plant using data-driven modeling and residual space analysis

An advanced fault detection and isolation (FDI) technique was developed for the steam generator system of a typical pressurized water reactor (PWR) plant. The system characterization module used the group method of data handling method for modeling the interrelationship among the various process variables associated with a U-tube steam generator (UTSG). A principal component analysis algorithm was utilized for the generation of representative fault signatures. The accuracy of fault identification was quantified using normalized vector projections on to fault spaces. Other classical pattern classification methods were executed in parallel to increase the robustness of FDI results. Six typical static faults and one transient fault were successfully detected and isolated using a full-scope simulation of a four-loop PWR. The results demonstrate the implementation of the FDI algorithm for both instrument and actuator monitoring.     

基于KNN改进算法的闪烁体探测器故障诊断

为研究核探测器的可靠性,本文提出了一种基于K近邻（KNN）算法的闪烁体探测器故障诊断方法。首先通过提取不同工况下的核脉冲信号的下降沿时间、信号幅值及能谱信号的能峰位置和低道址计数等特征参数,建立故障核信号统计特征信息库。通过修正权重因子,改变邻点距离计算方式等方法改进KNN算法建立闪烁体探测器故障诊断模型,并搭建故障数据采集验证系统,提取探测器输出信号的特征信息放入到模型中进行诊断实验。实验结果表明,该方法不仅能实现对探测器故障类别的智能诊断,而且能对不同故障的严重程度做出良好的判别。     

基于入侵性野草算法的核动力装置故障诊断

针对船用核动力装置故障原因与相应故障征兆之间并非完全一一对应的特点,提出了一种将入侵性野草算法和概率因果模型相结合的故障诊断方法,该方法将概率因果模型中的似然函数作为入侵性野草算法的适应函数,从而将复杂系统的故障诊断转化为优化问题。结果表明,该方法能用于诊断过程中出现的不确定性问题,也可实现通过多个征兆来诊断多个故障的目的,且具有较高的诊断可靠性与实用性。     

基于粗糙决策模型的核电厂故障诊断方法

将粗糙集和信息熵相结合,提出了改进的决策树ID3算法,并提出了基于此算法和粗糙集约简算法的故障诊断方法.将该方法用于核电厂故障诊断.结果表明,该方法具有良好的故障诊断性能,可在核电厂故障诊断领域中应用.