基于灰色马尔可夫链的核设备退化趋势预测

核设备的状态退化趋势预测是确定其在役检查以及维修计划的重要依据,但由于核设备样本小、退化数据缺乏、退化轨迹具有波动性,难以采用传统的概率统计模型对其退化趋势进行精确预测。为此,本文提出应用灰色马尔可夫链模型对核设备退化趋势进行预测的方法,该方法充分利用GM（1,1）和马尔可夫链的优点,能够有效提高核设备退化趋势预测的精度。并以屏蔽泵的退化数据为样本,精确预测了屏蔽泵的退化趋势,同时与GM（1,1）模型的预测结果进行了对比。结果表明,灰色马尔可夫链模型的预测精度更高,能够对核设备的退化趋势进行精确预测。     

一种自适应的控制棒驱动机构滚轮丝杠副剩余寿命预测方法研究

针对反应堆控制棒驱动机构（CRDM）滚轮丝杠副剩余使用寿命（RUL）预测中如何选取有效的健康状态指标和合理构建预测模型的难题,提出了一种新的滚轮丝杠副RUL预测模型。采用基于生成拓扑映射算法（GTM）的负对数似然概率（NLLP）指标作为滚轮丝杠副的健康状态指标,利用K均值聚类算法对NLLP指标进行状态划分。利用历史数据和在线监测数据构建基于Markov模型和最小均方算法（LMS）的自适应预测模型,根据设定的阈值预测得出剩余寿命。通过实验验证,结果表明本文选取的健康状态指标能够有效地反映设备状态,所给出的自适应预测模型比一般的预测模型的预测精度高,为合理构建RUL预测模型提供了依据。      

核级电动闸阀样机端部加载试验技术要点探讨

核级电动闸阀端部加载试验是通过对阀门管端施加载荷,测定载荷施加过程中力和应变的变化,考核阀门的刚度、强度以及在载荷作用时和作用后的性能指标,验证当阀门受到所有管端载荷及压力和自重在内的正常工作载荷作用时阀门的可操作性。本文探讨了核级电动闸阀样机端部加载试验的技术要点,并以一实际样机试验为例,描述了核级电动闸阀样机端部加载试验的试验方法及其试验结果。     

基于SWT与相空间曲变的控制棒驱动机构退化状态跟踪和识别

控制棒驱动机构（CRDM）是保证反应堆正常运行与安全的重要设备。针对其转速低、性能退化特征微弱且易受其他信号源干扰的特点,本文采用小波半软阈值（SWT）与相空间曲变相结合的方法对其退化状态进行了跟踪与识别。通过与基于振动幅值均方根（RMS）方法的对比,表明该方法能够克服传统基于线性信号处理理论对微弱性能退化状态不能很好地跟踪和识别的缺点,可以较好地对强背景噪声下的旋转机械性能退化状态进行跟踪和识别。      

HHT变换联合人工神经网络的核信号时频特征分析与识别

用HHT变换联合BP人工神经网络分析处理252Cf自发裂变中子源驱动式核信息系统输出的随机核信号,获得信号时频特征,并分类识别核信号样本.理论分析和实验结果表明,基于HHT变换边际谱的特征提取方法,HHT反映信号时频域联合分布能较好地提取实际非平稳随机核信号的时频特征;用BP神经网络分类器对核信号样本分类识别,取得较高正确率,验证了此方法的有效性和合理性.     

基于径向基代理模型与小生境遗传算法的铅铋反应堆堆芯智能优化

为解决铅铋反应堆多因素耦合影响下的复杂非线性多维优化问题,构建了基于径向基（RBF）代理模型预测、正交拉丁超立方抽样（OLHS）和小生境遗传算法（NGA）寻优的堆芯智能优化方法,开发了包含抽样、蒙卡程序耦合处理、堆芯参数预测寻优等功能的铅铋反应堆设计优化平台,并以堆芯最小燃料装载量为优化目标进行方案寻优验证。研究结果表明:RBF代理模型可准确快速地预测铅铋反应堆堆芯特性参数,与蒙卡程序计算值比较,其预测的堆芯有效增殖因子（k_eff）相对误差在±0.1%以内;该智能优化方法应用于铅铋反应堆堆芯优化是可行的,能找到多因素共同变化约束下的最优目标方案,且极大缩减了设计方案的搜索计算时间。本研究建立的堆芯智能优化方法可为铅铋反应堆多物理、多变量、多约束耦合影响的优化设计提供思路。      

毛发中铀的二次离子质谱分析研究

毛发对人体微量元素有一定的蓄积效应,且作为一种无创生物检材,在核取证、环境监测、职业暴露检测、医学等领域逐渐受到重视。目前,大多采用整体分析技术测量毛发中元素的平均含量,毛发微区分析相关研究较少。为拓展毛发分析方法应用,本研究通过喂养大鼠硝酸铀酰溶液的方式获得含铀量较高的鼠毛样品,对毛发样品预处理方法、毛发中铀的二次离子质谱（SIMS）测量方法开展研究。结果表明,使用SIMS测量毛发时,样品靶表面刻槽能有效提升信号强度稳定性;样品表面镀铜能有效降低表面电荷积累并减少多原子离子对铀离子信号干扰。使用多接收电感耦合等离子体质谱仪（MC-ICP-MS）和SIMS对鼠毛中铀分布、平均铀含量及同位素比进行测量。结果表明,该方法能有效判断大鼠是否摄铀,分辨大鼠摄铀类型。本研究可为毛发样品溯源分析提供技术支持,对毛发中微量元素的分析研究具有重要意义。     

高频地震激励对核级阀门危害性分析

利用试验和修正后的集中质量有限元模型预测安装在管道中阀门在不同频率成分地震激励下的响应,研究高频地震激励对管道中质量较大核级阀门的危害性。研究结果表明：高频地震激励对核级阀门的危害在于使阀门以其自身固有振型发生共振,此时阀门顶部取代阀门与管道连接位置成为阀门中响应最大的位置,这会导致安装于阀门顶端的驱动机构遭受苛刻的地震工况。增加管道阻尼和阀门刚度能有效降低高频激励对阀门的危害,但增加阀门刚度会导致管道响应增大。利用等效静力法对阀门进行抗震鉴定时,分析结果对阀体水平部位内力估计不足,对阀体垂直部分、阀盖等阀门上部构件的内力估计结果具有较大裕度。     

主给水隔离方式对主蒸汽管道破裂事故质能释放与安全壳行为的影响分析

以秦山第二核电厂3、4号机组为对象,采用THEMIS程序分析了隔离给水管线上阀门和停运主给水泵2种给水隔离方式对主蒸汽管道破裂（MSLB）事故质能释放的影响,并采用PAREO9程序模拟了安全壳热工水力现象。结果表明,采用隔离阀门的方式能够更加有效地缓解MSLB的质能释放,采用停运主给水泵方式时提高浓硼箱中的硼浓度可以在一定程度上缓解MSLB的质能释放,避免安全壳超压。      

基于PCA与SDG的反应堆一回路系统故障诊断方法研究

反应堆一回路系统复杂,运行参数耦合多变,安全问题突出。为了保障运行安全、快速定位故障源,提出基于主元分析（PCA）与符号有向图（SDG）的一回路系统故障诊断模型。以一回路系统为诊断研究对象建立PCA-SDG模型,通过PCA分析监测参数的残差,判断故障的发生;然后采用SDG模型进行反向推理,找到潜在故障类型。通过模拟机仿真试验验证,该方法能够有效诊断故障,并提供报警传递路径。该方法可用于运行人员辅助决策,对运行装置的状态监测、报警分析和故障诊断具有重要意义。      

Hazard Analysis of Nuclear Safety Class Valve Induced by High Frequency Seismic Excitation

The test and the updated lumped mass finite element model were used to predict the response of the valve installed in the pipeline under the seismic excitation of different frequency components, and the hazard of high frequency seismic excitation to large-mass nuclear safety class valves in the pipeline was studied. The results show that the high frequency seismic excitation causes the nuclear safety class valve to resonate with its own mode of vibration. At this moment, the top of the valve replaces the position where the valve is connected to the pipe to become the position with the largest response amplitude in the valve, which causes the drive mechanism installed on the top of the valve to suffer severe seismic conditions. Increasing the pipe damping and valve stiffness can effectively reduce the hazard of high frequency excitation to the valves, but increasing the valve stiffness will lead to the increase of the pipe response amplitude. When the equivalent static method is used for seismic identification of the valve, the analysis result is insufficient to estimate the internal force of the horizontal part of the valve body, and has a large margin to estimate the internal force of the vertical part of the valve body, the valve cover and other upper parts of the valve.     

核电厂MSIV电磁阀性能检测系统开发

针对压水堆核电厂MSIV上电磁阀性能检测的需要,研制了电磁阀开启/关闭特性及密封性等综合性能检测系统,介绍了检测原理方法和系统组成。该系统能有效检测出MSIV上电磁阀的综合性能,继而指导电磁阀后续的维修工作方向,有效避免人因失误,可作为一项重要的检修专用工具。      

Engineering analysis of mass flow rate for turbine system control and design

The mass flow rate is determined in the steam turbine system by the area formed between the stem disk and the seat of the control valve. For precise control the steam mass flow rate should be known given the stem lift. However, since the thermal hydraulic characteristics of steam coming from the generator or boiler are changed going through each device, it is hard to accurately predict the steam mass flow rate. Thus, to precisely determine the steam mass flow rate, a methodology and theory are developed in designing the turbine system manufactured for the nuclear and fossil power plants. From the steam generator or boiler to the first bunch of turbine blades, the steam passes by a stop valve, a control valve and the first nozzle, each of which is connected with piping. The corresponding steam mass flow rate can ultimately be computed if the thermal and hydraulic conditions are defined at the stop valve, control valve and pipes. The steam properties at the inlet of each device are changed at its outlet due to geometry. The Compressed Adiabatic Massflow Analysis (CAMA) computer code is written to predict the steam mass flow rate through valves. The Valve Engineered Layout Operation (VELO) test device is built to experimentally study the flow characteristics of steam flowing inside the control valve with the CAMA input data. The Widows’ Creek type control valve was selected as reference. CAMA is expected to be commercially utilized to accurately design and operate the turbine system for fossil as well as nuclear power plants.     

An analytical study of the dynamics and stability of a spring loaded safety valve

Spring loaded self-actuating safety valves are employed as part of the overpressure protection systems in various industrial applications. In order to design and predict their performance it is necessary to study the dynamic behavior of the valve over a range of fluid and system conditions. A one-dimensional model has been developed to study the effects of different valve parameters such as the spring-mass characteristics, geometry of internal parts, adjustment ring settings, bellows etc. which influence the dynamic behavior and stability of the valve. Analytical results for steam flow conditions are presented to demonstrate the relative effects of these parameters on the valve opening time, maximum lift, blowdown (upstream pressure differential between the valve opening and closing) and any oscillations of the valve stem. If the valve is not properly backpressure compensated, it may become unstable as the stagnation pressure at the valve inlet decreases. Lowering of the guide adjustment ring position or raising the nozzle adjustment ring generally results in improved stability, shorter valve opening time, higher lift and longer blowdown. The effect of damping on the valve stability is also demonstrated. The model can be used to evaluate the design of safety valves and damping devices to eliminate unstable valve dynamic behavior.     

Development of diagnosis algorithm for the check valve with spectral estimations and neural network models using acoustic signals

We have developed a method for detecting and diagnosing a disk wear failure and a foreign object failure among the various failure modes of check valves. The method is based on the acoustic emission sensors which can detect the sound wave of the leakage flow and the estimation of the power spectral densities with an auto-regressive model. For validating the method, we implemented a hydraulic test loop with an artificially failed check valve. We have found that the frequency spectrums from the acoustic signals are strongly dependent on the failure modes of the check valve and that they are nearly independent of the failure size and operating pressure through an estimation of the power spectral density with an auto-regressive signal processing model. In addition, the root mean square values of the acoustic signal and the amplitudes of the power spectral density as well as the loop pressure have a strong dependency on the failure size in each failure mode of the check valve. We developed a diagnosis algorithm by using neural network models in order to identify the type and size of the failure in the check valve. The diagnosis algorithm consists of a hierarchical model composed of three back-propagation neural networks. The results of our research and the experiments show that the diagnosis algorithm is proven to be a good solution for identifying the failures of the check valves without any disassembling work.     

直动电磁阀参数设计与分析

控制棒水压驱动技术是清华大学核能与新能源技术研究院具有自主知识产权的一项新型发明专利技术,组合阀属于该项技术的关键部件。组合阀由3个直动电磁阀组成,电磁阀的性能直接影响组合阀的性能,从而影响控制棒水压驱动技术的运行性能。本工作运用二次回归正交试验拟合方法和复合形优化方法,对直动电磁阀的参数进行了设计和分析,并进行了部分实验验证。分析结果表明：输入直动电磁阀的电流数值对电磁力大小影响最大;通过优化方法可以得出直动电磁阀的优化设计参数和最大电磁力。     

AP1000多根蒸汽发生器传热管破裂分析

利用RELAP5/MOD3程序对AP1000多根蒸汽发生器传热管破裂事故进行了分析。基于最佳估算方法,分析了1~5根蒸汽发生器传热管破裂的工况。分析结果表明:在大气释放阀可用的情况下,主蒸汽安全阀(MSSV)始终保持关闭状态,从而不会旁通安全壳。每个工况的堆芯补水箱水位均未出现下降,不会产生自动卸压信号。即使假设MSSV卡开,堆芯也从未出现裸露,仍保持可冷却状态。     

基于边际谱的功率谱估计方法

针对传统傅里叶谱估计方法不适合非平稳信号和对短信号谱分析能力弱的问题,提出一种基于边际谱的功率谱估计方法,给出了边际谱的求解过程,证明了边际谱的线性性质,并在与Fourier谱比较的基础上给出了边际谱的物理意义和频率分辨率.详细讨论了谱估计的算法和步骤,并用短信号和非平稳信号进行了仿真验证.理论与仿真结果表明:在短信号...