核动力蒸汽发生器水位控制方法分析

核动力蒸汽发生器(SG)是一个高度复杂的非线性时变系统.SG在瞬态、启动和低功率运行工况下的"收缩"与"膨胀"现象引起的逆动力学效应使SG的水位控制变得复杂.文章分析了SG水位控制方法的特点,重点分析了SG水位模糊控制方法与神经网络控制方法.指出了传统的PI(D)水位控制方法存在的问题,就SG水位控制发展趋势提出了看法.     

基于elman神经网络的蒸汽发生器水位重构

蒸汽发生器(SG)水位指示仪表出现虚假指示或者丧失指示的情况时有发生,严重影响操纵员对核动力装置运行情况的判断。elman神经网络是典型的动态神经网络,在处理复杂非线性对象中能直接反映动态过程系统的特性。本文提出用elman神经网络重构蒸汽发生器水位的方法,以主蒸汽管道破口事故下重构蒸汽发生器水位为例建模求解,与仿真数据进行对比,结果表明elman神经网络对SG水位重构的相对误差小、精度高,能满足实际需要。     

核电厂蒸汽发生器的容错控制

基于考虑了随机噪声的蒸汽发生器非线性模型,利用一组扩展卡尔曼滤波器对传感器的状态进行监测,通过实时检测、分离单传感器故障并重构传感器的输出值,达到蒸汽发生器水位稳定的容错控制目的。结果表明,基于扩展卡尔曼滤波器组对蒸汽发生器进行容错控制设计的方法是可行的。     

应用H∞最优线性滤波器估计时变反应性

本文应用一种Ｈ∞整定的线怀滤波方法，解决了由中子注量测量值来估计点堆动力学模型时变堆芯反应性的问题。为应用该方法，建立了一个相应的动态反应性模型，该模型可作为一个状态变量，嵌入到反应堆模型之中，通过使估计误差功率谱的Ｈ∞范数为最小的滤波器对被噪声污染了中子密度的测量值进行运算。给出动态堆芯反尖性的估算值。为了得向这种Ｈ∞最优滤波器的基本特性，进行了一些计算机仿真，仿真的结果表明，该滤波器可用于通过     

蒸汽发生器水位双PI控制的改进研究

蒸汽发生器（SG）水位的控制问题是压水堆核电站紧急停堆的1个重要原因,尤其在低功率情况下,缩胀效应致使水位控制较难。为了避开水位变化初期的假水位的影响,在控制过程中加入判别器来决定水位信号是否参与控制。采用遗传算法对控制器参数进行整定,采用的SG水位模型是简化的数学模型。对控制器中有无判别器的控制效果进行了对比。对比结果表明：加入判别器后,缩短了调节时间,降低了超调量。     

基于神经网络的蒸汽发生器过程辨识方法研究

在核电站运行过程中,由于蒸汽流量随负荷变化,蒸汽发生器内沸腾区域的气泡数量因局部压力变化而变化,水位呈现瞬时"虚假水位"现象,给蒸汽发生器的水位特性辨识带来困难.如果处理不当,就会严重影响核电站的安全运行.为了提高蒸汽发生器水位特性的辨识效果,对基于神经网络的蒸汽发生器水位辨识方法进行了研究.辨识模型采用串-并联型辨识结构.网络训练采用Levenberg-Marququardt BP学习算法(LMBP).仿真结果表明,所提出的方法具有良好的辨识性能.     

蒸汽发生器双级模糊积分智能水位控制器设计

为了实现对蒸汽发生器水位的有效控制,设计了双级模糊积分智能水位控制器.在常规模糊控制器之前,增加了"假水位"判别模糊控制器,以实现在"假水位"阶段对水位进行有效控制.为实现对水位的最优控制,将模糊控制器和积分控制器结果协调输出.通过对控制响应曲线的分析,验证本论文所设计的双级模糊积分智能控制器能有效克服"假水位"现象对水位控制带来的不利影响,且控制超调量小,稳定时间短.     

蒸汽发生器水位全程控制系统数字化及仿真实现

采用单冲量和三冲量的水位控制方案设计了蒸汽发生器(SG)水位的全程数字化控制系统,提出一套利用软件模块组态的方法,实现了水位控制策略。并将此方案应用于核电仿真机的运行。仿真结果曲线表明,设计的控制方案能使SG水位在稳定工况时保持恒定;变负荷时,水位能随着负荷的变化而产生变化并最终保持在恒定值上。     

基于结构奇异值综合的蒸汽发生器水位鲁棒控制器设计

针对U型管蒸汽发生器的给水阀开度-水位动力学特性随其本身和给水泵运行工况改变而变化的特性,建立了能定量描述上述原因导致的参数摄动和未建模动态的给水阀开度-水位对象的不确定状态空间模型。基于结构奇异值综合方法设计蒸汽发生器水位鲁棒控制器,并通过选择合适的性能加权函数优化控制系统的鲁棒性能。仿真结果表明,所设计控制器可满足蒸汽发生器从零负荷至满负荷及各种给水泵运行工况下水位自动控制的要求,具有较好的控制品质。     

池式堆供热系统建模与仿真方法研究

以低温、常压的池式堆供热系统为研究对象,经过系统划分和对实际物理设备进行合理简化和假设后,建立了覆盖其基本功能的动态数学模型。该模型包含点堆中子动力学模型、热工水力学模型及堆外热力设备模型。本文基于vPower仿真平台对该模型进行了论证,仿真结果表明该模型能正确反映池式堆供热系统的动态特性。该仿真模型可进一步用于研究运行工况及验证设计数据的合理性。     

基于SPSA的蒸汽发生器液位MPC系统性能优化方法研究

核电厂蒸汽发生器（SG）液位变化过程具有强非线性且存在“虚假水位”现象,传统SG液位控制系统多采用固定参数比例-积分-微分（PID）控制器,但传统PID控制方法不具备自优化、自适应、自学习等能力,使得控制系统性能难以达到并保持最佳。为提高机组瞬态响应能力以及核电厂的稳定性、安全性和经济性,提出了一种基于并行摄动随机逼近（SPSA）算法的模型预测控制（MPC）方法。该方法采用MPC系统代替传统PID控制系统,并利用SPSA实现液位控制系统参数的整定优化,从而实现SG液位控制系统的性能优化。通过仿真试验验证了本方法能够有效提高SG液位控制参数的整定效率以及控制系统稳定性。     

Adaptive estimator-based dynamic sliding mode control for the water level of nuclear steam generators

Steam Generator (SG) is a crucial component of nuclear power plant. The proper water level control of a nuclear steam generator is of great importance in order to secure the sufficient cooling source of the nuclear reactor and to prevent damage of turbine blades. The water level control problem of steam generators has been a main cause of unexpected shutdowns of nuclear power plants which must be considered for plant safety and availability. The control problem is challenging, especially at low power levels due to shrink and swell phenomena and flow measurement errors. Moreover, the dynamics of steam generator vary as the power level changes. Therefore, it is necessary to improve the water level control system of SG. In this paper, an adaptive estimator-based dynamic sliding mode control method is developed for the level control problem. The proposed method exhibits the desired dynamic properties during the entire output tracking process independent of perturbations. Simulation results are presented to demonstrate the effectiveness of the proposed controller in terms of performance, robustness and stability. Simulation results confirm the improvement in transient response obtained by using the proposed controller.     

Gain scheduled dynamic sliding mode control for nuclear steam generators

U-Tube Steam Generator (UTSG) is one of the most important facilities in a pressurized-water nuclear reactor. Poor control of the Steam Generator (SG) water level in the secondary circuit of a nuclear power plant can lead to frequent reactor shutdowns or damage of turbine blades. The control problem is challenging, especially at low power levels due to shrink and swell phenomena and flow measurement errors. In addition, the dynamics of steam generator vary as the power level changes. Therefore, designing a suitable controller for all power levels is a necessary step to enhance the plant availability factor. The purpose of this paper is to design, analyze and evaluate a water level controller for U-tube steam generators using dynamic sliding mode control. The employed method is easy to implement in practical applications and moreover, the dynamic sliding mode control exhibits the desired dynamic properties during the entire output-tracking process independent of perturbations. Gain scheduling is used to obtain a global water level controller. Simulation results are presented to demonstrate the performance, robustness, and stability of the proposed controller.Computer simulations show that the proposed controller improves the transient response of steam generator water level and demonstrates its superiority to existing controllers.     

基于小波神经网络的蒸汽发生器动态过程辨识

在核动力蒸汽发生器(SG)运行过程中，其逆动力学效应使其动态特性难以辨识。为提高蒸汽发生器动态特性辨识的效果，提出了基于小波神经网络的蒸汽发生器动态过程辨识的新方法。辨识模型采用串并联型辨识结构，网络训练采用Levenberg Marququardt学习算法（LMBP）。对蒸汽发生器典型运行工况的辨识结果表明，所提出的方法能够正确地辨识蒸汽发生器的动态特性且具有较高的辨识精度。     

Design of stability-guaranteed neurofuzzy logic controller for nuclear steam generators

A neurofuzzy logic controller (NFLC), which is implemented by using a multilayer neural network with special types of fuzzifier, inference engine and defuzzifier, is applied to the water level control of a nuclear steam generator (SG). This type of NFLC has the structural advantage that arbitrary two-input, single-output linear controllers can be adequately mapped into a set of specific control rules of the NFLC. In order to design a stability-guaranteed NFLC, the stable sector of the given linear gain is obtained from Lyapunov's stability criteria. Then this sector is mapped into two linear rule tables that are used as the limits of NFLC control rules. The automatic generation of NFLC rule tables is accomplished by using the back-error-propagation (BEP) algorithm. There are two separate paths for the error back propagation in the SG. One considers the level dynamics depending on the tank capacity and the other takes into account the reverse dynamics of the SG. The amounts of error back propagated through these paths show opposite effects in the BEP algorithm from each other for the swell-shrink phenomenon. Through computer simulation it is found that the BEP algorithm adequately generates NFLC rule tables according to given learning parameters.     

蒸汽排放系统蒸汽冷凝器动态特性仿真研究

根据核动力商船高压定压蒸汽冷凝器的结构及工作特性,建立了适合蒸汽冷凝器实时仿真分析计算的两相流仿真模型。利用该仿真模型对蒸汽冷凝器进行了稳态计算和动态仿真分析,并将其稳态计算结果与蒸汽冷凝器试验结果进行对比验证。结果表明：该模型能准确模拟蒸汽冷凝器的动态特性,满足核动力装置蒸汽排放系统实时仿真分析要求。研究结果对二回路系统启动及停运的运行和控制系统设计具有指导意义。     

集成神经网络方法在蒸汽发生器故障诊断中的应用

针对蒸汽发生器传统故障检测与诊断方法的不足,提出了基于集成神经网络的蒸汽发生器故障检测与诊断的新方法。该方法采用两个神经网络。一个神经网络作为蒸汽发生器的动力学模型,用于蒸汽发生器的重要运行参数的预测,其原理是通过检测蒸汽发生器运行参数监测信号值与相应的蒸汽发生器神经网络模型预测值之间的偏差来确定是否发生了异常,如果某一参数偏差超过了预先给定的极限,就认为发生了异常。另一个神经网络作为故障分类模型,用以对蒸汽发生器故障进行分类,给出故障的类型。由两个神经网络监测和诊断结果的融合给出蒸汽发生器故障较为清晰的信息。仿真结果表明,该方法能够提高蒸汽发生器监测与诊断的能力。     

CPR1000型核电站给水泵事故瞬态分析

以CPR1000型核电站3×50%电动给水泵为研究对象,采用基于RELAP5和Simulink程序开发的CPR1000数字化仪控系统仿真试验台,详细计算分析了给水泵单泵故障和双重故障对反应堆运行的影响及相应的缓解措施。结果表明,给水泵单泵故障对反应堆运行的影响较小,各相关参数能够很快重回事故前的稳态工况。在给水泵双重故障情况下：初始核功率在75%FP及以下时,不会出现蒸汽发生器(SG)低-低水位;初始核功率高于75%FP、汽机初始负荷在90%FP及以下时,需将汽机负荷阶跃降至50%FP,才不会出现SG低-低水位;汽机初始负荷在90%FP以上时,建议停堆。     

Application of Particle Filtering for Estimating the Dynamics of Nuclear Systems

System dynamics estimation is a crucial issue for the safe operation and control of nuclear power plants. Typically, the estimation is based on a model of the plant dynamics and related measurements. In practice, the non-linearity of the dynamics and non-Gaussianity of the noise associated to the process and measurements lead to inaccurate results even with advanced approaches, such as the Kalman, Gaussian-sum and grid-based filters. On the contrary, accurate results may be obtained with Monte Carlo-based estimation methods, also called particle filters. The present paper illustrates the developments of a previous work by the same authors with regards to the comparison of the so called sampling importance resampling filter method with the standard and extended Kalman filtering techniques. Two case studies are analyzed to separately highlight the effect of non-linearity and non-Gaussianity in the process noise.