Prediction of residual stress in the welding zone of dissimilar metals using data-based models and uncertainty analysis

Since welding residual stress is one of the major factors in the generation of primary water stress-corrosion cracking (PWSCC), it is essential to examine the welding residual stress to prevent PWSCC. Therefore, several artificial intelligence methods have been developed and studied to predict these residual stresses. In this study, three data-based models, support vector regression (SVR), fuzzy neural network (FNN), and their combined (FNN + SVR) models were used to predict the residual stress for dissimilar metal welding under a variety of welding conditions. By using a subtractive clustering (SC) method, informative data that demonstrate the characteristic behavior of the system were selected to train the models from the numerical data obtained from finite element analysis under a range of welding conditions. The FNN model was optimized using a genetic algorithm. The statistical and analytical uncertainty analysis methods of the models were applied, and their uncertainties were evaluated using 60 sampled training and optimization data sets, as well as a fixed test data set.     

Calculation of the power peaking factor in a nuclear reactor using support vector regression models

Local power density (LPD) at the hottest part of a hot nuclear fuel rod should be estimated accurately to confirm that the rod does not melt. The power peaking factor (PPF) is defined as the highest LPD divided by the average power density in the reactor core. In this paper, the PPF is calculated by support vector regression (SVR) models using numerous measured signals from the reactor cooling system. SVR models are regression analysis models using a kernel function for artificial neural networks. Their neural network weights are found by solving a quadratic programming problem under linear constraints. SVR models are trained using a training data set and then verified against another test data set. The proposed SVR models were applied to the first fuel cycle of the Yonggwang nuclear power plant unit 3. The root mean square errors of the SVR model, with and without in-core neutron flux sensor signal inputs, were 0.1113% and 0.0968%, respectively. This level of errors is sufficiently low for use in LPD monitoring.     

Prediction of critical heat flux for water flow in vertical round tubes using support vector regression model

This paper deals with ν-support vector regression (ν-SVR) based prediction model of critical heat flux (CHF) for water flow in vertical round tubes. The dataset used in this paper is obtained from available published literature. The dataset is partitioned into two independent sets, a training data set and a test data set, to avoid overfitting problem. To train the ν-SVR models with more informative data, the training data is selected using a subtractive clustering (SC) scheme, and then the remaining data is used as test data to evaluate the performance of the ν-SVR models. Next, the parametric trends of CHF are investigated using the ν-SVR models. The results obtained from the ν-SVR models are compared with those obtained from the radial basis function (RBF) network, which is a kind of artificial neural networks (ANNs). It is found that the results of the ν-SVR models are not only in better agreement with the experimental data than those of the RBF network, but also follow the general understanding. The analysis results indicate that the ν-SVR models can be successfully applied to CHF prediction.     

核电厂非能动余热排出系统的失效概率评价

先进核电厂设计中大量采用非能动安全系统提高反应堆安全性。但目前尚无系统性评价非能动系统的成熟方法,而且概率安全评价(PSA)也未考虑非能动系统自然循环现象不确定性导致的功能失效。在欧盟非能动系统可靠性评价研究项目(RMPS)研究成果的基础上,以压水堆二次侧非能动余热排出系统(PRS)为研究对象,基于统计学和热工水力计算确定了影响性能的参数重要度,进而利用蒙特卡罗抽样和响应面分析对全厂断电事故下的PRS自然循环失效概率进行了量化分析评价。初步评价结果表明:非能动系统功能失效概率为2.14×10-3,在PSA中应当充分考虑各种非能动系统的功能失效。本文的评价方法还可以为非能动安全系统设计优化提供支持。     

Calibration curve and support vector regression methods applied for quantification of cement raw meal using laser-induced breakdown spectroscopy

Laser-induced breakdown spectroscopy(LIBS) is a qualitative and quantitative analytical technique with great potential in the cement industrial analysis. Calibration curve(CC) and support vector regression(SVR) methods coupled with LIBS technology were applied for the quantification of three types of cement raw meal samples to compare their analytical concentration range and the ability to reduce matrix effects, respectively. To reduce the effects of fluctuations of the pulse-to-pulse, the unstable ablation and improve the reproducibility, all of the analysis line intensities were normalized on a per-detector basis. The prediction results of the elements of interest in the three types of samples, Ca, Si, Fe, Al, Mg, Na, K and Ti, were compared with the results of the wet chemical analysis. The average relative error(ARE),relative standard deviation(RSD) and root mean squared error of prediction(RMSEP) were employed to investigate and evaluate the prediction accuracy and stability of the two prediction methods. The maximum average ARE of the CC and SVR methods is 34.62% instead of 6.13%,RSD is 40.89% instead of 7.60% and RMSEP is 1.34% instead of 0.43%. The results show that SVR method can accurately analyze samples within a wider concentration range and reduce the matrix effects, and LIBS coupled with it for a rapid, stable and accurate quantification of different types of cement raw meal samples is promising.     

Residual stress analysis of an Overlay weld and a repair weld on the dissimilar Butt weld

Both the experiment and FE analysis were performed to estimate the residual stresses at the parts of the dissimilar metal welds. The specimen of the dissimilar Butt welds was manufactured, and the residual stresses of this specimen were measured by the X-ray method and a Hole Drilling Technique. The values measured by two experimental methods showed a big deviation at the SUS 316L plate. Consequently, the experimental methods to estimate the residual stresses are not a superior method. The Butt FEM Model on this specimen was developed and analyzed by the ABAQUS Code. The results of the FE analysis were compared with those of the experimental methods. As a whole, the values of the Butt FEM Model showed a trend which was in agreement with the experimental values and the values of FE analysis were found reasonable. The Repair FEM Model and the Overlay FEM Model were developed and analyzed by the ABAQUS Code. The values of these results were also found reasonable data even if the experimental methods be not performed. Therefore, the residual stresses for the dissimilar metal welds can be estimated by an analysis with an appropriate FEM Model without the experimental methods.     

In-reactor stress relaxation of selected metals and alloys at low temperatures

Sttess relaxation of bent beam specimens under fast neutron irradiation at 340 and 570 K has been studied for a range of materials, as follows: several stainless steels, a maraged steel, AISI4140, Ni, Inconel X-750, Ti, Zircaloy-2, Zr-2.5% Nb and Zr₃ Al. All specimens were in the annealed or solution-treated condition. Where comparisons were possible, the creep coefficients derived from the stress relaxation tests were found to be consistent with other studies of irradiation-induced creep. The steels showed the lowest rates of stress relaxation; the largest rates were observed with Zr-Nb, Ti and Ni. For most materials, the creep coefficient at 340 K was equal to or greater than that at 570 K. Such weak temperature dependence is not easily reconciled with existing models of irradiation creep based on dislocation climb, such as SIPA or climb-induced glide. Rate theory calculations indicate that because the vacancy mobility becomes very low at the lower temperature, recombination should dominate point defect annealing, resulting in a very low creep rate compared to that at the higher temperature. It is shown that the weak temperature dependence observed experimentally cannot be accounted for by the inclusion of more mobile divacancies in the calculation.     

某核电厂液体贮存罐泄漏~3H的迁移计算

运用Visual modflow建立三维地下水流场,选用某砂岩介质的核电厂址,模拟3H在其中的迁移。模拟结果表明:3H在砂岩介质迁移速度较慢,约经过8 a其浓度峰迁移到河流中。3H在水平方向以迁移为主,在垂直方向迁移很小。     

Simultaneous and multi-criteria optimization of TS requirements and maintenance at NPPs

One of the main concerns of the nuclear industry is to improve the availability of safety-related systems at nuclear power plants (NPPs) to achieve high safety levels. The development of efficient testing and maintenance has been traditionally one of the different ways to guarantee high levels of systems availability, which are implemented at NPP through technical specification and maintenance requirements (TS&M). On the other hand, there is a widely recognized interest in using the probabilistic risk analysis (PRA) for risk-informed applications aimed to emphasize both effective risk control and effective resource expenditures at NPPs. TS&M-related parameters in a plant are associated with controlling risk or with satisfying requirements, and are candidate to be evaluated for their resource effectiveness in risk-informed applications. The resource versus risk-control effectiveness principles formally enter in optimization problems where the cost or the burden for the plant staff is to be minimized while the risk or the availability of the safety equipment is constrained to be at a given level, and vice versa. Optimization of TS&M has been found interesting from the very beginning. However, the resolution of such a kind of optimization problem has been limited to focus on only individual TS&M-related parameters (STI, AOT, PM frequency, etc.) and/or adopting an individual optimization criterion (availability, costs, plant risks, etc.). Nevertheless, a number of reasons exist (e.g. interaction, similar scope, etc.) that justify the growing interest in the last years to focus on the simultaneous and multi-criteria optimization of TS&M. In the simultaneous optimization of TS&M-related parameters based on risk (or unavailability) and cost, like in many other engineering optimization problems, one normally faces multi-modal and non-linear objective functions and a variety of both linear and non-linear constraints. Genetic algorithms (GAs) have proved their capability to solve these kinds of problems, although GAs are essentially unconstrained optimization techniques that require adaptation for the intended constrained optimization, where TS&M-related parameters act as the decision variables. This paper encompasses, in Section 2, the problem formulation where the objective function is derived and constraints that apply in the simultaneous and multi-criteria optimization of TS&M activities based on risk and cost functions at system level. Fundamentals of a steady-state GA (SSGA) as an optimization method is given in Section 3, which satisfies the above requirements, paying special attention to its use in constrained optimization problems. A simple case of application is provided in Section 4, focussing on TS&M-related parameters optimization for a stand-by safety-related system, which demonstrates how the SSGA-based optimization approach works at the system level, providing practical and complete alternatives beyond only mathematical solutions to a particular parameter. Finally, Section 5 presents our conclusions.     

Predicting welding residual stresses in a dissimilar metal girth welded pipe using 3D finite element model with a simplified heat source

Dissimilar metal welds are commonly used in nuclear power plants to connect low alloy steel components and austenitic stainless steel piping systems. The integrity assessment and life estimation for such welded structures require consideration of residual stresses induced by manufacturing processes. Because the fabrication process of dissimilar metal weld joints is considerably complex, it is very difficult to accurately predict residual stresses. In this study, both numerical simulation technology and experimental method were used to investigate welding residual stress distribution in a dissimilar metal pipe joint with a medium diameter, which were performed by a multi-pass welding process. Firstly, an experimental mock-up was fabricated to measure the residual stress distributions on the inside and the outside surfaces. Then, a time-effective 3-D finite element model was developed to simulate welding residual stresses through using a simplified moving heat source. The simplified heat source method could complete the thermo-mechanical analysis in an acceptable time, and the simulation results generally matched the measured data near the weld zone. Through comparing the simulation results and the experimental measurements, we can infer that besides the multi-pass welding process other key manufacturing processes such as cladding, buttering and heat treatment should also be taken into account to accurately predict residual stresses in the whole range of the dissimilar metal pipe.     

Residual stress in a laser welded EUROFER blanket module assembly using non-destructive neutron diffraction techniques

Whilst the structural integrity and lifetime considerations in welded joints for blanket modules can be predicted using finite element software, it is essential to prove the validity of these simulations. This paper provides detailed analysis for the first time, of the residual stress state in a laser-welded sample with integral cooling channels. State-of-the-art non-destructive neutron diffraction was employed to determine the triaxial stress state and to understand microstructural changes around the heat affected zone. Synchrotron X-ray diffraction was used to probe the variation of strain-free lattice reference parameter around the weld zone allowing correction of the neutron measurements. This paper details an important experimental route to validation of predicted stresses in complex safety-critical reactor components for future applications.     

Thermal stress prediction in divertor cooling finger using the local heat transfer distribution

An approach that combines computational fluid dynamics (CFD) and structural mechanics simulation is used for evaluating stresses in a divertor cooling finger. Local distributions of heat transfer coefficient (HTC) between the helium and inner surface of the thimble are used as a boundary condition for the structural mechanics analysis. Stresses calculated with a realistic non-homogeneous HTC distribution are compared with the results where averaged HTC values in three representative zones of inner thimble surface are used. The obtained maximum stresses are considerably higher in the case of realistic non-homogeneous HTC boundary condition. The highest thermal stresses in the tile–thimble assembly are obtained on the thimble inner surface, in the region where the highest thermal gradients due to jet cooling can be observed. The results also show that tetrahedral based finite element model underestimate the computed stresses, especially when a coarse mesh is used.     

CPR1000核电站堆芯出口冷却剂过冷度测量不确定度评定

中国百万千瓦级先进压水堆(CPR1000)核电站反应堆通过堆芯冷却监测系统(CCMS)测量堆芯出口冷却剂的过冷度。本文分析了堆芯出口冷却剂过冷度测量过程中的各种误差来源,对饱和状态下堆芯出口冷却剂温度测量的不确定度进行评定,得到不确定度区间边界随一回路压力变化的曲线,给出了用于判断堆芯冷却状态的堆芯出口冷却剂过冷度测量的误差ε曲线的确定方法,该方法已在CPR1000核电站中得到实际应用。     

Life prediction of steam generator tubing due to stress corrosion crack using Monte Carlo Simulation

The failure of steam generator tubing is one of the main accidents that seriously affects the availability and safety of a nuclear power plant. In order to estimate the probability of the failure, a probabilistic model was established to predict the whole life-span and residual life of steam generator (SG) tubing. The failure investigated was stress corrosion cracking (SCC) after the generation of one through-wall axial crack. Two failure modes called rupture mode and leak mode based on probabilistic fracture mechanics were considered in this proposed model. It took into account the variance in tube geometry and material properties, and the variance in residual stresses and operating conditions, all of which govern the propagations of cracks. The proposed model was numerically calculated by using Monte Carlo Simulation (MCS). The plugging criteria were first verified and then the whole life-span and residual life of the SG tubing were obtained. Finally, important sensitivity analysis was also carried out to identify the most important parameters affecting the life of SG tubing. The results will be useful in developing optimum strategies for life-cycle management of the feedwater system in nuclear power plants.     

Experimental stress analysis at reactor and plant components using hard- and software

Increasing the reliability of pressure vessels the methods of experimental stress analysis are world-wide used for additional data. Components and software programs extend and improve the possibilities of the experimental stress analysis continuously. On-line display of stress results during applications is now possible. Field portable modular equipment with numerous instrument configurations meet the plant and field environmental constraints. The following paper describes a transportable quasistatic multipoint measuring unit with data processing and its use during the first pressure test of a nuclear reactor containment and further applications.     

Generation of nuclear data using Gaussian process regression

ABSTRACT

A new approach for generating nuclear data from experimental cross-section data is presented based on Gaussian process regression. This paper focuses on the generation of nuclear data for proton-induced nuclide production cross-sections with a nickel target. Our results provide reasonable regression curves and corresponding uncertainties and demonstrate that this approach is e?ective for generating nuclear data. Additionally, our results indicate that this approach can be applied in experimental design to reduce the uncertainty of generated nuclear data.     

Remaining life prediction of the core shroud due to stress corrosion cracking failure in BWRs using numerical simulations

Stress corrosion cracking (SCC) of the welded joints in a reactor core shroud is the primary result of the residual stresses caused by welding, corrosion and neutron irradiation in a boiling water reactor (BWR). Therefore, the evaluation of SCC propagation is important for the safe maintenance of the core shroud. This paper attempts to predict the remaining life of the core shroud due to SCC failures in BWR conditions via SCC propagation time calculations. First, a two-dimensional finite element method model containing H6a girth weld in the core shroud was constructed, and the weld processing was simulated to determine the weld's residual stress distribution. Second, using a basic weld residual stress field, the SCC propagation was simulated using a node release option and the stress redistribution was calculated. Combined with the J-integral method, the stress intensity factors were calculated at depths of 2, 3, 4, 8, 12, 16, 19, 22, 25 and 30 mm in the crack setting inside the core shroud; then, the SCC propagation rates were determined using the relation between the SCC propagation rate and the stress intensity factor. The calculations show that the core shroud could safely remain in service after 9.29 years even when a 1-mm-deep SCC has been detected.     

Prediction of residual stress distributions due to surface machining and welding and crack growth simulation under residual stress distribution

In nuclear power plants, stress corrosion cracking (SCC) has been observed near the weld zone of the core shroud and primary loop recirculation (PLR) pipes made of low-carbon austenitic stainless steel Type 316L. The joining process of pipes usually includes surface machining and welding. Both processes induce residual stresses, and residual stresses are thus important factors in the occurrence and propagation of SCC. In this study, the finite element method (FEM) was used to estimate residual stress distributions generated by butt welding and surface machining. The thermoelastic-plastic analysis was performed for the welding simulation, and the thermo-mechanical coupled analysis based on the Johnson-Cook material model was performed for the surface machining simulation. In addition, a crack growth analysis based on the stress intensity factor (SIF) calculation was performed using the calculated residual stress distributions that are generated by welding and surface machining. The surface machining analysis showed that tensile residual stress due to surface machining only exists approximately 0.2 mm from the machined surface, and the surface residual stress increases with cutting speed. The crack growth analysis showed that the crack depth is affected by both surface machining and welding, and the crack length is more affected by surface machining than by welding.     

中国先进研究堆中子残余应力谱仪校准

为校准中国先进研究堆（CARR）中子残余应力谱仪,获得一维位置灵敏探测器（PSD）的灵敏度系数、位置坐标与衍射角的几何关系及零点等参数,设计了校准实验和软件,并将获得的校准参数用于标准Fe粉前4个衍射峰的数据预处理,校正衍射峰位与计算衍射角的差值符合残余应力谱仪的误差要求。结果表明,该工作能准确获得校准参数,并可用于CARR中子残余应力谱仪的数据预处理。