Sobol′ Sensitivity Analysis Using a Neural Network Model of a LB-LOCA in the ZION Nuclear Power Plant With CATHARE-2 V2.5 Code

2013 ◽  
Author(s):  
Fabrice Fouet ◽  
Pierre Probst
Keyword(s):  
Neural Network ◽  
Sensitivity Analysis ◽  
Nuclear Power ◽  
Uncertainty Propagation ◽  
Correlation Coefficients ◽  
Computer Code ◽  
Numerical Application ◽  
Loss Of Coolant Accident ◽  
Input Parameters ◽  
Artificial Neural Network Ann

In nuclear safety, the Best-Estimate (BE) codes may be used in safety demonstration and licensing, provided that uncertainties are added to the relevant output parameters before comparing them with the acceptance criteria. The uncertainty of output parameters, which comes mainly from the lack of knowledge of the input parameters, is evaluated by estimating the 95% percentile with a high degree of confidence. IRSN, technical support of the French Safety Authority, developed a method of uncertainty propagation. This method has been tested with the BE code used is CATHARE-2 V2.5 in order to evaluate the Peak Cladding Temperature (PCT) of the fuel during a Large Break Loss Of Coolant Accident (LB-LOCA) event, starting from a large number of input parameters. A sensitivity analysis is needed in order to limit the number of input parameters and to quantify the influence of each one on the response variability of the numerical model. Generally, the Global Sensitivity Analysis (GSA) is done with linear correlation coefficients. This paper presents a new approach to perform a more accurate GSA to determine and to classify the main uncertain parameters: the Sobol′ methodology. The GSA requires simulating many sets of parameters to propagate uncertainties correctly, which makes of it a time-consuming approach. Therefore, it is natural to replace the complex computer code by an approximate mathematical model, called response surface or surrogate model. We have tested Artificial Neural Network (ANN) methodology for its construction and the Sobol′ methodology for the GSA. The paper presents a numerical application of the previously described methodology on the ZION reactor, a Westinghouse 4-loop PWR, which has been retained for the BEMUSE international problem [8]. The output is the first maximum PCT of the fuel which depends on 54 input parameters. This application outlined that the methodology could be applied to high-dimensional complex problems.

Sensitivity Analysis by the Use of a Surrogate Model During Large Break LOCA on ZION Nuclear Power Plant With CATHARE-2 V2.5 Code

2009 ◽  
Author(s):  
Fabrice Fouet ◽  
Pierre Probst
Keyword(s):  
Sensitivity Analysis ◽  
Power Plant ◽  
Nuclear Power Plant ◽  
Surrogate Model ◽  
Nuclear Power ◽  
Uncertainty Propagation ◽  
Correlation Coefficients ◽  
General Rule ◽  
Computer Code ◽  
Input Parameters

In nuclear safety, the Best-Estimate (BE) codes may be used in demonstration and licensing, provided that uncertainties are added to the relevant output parameters before comparing them with the acceptance criteria. The uncertainty of output parameters, which comes mainly from the lack of knowledge of the input parameters, is evaluated by estimating the 95% percentile with a high degree of confidence. IRSN, technical support of the French Safety Authority, develop a method of uncertainty propagation and chose to apply it to the calculation of the Peak Cladding Temperature (PCT) with CATHARE-2 V2.5 code during a Large Break (LB) LOCA event for ZION, a 4-loop PWR of Westinghouse design. As a general rule the Global Sensitivity Analysis (GSA) is done with linear correlation coefficients. This paper presents a new approach to perform a more accurate GSA to determine and to classify the main uncertain parameters: the SOBOL methodology. This technique requires simulating many sets of parameters to propagate uncertainties correctly, which makes of it a time-consuming approach. Therefore, it is natural to replace the complex computer code by an approximate mathematical model, called response surface or surrogate model. Kriging methodology (with simulated annealing optimization) for its construction and the SOBOL methodology for the GSA are used. The paper presents the application of the previously described methodology on a LB-LOCA scenario in ZION reactor, associated with 54 input parameters. The output is the first maximum peak cladding temperature of the fuel. Results show that the methodology could be applied to both high-dimensional complex problems and real nuclear power plant calculations.

scholarly journals Prediction Model of Iron Ore Pellet Ambient Strength and Sensitivity Analysis on the Influence Factors

Metals ◽  
2018 ◽  
Vol 8 (8) ◽  
pp. 593 ◽  
Author(s):  
Qiangjian Gao ◽  
Yingyi Zhang ◽  
Xin Jiang ◽  
Haiyan Zheng ◽  
Fengman Shen
Keyword(s):  
Neural Network ◽  
Sensitivity Analysis ◽  
Prediction Model ◽  
Bp Neural Network ◽  
Back Propagation ◽  
Influence Factors ◽  
Principal Component ◽  
Bp Network ◽  
Input Variables ◽  
Artificial Neural Network Ann

The Ambient Compressive Strength (CS) of pellets, influenced by several factors, is regarded as a criterion to assess pellets during metallurgical processes. A prediction model based on Artificial Neural Network (ANN) was proposed in order to provide a reliable and economic control strategy for CS in pellet production and to forecast and control pellet CS. The dimensionality of 19 influence factors of CS was considered and reduced by Principal Component Analysis (PCA). The PCA variables were then used as the input variables for the Back Propagation (BP) neural network, which was upgraded by Genetic Algorithm (GA), with CS as the output variable. After training and testing with production data, the PCA-GA-BP neural network was established. Additionally, the sensitivity analysis of input variables was calculated to obtain a detailed influence on pellet CS. It has been found that prediction accuracy of the PCA-GA-BP network mentioned here is 96.4%, indicating that the ANN network is effective to predict CS in the pelletizing process.

Sensitivity Analysis for Dynamical Response of Reactor Coolant System Based on OPTIMUS

2020 ◽  
Author(s):  
Yuan Yanli ◽  
Ye Xianhui ◽  
Li Lijuan ◽  
Yuan Feng
Keyword(s):  
Sensitivity Analysis ◽  
Steam Generator ◽  
Nuclear Power ◽  
Reactor System ◽  
Response Analysis ◽  
Dynamical Response ◽  
Calculation Error ◽  
Reactor Coolant ◽  
Input Parameters ◽  
Coolant System

Abstract The sensitivity analysis of the dynamical response of reactor coolant system to the input parameters is an important precondition for the design optimization. In this paper, the sensitivity of the dynamical loads at the nozzles of the equipment under seismic conditions is analyzed with an integrated platform called OPTIMUS, taking the stiffness of the dampers in the steam generator and the main pump as the input variables. The key parameters of the reactor system are usually different from the design value due to the calculation error, random and other uncontrollable errors in the manufacturing process and installation process. In a nuclear power project, the measured stiffness values of the dampers on the steam generator and the main pump in the manufacturer are deviated from the requirements in the equipment specification, and it is necessary to evaluate the influence of the deviation on the dynamical response analysis of the reactor system. According to the traditional method, it is necessary to establish the models of the reactor coolant system for nonlinear analysis according to the different stiffness of the dampers, and then the calculation results are compared by EXCEL. In this paper, the sensitivity analysis of output parameters which are the loads at the nozzles of the equipment to the input parameters which are the stiffness of the dampers on the steam generator and pump is realized by OPTIMUS, which is a kind of integration platform. Not only can ANSYS simulation calculations be carried out automatically on the OPTIMUS, but also the output data can be processed rapidly automatically, and the influence of manufacturing deviation of the stiffness of the dampers on the dynamical response of the reactor coolant system can be analyzed quantitatively in the above-mentioned problems, and the data support is provided for the determination of the design variables for subsequent optimization analysis.

Uncertainty Analysis of Scaling Calculations From LSTF Small Break LOCA Tests With Steam Generator Intentional Depressurization Applying to a Four-Loop PWR

2018 ◽  
Author(s):  
Ikuo Kinoshita
Keyword(s):  
Steam Generator ◽  
Uncertainty Propagation ◽  
Correlation Coefficients ◽  
Injection System ◽  
Plant Model ◽  
Input Uncertainty ◽  
Loss Of Coolant Accident ◽  
Output Uncertainty ◽  
Management Procedure ◽  
Uncertainty Parameters

The Best Estimate Plus Uncertainty (BEPU) method is applied to analysis of the “intentional depressurization of steam generator secondary side” which is an accident management procedure in a small break loss-of-coolant accident with high pressure injection system failure. In the present study, scaling calculations from the LSTF small break LOCA tests were carried out for a conventional Westinghouse type four-loop PWR. The two test cases were selected with different break size and different depressurization conditions to ensure the reliability for the analyses for the accident scenario. The uncertainty propagation analyses were performed for the PWR. The dominating input uncertainty parameters, i.e. those few with the highest influence on the output uncertainty were identified. The analysis results were compared with those for the LSTF to address the scaling up capability and the similarity evaluation between the LSTF model and the PWR plant model. It was found that the PWR plant model results had overall agreements with the LSTF model, and the uncertainty of the predicted PCT included the measured PCT. Furthermore, the correlation coefficients between the input uncertainty parameters and the PCT for the PWR plant model had similarities with those for the LSTF model.

Prediction of Surface SO2 Concentration in Shanghai Using Artificial Neural Network

2014 ◽  
Vol 522-524 ◽  
pp. 44-47 ◽  
Author(s):  
Dan Xue ◽  
Qian Liu
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Daily Precipitation ◽  
Correlation Coefficients ◽  
Ann Model ◽  
Sea Level Pressure ◽  
Neuron Number ◽  
Artificial Neural ◽  
Artificial Neural Network Ann ◽  
Lm Algorithm

Air quality has been deteriorated seriously in Shanghai as a result of urbanization and modernization. A three-layer Artificial Neural Network (ANN) model was developed to forecast the surface SO2 concentration. The subsequent SO2 concentration being the output parameter of this study was estimated by six input parameters such as preceding SO2 concentrations, average daily temperature, sea-level pressure, relative humidity, average daily wind speed and average daily precipitation. Levenberg-Marquarde (LM) backpropagation was tested as the best algorithm and the optimal neuron number for the LM algorithm was found to be eight. ANN testing outputs were proven to be satisfactory with correlation coefficients of about 0.765.

A Review of Neural Network Application for Fault Diagnosis of NPP

2009 ◽  
Author(s):  
He Wang
Keyword(s):  
Neural Network ◽  
Fault Diagnosis ◽  
Nuclear Power ◽  
Layer Structure ◽  
Engineering Application ◽  
Model Performance ◽  
Training Sample ◽  
Feed Forward Neural Network ◽  
Artificial Neural Network Ann ◽  
Hidden Layer

Artificial Neural Network (ANN) with its self-learning capabilities, nonlinear mapping ability and generalization ability, has been widely applied for fault diagnosis of complex system like Nuclear Power Plant (NPP). In this paper, an overview of the application of supervised multi-layer feed-forward neural network for fault diagnosis of NPP is presented, including the following aspects: the acquisition of the training sample data, the determination of appropriate input and output data, the choice of hidden layer structure and the evaluation of network model performance. Finally, a number of key issues about the engineering application of neural network fault diagnosis in practice were discussed.

scholarly journals Assessing the Impact of Expert Labelling of Training Data on the Quality of Automatic Classification of Lithological Groups Using Artificial Neural Networks

2020 ◽  
Vol 25 (2) ◽  
pp. 145-152
Author(s):  
Yan Kuchin ◽  
Ravil Mukhamediev ◽  
Kirill Yakunin ◽  
Janis Grundspenkis ◽  
Adilkhan Symagulov
Keyword(s):  
Neural Network ◽  
Automatic Classification ◽  
Training Data ◽  
Classification Problems ◽  
Expert Opinions ◽  
Input Parameters ◽  
Artificial Neural ◽  
Artificial Neural Network Ann ◽  
The Impact

AbstractMachine learning (ML) methods are nowadays widely used to automate geophysical study. Some of ML algorithms are used to solve lithological classification problems during uranium mining process. One of the key aspects of using classical ML methods is causing data features and estimating their influence on the classification. This paper presents a quantitative assessment of the impact of expert opinions on the classification process. In other words, we have prepared the data, identified the experts and performed a series of experiments with and without taking into account the fact that the expert identifier is supplied to the input of the automatic classifier during training and testing. Feedforward artificial neural network (ANN) has been used as a classifier. The results of the experiments show that the “knowledge” of the ANN of which expert interpreted the data improves the quality of the automatic classification in terms of accuracy (by 5 %) and recall (by 20 %). However, due to the fact that the input parameters of the model may depend on each other, the SHapley Additive exPlanations (SHAP) method has been used to further assess the impact of expert identifier. SHAP has allowed assessing the degree of parameter influence. It has revealed that the expert ID is at least two times more influential than any of the other input parameters of the neural network. This circumstance imposes significant restrictions on the application of ANNs to solve the task of lithological classification at the uranium deposits.

Improving the Mechanical Properties of Wire-Rope Silk Scaffold by Artificial Neural Network in Tendon and Ligament Tissue Engineering

2015 ◽  
Vol 10 (3) ◽  
pp. 155892501501000 ◽  
Author(s):  
Elham Naghashzargar ◽  
Dariush Semnani ◽  
Saeed Karbasi
Keyword(s):  
Neural Network ◽  
Mechanical Properties ◽  
Artificial Neural Network ◽  
Back Propagation ◽  
Wire Rope ◽  
Elongation At Break ◽  
Input Parameters ◽  
Artificial Neural ◽  
Artificial Neural Network Ann ◽  
Silk Yarn

Finding an appropriate model to assess and evaluate mechanical properties in tissue engineered scaffolds is a challenging issue. In this research, a structurally based model was applied to analyze the mechanics of engineered tendon and ligament. Major attempts were made to find the optimum mechanical properties of silk wire-rope scaffold by using the back propagation artificial neural network (ANN) method. Different samples of wire-rope scaffolds were fabricated according to Taguchi experimental design. The number of filaments and twist in each layer of the four layered wire-rope silk yarn were considered as the input parameters in the model. The output parameters included the mechanical properties which consisted of UTS, elongation at break, and stiffness. Finally, sensitivity analysis on input data showed that the number of filaments and the number of twists in the fourth layer are less important than other input parameters.

scholarly journals Forecasting Wastewater Temperature Based on Artificial Neural Network (ANN) Technique and Monte Carlo Sensitivity Analysis

2020 ◽  
Vol 12 (16) ◽  
pp. 6386 ◽  
Author(s):  
Farzin Golzar ◽  
David Nilsson ◽  
Viktoria Martin
Keyword(s):  
Neural Network ◽  
Sensitivity Analysis ◽  
Artificial Neural Network ◽  
Monte Carlo ◽  
Wastewater Treatment ◽  
Wastewater Treatment Plant ◽  
Heat Recovery ◽  
Historical Data ◽  
Treatment Plant ◽  
Artificial Neural Network Ann

Wastewater contains considerable amounts of thermal energy. Heat recovery from wastewater in buildings could supply cities with an additional source of renewable energy. However, variations in wastewater temperature influence the performance of the wastewater treatment plant. Thus, the treatment is negatively affected by heat recovery upstream of the plant. Therefore, it is necessary to develop more accurate models of the wastewater temperature variations. In this work, a computational model based on artificial neural network (ANN) is proposed to calculate wastewater treatment plant influent temperature concerning ambient temperature, building effluent temperature and flowrate, stormwater flowrate, infiltration flowrate, the hour of day, and the day of year. Historical data related to the Stockholm wastewater system are implemented in MATLAB software to drive the model. The comparison of calculated and observed data indicated a negligible error. The main advantage of this ANN model is that it only uses historical data commonly recorded, without any requirements of field measurements for intricate heat transfer models. Moreover, Monte Carlo sensitivity analysis determined the most influential parameters during different seasons of the year. Finally, it was shown that installing heat exchangers in 40% of buildings would reduce 203 GWh year−1 heat loss in the sewage network. However, heat demand in WWTP would be increased by 0.71 GWh year−1, and the district heating company would recover 176 GWh year−1 less heat from treated water.

Sign in / Sign up

Export Citation Format

Share Document