Code Input Parameters Research Articles

Sensitivity calculation for Monte Carlo particle simulations of neutrals in the plasma edge

AbstractAlgorithmic differentiation (AD) is a promising tool to accurately and efficiently compute sensitivities of several code outputs w.r.t. the code input parameters for the complex Monte Carlo code EIRENE. We illustrate the procedure by assessing the sensitivity of the total atom content w.r.t. the atomic and molecular cross‐sections and rate coefficients. AD does not suffer from decorrelation between primal and perturbed particle trajectories, which easily happens for finite difference (FD) sensitivity calculations. Consequently, the resulting statistical errors on AD sensitivities are up to five orders of magnitude smaller than the statistical errors on FD sensitivities for a JET low‐recycling background plasma. Although this statistical error reduction is promising for the application of the AD sensitivities in a future optimization or uncertainty quantification framework, we show that the AD sensitivities can blow up for a minority of long‐lived particles, especially occurring in detachment. Finally, we conclude that the Monte Carlo simulation and estimator type possibly need to be adapted to improve the accuracy of the sensitivities.

Comparison of inverse uncertainty quantification methods for critical flow test

A problem of epistemic uncertainties introduced by code input parameters that cannot be estimated other than by user expertise has existed in nuclear reactor systems analyses since first uncertainty analyses were conducted. Inverse Uncertainty Quantification (IUQ) methods aim at providing an estimation of the distributions of such parameters. This paper compares and assesses two approaches to quantify the uncertainty of critical flow model parameters available in the TRACE code. First novel approach is based on a machine-learning algorithm, using a Random Forest classifier to assign the results of calculations to one of the defined classes of prediction accuracy with respect to experimental data. The second approach is based on Markov Chain Monte Carlo sampling and Bayesian inference. Both methods allows to assign probability distribution functions to TRACE internal variables, which is a goal for IUQ methods.Both approaches were used to estimate uncertainties for four critical flow parameters available in the TRACE code based on the Marviken critical flow experiment. In case of two parameters both methods produced similar results of the probability density functions. Based on the results of machine learning two additional parameters were deemed noninfluential, while for the second method such definitive assessment could not be made.

A large reduction of direct aerosol cooling over Poland in the last decades

AbstractThis paper presents an analysis of the long‐term (1982–2015) variability of aerosol optical properties, as well as aerosol and greenhouse gases (GHG) direct radiative forcing (RF) in central Europe on the basis of MERRA‐2 reanalysis and ground‐based observation. Calculations of longwave (LW) and shortwave (SW) RF were made with the use of the off‐line Fu‐Liou radiative transfer model and were preceded by the sensitivity study of the code's input parameters. Then, the long‐term mean as well as the annual variability for selected regions of aerosol optical properties and radiative forcing were analysed. The mean AOD trend was −0.056 ± 9% per decade and AOD was reduced by 48% between 1982 and 2015. While for 1982–1990 the trend per decade was −0.12 ± 20%, for 1991–2000 it was −0.17 ± 17% and only −0.02 ± 31% in the last 15 years (2001–2015). The trend of the aerosol radiative forcing (ARF) was 1.52 ± 0.12 W·m−2/10 year and 1.21 ± 0.19 W·m−2/10 year at top of the atmosphere (TOA) and at Earth's surface, respectively. The trend for GHG was significantly smaller and it equalled 0.27 ± 0.01 W·m−2/10 year at TOA and 0.17 ± 0.01 W·m−2/10 year at the surface. Changes in GHG and aerosol direct effect produced additional 3.2 ± 0.2 W·m−2 at TOA, which could be associated with the observed regional climate warming. The change of the direct aerosol effect was about 4 times the GHG RF. The influence of aerosol loading reduction on the radiation budget was significantly higher in the 1990s of the 20th century in comparison to the first decades of the 21st. The effect of aerosol reduction has significant impact on air temperature changes during warm season and negligible during winter.

The use of machine learning for inverse uncertainty quantification in TRACE code based on Marviken experiment

Inverse Uncertainty Quantification (IUQ) methods are gaining recognition in the application of Best Estimate Plus Uncertainty methodologies. Typically, the code input parameters constituted the sources of uncertainties that cannot be estimated other than by user expertise. The IUQ methods aim at providing a more robust and scientific estimation of the distributions of those parameters. In this paper, an approach has been proposed to use machine learning to quantify the uncertainty of critical flow parameters available in the TRACE code.Serving as the means to the ultimate goal of IUQ, the machine learning algorithm has also been utilized in the qualitative assessment of discrepancies between the experiment and calculations. First, a set of Monte Carlo calculations were performed for selected Marviken Critical Flow Tests with permutated values of input parameters of TRACE code. Then, utilizing a supervised machine learning model and a set of features describing the mass flow progression, a Random Forest classifier was used to assign the results to one of the defined classes of prediction accuracy. This analysis has been done using two different training datasets to present the influence of the training data selection on the classification.The machine learning classification resulted in creating a class for which the results with the best fit to the experiment were collected. Derivation of this class created an opportunity to propose ranges and probability distribution functions for CHM12 and CHM22 TRACE internal variables, which is a goal for IUQ methods. Lognormal distribution has been proposed for CHM12 with a mode equal to 0.87, while normal distribution was proposed for CHM22 with a mean equal to 0.9. Variables C1RC2 and C2RC2 proved to be uninfluential on the results. The PDFs for CHM12 and CHM22 have been verified on all 26 Marviken tests. The article proposes a method that can be used in best-estimate analyses to obtain reliable distributions of input variables.

Coalitional Game-Based Cooperative Computation Offloading in MEC for Reusable Tasks

Mobile-edge computing (MEC) has been a promising solution for Internet-of-Things (IoT) applications to obtain latency reduction and energy savings. In view of the loosely coupled application, multiple devices can use the same task code and different input parameters to obtain diverse results. This motivates us to study the cooperation between devices for eliminating the repeated data transmission. Leveraging coalitional game theory, we formalize the cooperative offloading process of a reusable task into a coalitional game to maximize the cost savings. In particular, we first propose an efficient coalitional game-based cooperative offloading (CGCO) algorithm for the single-task model, and then expand it into a CGCO-M algorithm for the multiple-task model with jointly applying a two-stage flow shop scheduling approach, which helps to obtain an optimal task schedule. It is proved that our CGCO and CGCO-M can achieve the Nash-stable solution with convergence guarantee, and CGCO can obtain an optimal solution. The simulations show that CGCO is equal to the optimal exhaustive search (ES) method and CGCO-M is close to ES in terms of cost ratios. Cost ratios of CGCO and CGCO-M are significantly down by 41.08% and 83.70% compared to local executions, respectively. Meanwhile, CGCO-M obtains 41.46% and 89.74% reductions when reuse factors are 0.1 and 1, which means CGCO-M can save more cost with higher reuse density.

A Comparison of Tritium Dispersion Methodology for Accident Analysis in U.S. Department of Energy Complex Facilities

A comparison of three United States (U.S.) Department of Energy (DOE) Standard DOE-STD-3009–2014 dispersion modeling protocol options has been performed assuming a ground-level release of tritium oxide source term. The options are characterized by differing sets of assumptions and inputs that allow incorporating greater user flexibility and realism into the modeling and subsequent analysis. The three options used to evaluate atmospheric dispersion include: (1) Use of U.S. Nuclear Regulatory Commission (NRC) Regulatory Guide 1.145; (2) Application of a DOE-approved toolbox code and application of conservative input parameters; and (3) Use of site-specific methods and parameters as defined in a site/facility specific DOE-approved modeling protocol.Option 1 dose results are the lowest of the three sets of results at close-in distances, but are the highest for distances beyond approximately 3,000 m, reflecting the distance-dependent NRC plume meander model. Option 1 doses also reflect a lower minimum wind speed and consideration of G stability. Option 3 dose results are consistently lower than the Option 2 results by a factor of 2.2 reflecting the higher vertical dispersion values calculated from the crediting site-specific surface roughness. Option 2 and 3 results are obtained with DOE Central Registry computer software reflect default parameters in Option 2, and more site-specific input with Option 3. An averaging time of two hours leads to dose results that are lower than those obtained with an averaging time of three minutes by a factor of 2.5 due to the higher crosswind dispersion parameter values. This effect is due to the larger crosswind dimension of the plume with increasing averaging time using the Gifford meander model. A sensitivity case study indicates appreciable differences are observed between results obtained with the NRC Regulatory Guide 1.145 temperature difference (ΔT) method and those with U.S. Environmental Protection Agency (EPA) EPA-454/R-99–005 methodology for stability class categorization. A second sensitivity case suggests that crediting deposition, hold-up or other retention of tritium may be difficult to defend from a regulatory perspective, recognizing region of transport characteristics and accounting for reemission phenomenon. In terms of recommending one of the three options for modeling tritium releases in Documented Safety Analysis (DSA) applications, the Option 2 approach (Application of a DOE-approved toolbox code and conservative input parameters – without crediting tritium deposition) is the simplest model for source to receptor distances of 500 m or greater. Option 3 requires additional resource commitment and DOE authority approval, but may provide regulatory relief for certain accident scenarios. These recommendations apply to deterministic DSA dispersion analysis but are not extended to best estimate, realistic analyses such as those supporting probabilistic safety analyses.

NODAL3 Sensitivity Analysis for NEACRP 3D LWR Core Transient Benchmark (PWR)

This paper reports the results of sensitivity analysis of the multidimension, multigroup neutron diffusion NODAL3 code for the NEACRP 3D LWR core transient benchmarks (PWR). The code input parameters covered in the sensitivity analysis are the radial and axial node sizes (the number of radial node per fuel assembly and the number of axial layers), heat conduction node size in the fuel pellet and cladding, and the maximum time step. The output parameters considered in this analysis followed the above-mentioned core transient benchmarks, that is, power peak, time of power peak, power, averaged Doppler temperature, maximum fuel centerline temperature, and coolant outlet temperature at the end of simulation (5 s). The sensitivity analysis results showed that the radial node size and maximum time step give a significant effect on the transient parameters, especially the time of power peak, for the HZP and HFP conditions. The number of ring divisions for fuel pellet and cladding gives negligible effect on the transient solutions. For productive work of the PWR transient analysis, based on the present sensitivity analysis results, we recommend NODAL3 users to use2×2radial nodes per assembly,1×18axial layers per assembly, the maximum time step of 10 ms, and 9 and 1 ring divisions for fuel pellet and cladding, respectively.

Standardisation of the 129I, 151Sm and 166mHo activity concentration using the CIEMAT/NIST efficiency tracing method

The 129I, 151Sm and 166mHo standardisations using the CIEMAT/NIST efficiency tracing method, that have been carried out in the frame of the European Metrology Research Program project “Metrology for Radioactive Waste Management” are described. The radionuclide beta counting efficiencies were calculated using two computer codes CN2005 and MICELLE2. The sensitivity analysis of the code input parameters (ionization quenching factor, beta shape factor) on the calculated efficiencies was performed, and the results are discussed. The combined relative standard uncertainty of the standardisations of the 129I, 151Sm and 166mHo solutions were 0.4%, 0.5% and 0.4%, respectively. The stated precision obtained using the CIEMAT/NIST method is better than that previously reported in the literature obtained by the TDCR (129I), the 4πγ-NaI (166mHo) counting or the CIEMAT/NIST method (151Sm).

Influence of seeding and SOL transport on plasma parameters in JET ITER-like wall H-mode discharges

This paper describes integrated numerical modeling applied to JET ITER-like wall H-mode discharges using the COREDIV code, which self-consistently solves 1D radial transport of plasma and impurities in the core region and 2D multi-fluid transport in the SOL. In this paper we focus on sensitivity studies to give insight on the influence of different code input parameters on the simulations results. Simulations show, that for a given power and density value, there is always a limit on the seeding level, above which the self-consistent solution does not exist. This limit seems to be related to the edge density limit and might impose severe consequences on the accessible JET operational domain. Therefore, numerical investigations have been done to check how this limit can be released. It has been found that changes in the edge density, and in particular in the SOL transport can somehow mitigate this effect.

Post-test calculation and uncertainty analysis of the experiment QUENCH-07 with the system code ATHLET-CD

In the frame of developmental assessment and code validation, a post-test calculation of the test QUENCH-07 was performed with ATHLET-CD. The system code ATHLET-CD is being developed for best-estimate simulation of accidents with core degradation and for evaluation of accident management procedures. It applies the detailed models of the thermal-hydraulic code ATHLET in an efficient coupling with dedicated models for core degradation and fission products behaviour. The first step of the work was the simulation of the test QUENCH-07 applying the modelling options recommended in the code User's Manual (reference calculation). The global results of this calculation showed a good agreement with the measured data. This calculation was complemented by a sensitivity analysis in order to investigate the influence of a combined variation of code input parameters on the simulation of the main phenomena observed experimentally. Results of this sensitivity analysis indicate that the main experimental measurements lay within the uncertainty range of the corresponding calculated values. Among the main contributors to the uncertainty of code results are the heat transfer coefficient due to forced convection to superheated steam–argon mixture, the thermal conductivity of the shroud isolation and the external heater rod resistance. Uncertainties on modelling of B 4C oxidation do not affect significantly the total calculated hydrogen release rates.

Processing Expert Judgements in Accident Consequence Modelling

In performing uncertainty anaylsis a distribution on the code input parameters is required. The construction of the distribution on the code input parameters for the joint CEC/USNRC Accident Consequence Code Uncertainty Analysis using Expert Judgement is discussed. An example from the food chain module is used to illustrate the construction. Different mathematical techniques have been developed to transform the expert judgements into the required format. Finally, the effect of taking account of correlations in performing uncertainty analysis is investigated.

Probabilistic Accident Consequence Uncertainty Analysis of the Dose Calculations Module in the COSYMA Package

Uncertainty analysis of the dose calculation module of the COSYMA accident consequence assessment code has been undertaken, involving the following steps: (1) Expert judgement techniques were applied to assess uncertainties in measurable parameters determining external and internal doses. (2) The data obtained were used to calculate distributions on the dose quantities required as code input parameters. (3) The effect of uncertainties in dose quantities was analysed for a range of COSYMA end points, including the extent of countermeasures and incidences of early and late health effects, and the most important uncertainties were identified for inclusion in an overall uncertainty analysis of COSYMA. Parameters identified as making the largest contributions to uncertainty included external doses and location factors, residence times of materials on skin, breathing rates, and respiratory tract deposition and retention parameters, for the extent of countermeasures and early health effects, and caesium and iodine retention parameters for late effects.

Uncertainty in compartmental models for hazardous materials — a case study

Performing uncertainty analysis on compartmental models is the main topic of this article. Elements of the methodology developed during a joint CEC/USNRC accident consequence code uncertainty analysis are introduced. The uncertainty is quantified using structured expert judgment. Experts are queried about physically observable quantities. Many code input parameters of the accident consequence codes are not physically observable but are used to predict observable quantities. Therefore, a probabilistic inversion technique was developed which `transfers' the uncertainty from the physically observable quantities to the code input parameters. The probabilistic inversion technique is illustrated using the compartmental model of systemic retention of Sr in the human body. The article is concluded with a discussion on capturing uncertainty via compartmental models.

Identification of resonant, two-electron, and interatomic features in K and L 23 NEXAFS spectra

Experimental Cl K NEXAFS spectra for Cl − hydrated in H 2O, and on passivated Al electrodes (below and above the pitting potential) have widely different spectral line shapes. Experimental K- and L 23-edge NEXAFS data for the alkali halides and rare gas solids are compared with FEFF6 calculated results to optimize code input parameters, test code validity, and identify many-body features in these model systems. Differences are shown to arise from many-body shake-up and charge-transfer excitations. Although relatively small in the model spectra, the latter interactomic excitations are believed to be primarily responsible for the different Cl − line shapes on the passivated Al electrodes.

Quantifying reactor safety margins part 3: Assessment and ranging of parameters

Comparisons of results from TRAC-PFl/MOD1 code calculations with measurements from Separate Effects Tests, and published experimental data for modeling parameters have been used to determine the uncertainty ranges of code input and modeling parameters which dominate the uncertainty in predicting the Peak Clad Temperature for a postulated Large Break Loss of Coolant Accident (LBLOCA) in a four-loop Westinghouse Pressurized Water Reactor. The uncertainty ranges are used for a detailed statistical analysis to calculate the probability distribution function for the TRAC code-predicted Peak Clad Temperature, as is described in an attendant paper (Part 4). Measurements from Separate Effects Tests and Integral Effects Tests have been compared with results from corresponding TRAC-PFl/MOD1 code calculations to determine globally the total uncertainty in predicting the Peak Clad Temperature for LBLOCAs. This determination is in support of the detailed statistical analysis mentioned above. The analyses presented here account for uncertainties in input parameters, in modeling and scaling, in computing and in measurements. The analyses are an important part of the work needed to implement the Code Scalability, Applicability and Uncertainty (CSAU) methodology. CSAU is needed to determine the suitability of a computer code for reactor safety analyses and the uncertainty in computer predictions. The results presented here are used to estimate the safety margin of a particular nuclear reactor power plant for a postulated accident. Specifically, this paper describes, first in principle and then through their application, steps 7 through 10 of the fourteen-step CSAU methodology.