CASL

Strategy

The CASL implementation strategy, consisting of six key elements, is the method CASL employs to bring its goals to a successful conclusion. Each of the six CASL strategic elements is briefly described below. Associated with each element of the CASL strategy is one or more strategic roadmaps (see Appendix E) outlining the logical steps that build upon one another in progressing toward CASL’s strategic goals.

Challenge Problems for PWR Core Phenomena

Currently industry uses predictive tools that are based on an analytical framework that was developed over the past several decades, parts of which have origins in the 1960s and 1970s. Over the years, these tools have been improved based on the existing framework and new models, many empirical, which have been implemented and validated against in-reactor and out-of-reactor experimental data. Typically, the traditional tools are uncoupled or loosely coupled and are applied in series using simple models and conservatisms in the analyses. To achieve the key CASL objectives of supporting the enabling of power uprates, increasing fuel burnup and cycle length, and achieving life extension for U.S. nuclear plants, there is a need to better understand operational and safety margins.

CASL is focused on a set of specific Challenge Problems (CPs) that encompass the key phenomena currently limiting the performance of PWRs, with the recognition that much of the capability developed will be broadly applicable to other types of reactors. CASL defines a Challenge Problem as one whose solution is (1) important to the nuclear industry and (2) amenable to or enabled by M&S. Table 1 provides a brief summary on each challenge problem and their respective simulation gaps and drivers. Further detail on this table is also given in Appendix D.

Charters and implementation plans have been developed for the Challenge Problems. Table 2 provides a summary of the path forward based on the charters and implementation plans. Figure 2 gives a high level visual representation of how the CASL tools will be applied and some of the sources for data validation. It is reasonable to expect that, over time, the importance of current challenges will change as solutions are achieved and that new ones will emerge. Appendix D presents examples of possible future Challenge Problems that would expand CASL relevance to other LWR types.

Table 1: CASL Challenge Problem descriptions and drivers

CP Problem Description and Impact Simulation Gaps and Drivers
CIPS/CILC CRUD-Induced Power Shift and CRUD-Induced Localized Corrosion. High uncertainties in crud source, thickness, boiling surface area, and margin to fuel leakage affect fuel management and thermal margin in many plants, limiting power uprates. More accurate and higher resolution models for boiling surface area, crud deposition, boron uptake and cladding oxidation rate for each rod in core, with boron feedback in neutronics.
GTRF Grid-To-Rod Fretting. Rod growth changes, flow induced vibration, irradiation-induced grid growth and spring relaxation affect wear, especially for fuel near the core shroud. Reliable predictions of grid to rod gap, turbulent flow excitation, and resulting rod vibration and wear at any location in core.
PCI Pellet-Clad Interaction. Cladding creep-down onto pellets, followed by pellet expansion, creates local cladding stresses at pellet imperfections, resulting in clad failure sometimes assisted by SCC Sufficient 3D geometric detail of fuel rod material property changes; pellet growth, cracking and fission product release; cladding stresses, creep, fracture and SCC attach; fuel-cladding binding; and coupling to neutronics and T-H.
FAD* Fuel Assembly Distortion. Forces from radiation-induced growth and fluid flow may cause distortion and alter power distributions, challenge fuel handling, retard control rod insertion, and restrict plant operation. Fully 3D structural models of fuel assemblies accounting for material property changes, growth and creep; coupled with neutronics and TH.
RV & Internals Integrity* Damage from radiation results in increased temperature for onset of brittle failure and higher failure probability due to thermal shock stresses with safety injection for RV, and damage from radiation, thermal & mechanical fatigue render upper internal package more susceptible to distortion under blow-down or seismic loads. 3D prediction of temperature, force, stress, fluence and resulting material property changes of reactor internal structures and vessel; and solid mechanics prediction of vessel fracture and internals distortion.
DNB Departure from Nucleate Boiling. Local clad surface dryout, affected by detailed flow patterns and mixing, cause dramatic reduction in heat transfer during transients (e.g., overpower and loss of coolant flow) leading to high cladding temperatures. 3D subchannel and CFD tools to model detailed flow patterns downstream of mixing grids for single and two-phase flow, coupled to detailed pin-resolved radiation transport and fuel performance models for application to DNB transients (e.g., RIA, Loss of Flow).
Cladding Integrity (RIA,LOCA) Reactivity Insertion Accident and Loss Of Coolant Accident. Physical changes during the transient (e.g., swelling and burst, oxidation mechanics), leading to clad failure followed by fuel dispersal. Improvements to current simplistic fuel rod models, with improved predictive capability for normal operations to obtain fuel initial conditions at initiation of accident simulation, and transient fuel rod behavior.
* Currently not addressed in CASL Phase 1 scope.

 

Table 2: Simulation targets and path forward for CASL Challenge Problems


CP Simulation Capability Targets Path Forward
CIPS Validated, predictive tools that reduce uncertainty for simulating CIPS and increase flexibility in incore fuel management.
  • Couple models for consistent feedback between neutronics, T-H, crud/chemistry accounting for boron and crud feedback
  • Calculate 3D pin resolved transport neutronics, subchannel TH and crud deposits
  • Calculate erosion of crud and crud chemistry
  • Calculate 3D mass of boron deposited in crud and CIPS throughout life
CILC Validated, predictive tools that reduce uncertainty for simulating CILC and increase flexibility in incore fuel management.
  • Zoom in to limiting region of core for crud deposition based on CIPS analysis
  • Provide azimuthal variation of heat transfer downstream of spacer grids
  • Predict 3D crud deposits, crud chemistry and clad temperature around limiting rods
  • Predict CILC and margin to leaker
GTRF Validated, predictive tools for simulating wear of any rod in core and to evaluate impact of spacer grid design features.
  • Predict turbulent excitation force at any location in core using 3D CFD modeling tools
  • Predict fuel rod and spacer grid structural material behavior as function of fast fluence
  • Predict fuel rod and spacer grid corrosion material behavior as function of fast fluence
  • Predict gap between grid cell support structures and the fuel rod versus time in core
  • Develop wear model that takes into consideration contact type and material condition
  • Perform structure-dynamic simulation to translate hydraulic forces to rod vibration
  • Predict rod wear margin for any location in core
PCI Validated, predictive models for 3D fuel rod behavior that increase flexibility in plant power maneuvering.
  • Couple fuel rod model (2-D) to core simulator (core-wide power/TH calculations) to calculate performance indicator throughout entire core
  • Develop method to couple full-length fuel rod model to local effects stress and failure potential simulation (3-D effects)
  • Develop fuel rod model to calculate fuel rod conditions during operation for full-length rods
  • Develop 3-D local effects (few pellets) modeling methods to evaluate conditions leading to PCI failure
FAD* Validated, predictive tools for fuel assembly distortion to evaluate impact of assembly skeleton design features.
  • Build 3D structural models for fuel in core
  • Couple structural model with neutronics and TH
DNB Validated tools to predict DNB using more advanced methods to evaluate safety margin, enhance understanding, and evaluate impact of spacer grid design features.
  • Couple neutronics and subchannel thermal hydraulic tools to better predict DNB for transients such as RIA
  • Implement and validate two-phase models in CFD tools for predicting DNB
  • Predict DNB with improved CFD tools and validate to available data
RIA Validated modeling improvements to more accurately assess fuel performance during RIA and LOCA and evaluate impact of Accident Tolerant Fuel features.
  • Implement pin-resolved modeling of the transient neutronics for RIA including the effects of local temperature reactivity feedback
  • Develop advanced transient fuel rod thermal-mechanical tools capable of modeling all of the characteristics of the initial and transient conditions and behaviors of the fuel rod such as fuel pellet cracking and relocation, fuel-to-pellet contact and binding, and cladding outer diameter oxidation due to normal in-service corrosion and hydrogen pickup.
  • Couple tools to simulate RIA and predict PCMI and margin to failure
LOCA Validated modeling improvements to more accurately assess fuel performance during RIA and LOCA and evaluate impact of Accident Tolerant Fuel features.
  • Perform same tasks as RIA but apply to LOCA transient
  • Utilize boundary conditions from RELAP5 System Code for LOCA
  • Evaluate maximum PCT and oxidation margin
  • Investigate benefits of ATF
RV & Internals Integrity* Validated, predictive tools for vessel performance during cold leg injection and for internal performance during hot leg blow-down or seismic event.
  • 3D prediction of temperature, stress, fluence, and growth of fuel and reactor internal structures and vessel
  • Predict any component vibration
  • LWRS Scope: Predict fatigue, SCC and radiation damage
* Currently not in CASL Phase 1 scope

Figure 2: Plan to address Challenge Problems

VERA: Virtual Environment for Reactor Applications

In order to address the Challenge Problems previously described, CASL is collectively developing a suite of simulation capabilities known as VERA. While VERA development is guided by the CASL Challenge Problems, its capabilities will be broadly applicable for other PWR analyses, and designed to be extensible, to allow future application to other reactor types and to challenges beyond normal operation.

The Physics Integration (PHI) focus area (FA) is responsible for overall VERA development, including the software development and testing processes, physics coupling strategies, and usability, including overall analysis workflow development that could differ significantly from current industry practice due to the integrated nature of the VERA tools.

VERA is not a single simulation tool but rather a collection of capabilities for scalable simulation of nuclear reactor core behavior (Figure 3). It is a flexible toolkit of components that can be exercised in various combinations for different Challenge Problems and for varying fidelity requirements or computational resources:

  • Baseline: Existing, and in most cases proprietary, industry codes in use by CASL partners today. These tools have been integrated to provide a baseline for comparison against new CASL-developed capabilities and useful insight into current industry analysis practices, as well as sensitivity analysis to guide development priorities. However, VERA baseline components are not undergoing further development but are available for CASL industrial partners’ usage under NDA restrictions.
  • Advanced: These are capabilities integrated into the common VERA software environment, where they are algorithmically coupled, share input/output, and are subjected to sensitivity analysis and UQ tools.
    • Leadership-class Platform capabilities – CASL’s commitment to higher fidelity understanding of reactor phenomena requires a more rigorous approach to radiation transport, fluid flow, fuel rod performance, and chemistry prediction, and this has spurred investments in 3D discrete ordinates (Sn) radiation transport methods; hybrid finite-volume/finite-element compressible/low-Mach flow multiphase solvers; 3D finite element thermal/solid mechanics models utilizing enhanced material modeling techniques; and 3D primary coolant/cladding surface chemistry.
    • Industry-class Platform capabilities – Recognizing the need for higher-fidelity simulations on an industry-sized computing platform, CASL has elected to provide additional capabilities using alternative, less computationally intensive methods such as 2D/1D Methods-of-Characteristics (MOC) and 3D Simplified PN (SPN) radiation transport; sub-channel flow solutions; 2D (r,z) finite element fuel rod mechanics utilizing enhanced material modeling techniques; and 2D (r,z) primary coolant/cladding surface chemistry.

Table 3 lists the codes currently integrated in or planned for advanced VERA, showing how their capabilities and attributes compare with tools currently in use by industry. Table 4 shows how specific combinations of these VERA capabilities are envisioned to address the current CASL Challenge Problems. A specific subset of advanced VERA capabilities can be grouped to encompass functionality similar to current industry tools often referred to as “core simulators.” VERA core simulator components (denoted by the dashed line in Figure 3) are typically of higher-fidelity with increased feedback between physics components and are designed to be scalable from small clusters to the largest supercomputers. They address quasi-steady-state LWR conditions and depletion, including models for neutron transport, cross sections, thermal-hydraulics, 2D fuel thermal/mechanics, and depletion. They can link to other physics (e.g., CFD, 3D fuel thermal/mechanics, CRUD models, structural, reactor system) but their principal function is to provide reactor conditions and distributions needed to solve the CASL Challenge Problems. Their distinguishing capability is to deplete the reactor core on a pin-by-pin resolved basis with actual reactor conditions, as shown in Table 5.

Figure 3: VERA - a toolkit of components for scalable simulation of reactor cores

 

Table 3: Advanced VERA component suite

Code Physics Current Practice CASL Capabilities

MPACT


Insilico

Radiation transport

  • MG lattice physics (2D MOC transport) 
  • FG core-wide physics (3D nodal diffusion) 
  • Pin-wise power/flux (reconstruction)
  • MG core-wide physics (2D MOC transport)/axial leakage (1D or 3D SPN)
  • 3D pin-resolved transport
  • Prototype hybrid Monte Carlo transport
  • Modern cross section/shielding treatments

Hydra-TH

Fluid dynamics

  • Limited CFD
  • 3D multiphase/multi-field equations solved with focus on bubble flow
  • Multiple LES and RANS models
  • Higher-order temporal and spatial treatments
  • State-of-art nonlinear solver routines

Cobra-TF

Subchannel thermal-hydraulics

  • Closed channel homogeneous equilibrium model
  • Limited subchannel
  • Updated capabilities with cross-flow
  • Modernized and parallelized with improved solvers

Peregrine

Fuel performance

  • 2D with experimentally derived closure models
  • 2D or 3D capabilities
  • Experimentally derived closure models
  • Micro/meso-scale modeling derived closure models

MAMBA & MAMBA-BDM

CRUD Chemistry & Deposition

  • 2D with limited chemical species
  • Experimentally derived models’ parameters
  • 2D or 3D capabilities
  • Increased number of chemical reactions and associated reactants and products treated
  • Detailed models of chimneys, crud porosity & structure, two-phase Darcy flow, solubility, & zirconium oxidation

SIERRA/SDSM

Structural Dynamics & Solid Mechanics

  • Limited FEM
  • FEM shell, solid, beam, spring, and gap contact elements
  • Quasi-static, transient, and eigen mode solvers with contact
  • Periodic constraints in conjunction with contact
  • General user material capability allowing radiation growth and creep models

Table 4: VERA components applied to CASL Challenge Problems

Challenge Problem Domain Time Scale VERA Components

CRUD-Induced Power Shift (CIPS)

Full core

years

MPACT – Cobra-TF – MAMBA-2D

CRUD-Induced Localized Corrosion (CILC)

Few pin-wide

years

MPACT/Insilico – Hydra-TH – Peregrine – MAMBA-3D

Grid-to-Rod Fretting (GTRF)

Few pin-wide

years + 50 Hz

MPACT – Hydra-TH – Peregrine – SIERRA/SDSM

Pellet-Clad Interaction (PCI)

Few pin-wide

years + min-hr

MPACT/Insilico – Hydra-TH – Peregrine

Departure from Nucleate Boiling (DNB)

System to assembly

years + sec-min

MPACT – Cobra-TF/Hydra-TH – Peregrine – RELAP5

Clad integrity during RIA

Few pin-wide

years + sec

MPACT – Hydra-TH – Peregrine

Clad integrity during LOCA

Pin-wide, years

years + sec-min

Peregrine (BCs from Westinghouse Retran/Cobra-TF)

Fuel Assembly Distortion (FAD)

Assembly

years

MPACT – Hydra-TH – Peregrine – SIERRA/SDSM

Structural integrity

R, internal structures

years

MPACT – Hydra-TH – SIERRA/SDSM

Table 5: Comparison of CASL’s core simulator with industry core simulators

Physics Model Industry Practice VERA Core Simulator Components

Neutron Transport

3D diffusion (core), 2 energy groups (core), 2D transport on single assembly

3D transport, 23+ energy groups

Power Distribution

nodal average with pin power reconstruction methods

pin-by-pin *

Thermal Hydraulics

1D assembly-averaged

subchannel with cross-flow

Fuel Temperatures

nodal average

pin-by-pin *  2D or 3D

Xenon / Samarium

nodal average with correction

pin-by-pin *

Depletion

infinite-medium cross sections, quadratic burnup correction, history corrections, spectral corrections, reconstructed pin exposures

pin-by-pin * with actual core conditions

Reflector Models

1D cross section models

actual 3D geometry

Target Platforms

workstation (single processing core)

1,000 – 300,000 processing cores

*pin-homogenized or pin-resolved, depending upon application

CASL will ultimately provide enhanced capabilities relative to current practices. VERA components offer predictive capabilities that utilize science-based approaches and micro- and meso-scale models to increase understanding and provide closure relationships. They incorporate higher resolution in space, time, energy and angle to achieve pin-resolved detail. Verification and validation, data assimilation, and uncertainty quantification are embedded in the development and application of these components. They implement modern, object-oriented, third-party software as part of the solvers and coupling infrastructure. To accommodate the rapid pace of CASL and evolving requirements, VERA is developed using modern Agile software development methodologies, which emphasize frequent communication within the development team (facilitated by the VOCC technologies described in Section 3.6) and close interaction with customers (users).

Development of VERA follows the TriBITS Lifecycle Model [http://web.ornl.gov/~8vt/TribitsLifecycleModel_v1.0.pdf] and uses the following terms to define the maturity of VERA capabilities:

  • Exploratory (EP): Prototype software to investigate new ideas and alternative approaches;
  • Research Stable (RS): Software under active development with strong software engineering practices but may lack usability attributes expected of production software;
  • Production Growth (PG): Software with all RS attributes but with improved usability (e.g., error checking and reporting, documentation) expected of production-quality software;
  • Production Maintenance (PM): Software not actively developed except for bug fixes and minor improvements.

Objectives for VERA by 2015 include:

  • Released and in use at Test Stands with supporting documentation and tutorials
  • Release and support model implemented
  • Functionalities integrated and coupled (varying degrees of maturity)
    • T-H: pin-by-pin subchannel, single- and multi-phase CFD
    • Neutronics: pin-homogenized transport, pin-resolved transport, hybrid transport
    • Fuel performance: advanced and upscaled capabilities
    • Coolant chemistry: CRUD formation and evolution
    • SA/UQ: advanced DAKOTA-based capabilities integrated and accessible
    • Structural mechanics: advanced general purpose capabilities demonstrated on GTRF and designed to address fuel assembly distortion and structural integrity issues
    • System: accessibility to RELAP5 capabilities
    • Infrastructure: build/test system, LIME and coupling model evaluators, data transfer and coupling toolkits, linear/nonlinear solvers, reactor aware I/O
  • Core Simulator: full functionality for all Benchmark Progression Problems (core follow and reload)

A detailed technology roadmap for VERA during Phase 1 of CASL is given in Appendix E.

Enabling R&D

Three (RTM, THM, MPO) of the six CASL focus areas (FAs) are dedicated to developing fundamental models of each relevant physical phenomenon linked through VERA, and one focus area (VUQ) is developing and utilizing Verification and Validation (V&V) and UQ capabilities.

Each of these science focus areas has an expected evolution of R&D accomplishments and technology capabilities as given in Appendix E roadmaps. Specific 2015 capability goals are presented in Table 6.

Table 6: 2015 capability goals for the CASL science Focus Areas

Focus Area 2015 Capability Goal
RTM Robust 3D pin-resolved transport and prototype hybrid Monte Carlo transport with modern cross section/shielding treatments and coupled by PHI focus area to T-H, fuel, and corrosion chemistry capabilities.
THM 3D single and multiphase capabilities with multiple LES and RANS models, including advanced closure models suitable for subcooled boiling processes in PWRs. This will be complemented with a computationally enhanced subchannel capability developed by PHI focus area.
MPO Mature 3D fuel performance capability with full assessment against CRUD/PCI/GTRF problems. Validated fuel performance models inform assessments of safety margin (PCI) and best operational practices (CRUD, GTRF). Functional capability and partial assessment for RIA- and LOCA-based transient problems.
VUQ Mathematical tools and methodologies integrated and accessible to enable quantifying sensitivities and uncertainties (following data assimilation) in full-scale multiphysics PWR simulations.

 

Radiation Transport

The overall radiation transport objective, owned by the RTM FA, is to develop pin-resolved predictions of flux, power density and interaction rates utilizing transport theory preferable over the entire core, but at a minimum over spatial regions of the core sufficient to address the Challenge Problems. "Pin-resolved" includes axial, radial, and azimuthal resolution within the fuel pins. Such spatial resolution is not only required to support the fuel performance and thermal-hydraulic model development activities, but also a number of other basic reactor analysis tasks such as excore detector response, vessel and internals packages fluences, control rod lifetime, incore instrument response, and fluence-based irradiation assembly bow. A three-prong approach is being taken to develop pin-resolved M&S capability. The main path is based upon the 2D/1D method, consisting of a full core 2D (radial plane) Method of Characteristics (MOC) transport model coupled with a 1D Cartesian diffusion (or nodal) axial model to obtain the axial leakage component. Alternative axial models based on low order transport theory are under development. An alternative path, which is also expected to result in a usable advanced technology for VERA over the first five-year term of CASL, is based upon a hybrid 2D (radial) Method of Characteristics (MOC) transport model coupled with a 3D Cartesian low order transport model (SPN) to obtain the axial leakage component. For the longer term, a 3D hybrid Monte Carlo (MC) modeling approach is being pursued, consisting of continuous energy Monte Carlo that is accelerated by adjoint-based weight windows, and/or CMFD (coarse mesh finite difference) and/or related acceleration methods. Underpinning these activities are advanced cross section processing and isotopic depletion methodologies, and multi-physics integration. Figure E4 provides further details and the time line for RTM development.

Thermal Hydraulics

The thermal hydraulics (T-H) objective, owned by the THM FA, is mainly focused on the development of single-phase and multi-phase 3D Navier Stokes computational fluid dynamics (CFD) capabilities, since these capabilities are required to analyze several of the Challenge Problems. The more traditional engineering subchannel model for T-H analysis is provided and integrated into VERA via the Cobra-TF code with the main purpose being to correct neutron cross-sections for T-H feedback. Subchannel model development, therefore, is limited within the CASL T-H objective, as the development and application of single- and multi-phase CFD is the principal focus. Single-phase CFD capability is relatively mature, aside from turbulence modeling for reactor-specific problems, and thus the near-term focus has been on developing such capability for public release. To better understand two-phase bubbly flow which occurs during subcooled-boiling and onset of DNB, complementary experimental and microscale modeling activities utilizing existing resources are underway, providing the foundations to develop the initial multi-phase CFD capability. In addition, M&S capabilities that are believed to be of lower accuracy but computationally efficient are being pursued, along with integrating thermal hydraulics into UQ methodologies.

Materials Science and Fuel Performance

The materials science and fuel performance objective, owned by the MPO FA, is to develop multiphysics, multiscale materials M&S capabilities that predict fuel performance during normal operation and a select set of design-basis accidents. The overarching objective is to develop improved predictive modeling of complex fuel degradation phenomena to better understand and define operating and safety margins, which in turn will increase the fuel utilization, power and safety of our current nuclear fleet. The current focus is on development of rigorous engineering scale modeling frameworks to directly address GTRF, CRUD (CIPS and CILC) and PCI challenge problems and simultaneously engaging a diverse team of materials scientists and physicists to better understand the underlying physics and chemistry of fuel degradation phenomena. For example, initially, fuel performance models will be developed based on the historical development of understanding of materials degradation and the implementation of semi-empirical based constitutive relationship. It is anticipated that the more fundamental, multiscale investigation will ultimately lead to improved material property models, or constitutive relationships, with improved physical fidelity to enable engineering scale fuel performance predictions. Although the focus for MPO is normal operation of LWR fuel (Figure E6), MPO envisions that once the advanced CASL LWR fuel performance models have been verified and validated, these tools will be used to assess the behavior of current and advanced fuels during accident conditions. As transient conditions are simulated, a number of additional degradation phenomena need to be incorporated. The MPO strategic plan to evaluate accident behavior and to assess the performance of advanced fuel forms, clad and assembly designs is depicted in the Figure E7 roadmap.

Validation and Uncertainty Quantification

The VUQ FA plays a critical role in achieving CASL’s goals in two key areas. First, decision-makers must be able to assess whether VERA predictions are accurate enough for the intended use, particularly when evaluating design and safety margins. Supplementing model predictions with confidence intervals derived from VUQ processes helps to facilitate this assessment. Second, VUQ can help to focus the efforts of model developers by executing a numerical Phenomena Identification and Ranking Table (PIRT) process to determine which models have a high impact on uncertainty, augmenting the expert-opinion-based PIRT process and more effectively concentrating CASL resources. Using CASL’s Challenge Problems as drivers, VUQ is addressing the difficulties of treating multi-physics and multiscale coupling, large parameter spaces, and nonlinear behaviors. Since each Challenge Problem is computationally intensive, and multiple cases are needed to develop uncertainty analyses, VUQ is pursuing techniques for the generation and use of surrogate models. Surrogate models can be constructed either mathematically or by calibration of lower fidelity models with higher fidelity model output. Comparing these physics-based surrogates with math-based surrogates and improving the accuracy of both will be one of the key VUQ focuses in a potential Phase 2 of CASL. Construction of the bridge between high fidelity high performance computing solutions and lower fidelity industry class computing solutions will be necessary to deliver advanced modeling and simulation tools to industry. Throughout the lifetime of CASL, VUQ will periodically assess the predictive maturity of VERA and report on its progress. To support this assessment, enhanced verification and validation capabilities are being developed, such as automated mesh refinement capability and high-order phase space characterization in support of code and solution verifications. Combining existing experimental data, new experiments, and high fidelity calculations to reduce uncertainty in the industry class computing tools will enable industry to improve the amount of nuclear power generated, its reliability, and its safety.

Technology Delivery

A key strategy for CASL is the proactive delivery of its technology to current and potential customers, clients, and users throughout the nuclear energy enterprise. In keeping with an agile philosophy and project management approach, CASL is committed to early and regular delivery of its technology (e.g., VERA) on fixed schedules, with the functionality of this technology evolving based on iterative feedback with those receiving and applying/testing the technology.

Release and Support of VERA

In accordance with its vision, namely that VERA will be deployed and applied broadly throughout the nuclear energy community, CASL envisions two strategic paths for release and support of VERA:

  • Continued availability and limited support of stable (static) or beta (exploratory) versions of VERA through Radiation Safety Information Computational Center (RSICC) distribution at ORNL and through open source distribution of the VERA infrastructure in accordance with practices established during active CASL execution; and
  • Periodic availability and full support of evolving, stable, industry-hardened, and user-friendly versions of VERA through distribution by a CASL core partner or a third party entity. Specific options that are under consideration for post-CASL distribution of VERA include:
    • National laboratory deployment and maintenance via multiple nonexclusive licenses or a single exclusive license;
    • RSICC deployment;
    • EPRI Maintained services;
    • Deployment in the form of multiple licenses managed by selected organizations;
    • Licensing CASL Tools to one or more nuclear engineering service companies for broader deployment;
    • Work with independent software vendors to provide CASL Tools as “shrink wrap” software;
    • Any combination of the above options depending on maturity of the software and the field-of-use definitions agreed upon by the parties to the IPMP.

In the past, nuclear simulation capabilities have been accessed by transfer of software and/or by access to HPC resources hosting the software. When utilizing some of the advanced models in VERA, the latter access approach may be necessary and commercially viable. The version(s) of VERA available to the public through RSICC, subject to export control and proprietary data access limitations, consist of all the CASL-developed releases created during active (funded) execution of the CASL Program. These releases are envisioned to be executed in the fourth quarter of every calendar year. The technology developed in the releases will have originated from CASL core partners and contributing members, with the majority originating from DOE laboratory and university institutions. A VERA User’s Group may be chartered to assist with improving the stability, adoption, and evolution of VERA.

Beyond CASL lifetime as an Energy Innovation Hub (2015 or 2020), subsequent nuclear technology federal programs (in the DOE or elsewhere) supporting M&S technology development and application would be encouraged to leverage and evolve the VERA releases. VERA updates that are released after the CASL Program is complete would be compliant with the CASL VERA licensing model and IP Management Plan. It is envisioned that a stable and user-friendly version of VERA will be available through one or more CASL core partners, or a third party entity, after the CASL Program is complete. This is further discussed in Section 5.3.

Test Stands

CASL Test Stands are functional prototype VERA installations at a CASL industry partner or Industry Council member site. The principal objective of the CASL Test Stands, ideally deployed during active VERA development, is to provide constructive feedback to CASL through critical exercise of CASL virtual reactor capabilities while addressing issues that are important to users. The Test Stands involve early deployment of VERA into an actual engineering/design environment to give the Test Stand host improved modeling and simulation capabilities and to give CASL enhanced feedback and testing. Pilot Projects, early demonstrations of CASL-developed capabilities to a problem of interest to industry, are identified and implemented in Test Stands as joint efforts with the Industry Council. Externally funded collaborative applications may eventually be proposed and sponsored through the Industry Council for problems not strictly scoped within CASL priorities.

Test Stand activities must in general be supported by both CASL (for development, release and refinement) and an industry partner (for application, post analysis, and possibly refinement). CASL funding will be used for applications that directly support CASL commitments, as defined in the CASL work plans, and direct industry support will be used for other applications that are clearly beyond CASL scope. For applications that indirectly support CASL commitments and work plans, a mixture of CASL funding and industry funding may be appropriate. Table 7 gives a tentative schedule for Test Stand deployment and initial simulation completion.

Table 7: Test Stand schedule

Institution Test Stand Deployment Date Initial Simulation Completion

Westinghouse Test Stand

June 28, 2013

November 30, 2013

EPRI Test Stand

October 31, 2013

March 31, 2014

TVA Test Stand

April 30, 2014

October 31, 2014

Non-Partner Test Stand

October 31, 2014

March 31, 2015

 

Education and Training

The CASL education and training objective is to educate today’s reactor designers in the use of advanced M&S tools, and to develop the next generation of nuclear engineers and scientists through development of new curricula.

The CASL education and training effort is the first of its kind to develop new curricula and programs in nuclear energy at the graduate level based on the M&S technologies being utilized by CASL. The CASL education and training strategy is being carried out with the following elements:

  • Promoting diversity and excellence among the nuclear engineering student population with active recruiting of top students at the undergraduate and graduate levels, underrepresented minorities and women to higher levels, and interdisciplinary students;
  • Mapping the CASL Challenge Problems into existing graduate course curricula;
  • Developing new chapter modules for courses based on CASL Challenge Problems course materials;
  • Developing and executing an industry education knowledge transfer plan;
  • Developing a template of courses for a new certificate in nuclear systems design; and
  • Executing annual programs for undergraduate scholars and graduate student interns.

Computing Infrastructure Requirements

CASL currently relies on in-kind computing contributions from systems at its national laboratory partners. CASL projects will have very large computing needs in the near future, with 100M core-hours estimated for FY13 and subsequently increasing by about 50% per year. Additional computing resources are also needed to support data visualization, data analysis, data storage, and other pre- and post-processing activities. CASL is exploring several options for meeting these needs, including:

  1. Continuing to use cycles donated by partner labs and applying for time on leadership class systems;
  2. Acquiring and operating a dedicated capacity system at a central location, and continuing to apply for time on leadership class systems;
  3. Acquire cloud computing services from scientific providers in the commercial marketplace; and
  4. Creating a NERSC-like model for all of the DOE Office of Nuclear Energy.

Although estimated costs vary substantially with option, there is considerable precedent for a DOE/NE-wide HPC facility with dedicated R&D staff. In the near term, CASL intends to pursue option 2 and to enlist the Board of Directors to advocate option 4 as the foundation for a DOE/NE commitment to the rapidly growing demand for HPC, leveraging the path-finding experience of NNSA and the Office of Science.

Intellectual Property Management

Intellectual Property (“IP”) means patents, copyrights, trade secrets, information or data that may be proprietary, sensitive, restricted, or confidential, as well as information that may be patentable under Title 35 of the U.S. Code or copyrightable under Title 17 of the U.S. Code. Intellectual Property includes any analysis technique, process, procedure, method of operation, design, discovery, invention or improvement, which is conceived and/or first actually reduced to practice, made, or generated in performance of the CASL Program, including Subject Inventions as defined in 37 CFR 401, patent applications, issued patents, copyrights, rights in any technical data, computer software, trademarks or mask works, which are first made or generated in performance of the CASL Program. New IP owned or created by a CASL Member (“CASL IP Owner”) with CASL Funding (“CASL IP”). Software, inventions, documentation and/or other information created with funds other than CASL Funding may be called out by the CASL Members as Background Intellectual Property (“Background IP”).

All CASL IP and Background IP will be managed through the implementation of an IP Management Plan (IPMP) and corresponding licenses to provide rights to practice, further develop, and/or distribute the CASL technologies. The principal goals of the IPMP for the CASL Program include:

  • Rapidly and successfully transferring nuclear reactor simulation and modeling technologies to the U.S. nuclear industry in order to facilitate the operational performance and longevity of today’s operating reactors and the design and analysis of next-generation reactors and fuel technologies;
  • Openly disseminating scientific reports and results for public benefit;
  • Broadly and rapidly disseminating information among the CASL Members to maximize productivity and progress, and
  • Complying with all federal intellectual property law.

The IPMP will also be utilized to reference the potential deployment concepts that could be implemented to move the CASL IP to market. As CASL IP is matured for commercialization, each concept will be evaluated to determine the most appropriate path for commercialization. Suitable license provisions to convey rights in CASL IP and Background IP will be negotiated for appropriate consideration in accordance with the goals and objectives of CASL, preservation of the US taxpayer investment in CASL, and any pre-existing encumbrances and/or license requirements that may exist for Background IP. These negotiations will be led by the commercialization executives who represent the CASL Members generating the IP.

Collaboration and Ideation

CASL’s virtual one-roof objective wholly relies upon the development and maturation of its Virtual Office, Community, and Computing (VOCC) venture (www.voccnet.org). The VOCC Lab is a technology environment that promotes collaboration and critical thinking, which fosters insight and ultimately leads to innovation. The VOCC mantra—innovation at the speed of insight—is accelerated innovation, realized by providing an optimal environment for collaborative CASL development and assisting in demonstrating CASL-developed M&S capabilities (VERA) to CASL partners, industry, and potential customers and users.

The VOCC delivers a unified collaboration platform and creative work environment to support CASL’s mission with two primary goals: (1) to bring together under one virtual roof the best scientists, researchers, students, and design and applications engineers; and (2) to successfully operate and mature a state-of-the-art scientific collaboration space (the VOCC Laboratory) that supports all CASL R&D use cases: business and program management; agile code development; VERA application and data analytics (of VERA simulation datasets); and lectern environment and research partnerships. The VOCC uses human-centered, immersive, and visual analytic design techniques and principles to build a physical workspace for the specific purpose of unifying geographically distributed CASL staff and computational (e.g., HPC) resources.

The CASL VOCC Laboratory supports CASL’s synchronous and asynchronous communications and visualization needs with traditional elements of a collaboration platform such as group communication tools, messaging, social networking and computing tools, as well as 2D visualization and 3D information manipulation. The VOCC Laboratory is where CASL researchers perform both physical and virtual collocation work and where they can think or ideate. The strategic VOCC goal is to provide virtual collaboration tools (accommodating large-scale data transfer, storage, and management) and multi-tenant portals that can be accessed by anyone, at any time, from anywhere on any device.