Download presentation
Presentation is loading. Please wait.
1
VHTR Methods R&D Plan: Overall Perspective
Richard R. Schultz, Program Manager Very High Temperature Reactor Design Methods Development & Validation Workshop: VHTR R&D Plan Hosted by: Academic Center of Excellence for Thermal Fluids and Reactor Safety, Oregon State University September 19-20, 2005
2
Outline Background Goals for VHTR
Needs: CFD and systems analysis software development & validation The Role of CFD & systems analysis software Approach for achieving goals Summary of experimental R&D Summary of PIRT
3
Energy Bill Was recently passed by Congress & signed by President
Calls for locating a prototype Very High Temperature Reactor (VHTR) at the INL. Specifies that a consortium of appropriate industrial partners will be organized that will carry out cost-shared research, development, design, and construction activities, and operate research facilities on behalf of Project (see Energy Bill—Section 643—Subtitle C: Next Generation Nuclear Plant Project).
4
Why at the Idaho National Laboratory?
52 nuclear test reactors First nuclear electricity to power a town 890 sq miles Lead Lab for Nuclear R&D Overview of the INEEL HAVE DONE POOR JOB ENGINEERS ASSUME TECHNOLOGY IS THE ANSWER!! BURMA, LIBERIA, US – NON METRIC Craters of the Moon National Monument is 25 miles away by road Largest recent (10,000 years) basaltic lava field in the continental United States INEEL Site is surrounded by wilderness
5
Energy Act… I’m not going to interpret anything in the Energy Act.
It is important to remember that although millions of dollars are mentioned in Energy Act—we won’t have dollars unless they are appropriated. For your reference regarding “…enabling research, development, and demonstration activities on technologies and components for reactor and balance-of-plant design, engineering, safety analysis, and qualification” see Sec 643 pages 622 through 627. R&D will be done for both pebble-bed and prismatic reference designs [Phase II, in which a design competition between concepts will be held, will not begin until NERAC determines that the objectives of Phase I have been achieved (see Sec 643 item (c)(3)(D), page 627)]. Appropriations—see pages 902 through 915.
6
The High Temperature Gas Reactor is Reference Design
Utilize inherent characteristics Helium coolant - inert, single phase Refractory coated fuel - high temp capability, low fission product release Graphite moderator - high temp stability, long response times Prismatic Simple modular design: Small unit rating per module Low power density Silo installation Passively safe design: Annular core Large negative temperature coefficient Passive decay heat removal No powered reactor safety systems Pebble-bed
7
Strengths & Weaknesses of Prismatic Design Relative to Pebble-Bed—INL Perception
Larger fabrication, operating, & licensing experience base in US. Flow paths are well known and relatively controllable due to fixed core design; peak fuel temperature may be more predictable. Placement of control rods in fuel region is easier. Weaknesses: Larger excess reactivity, higher control worth, and relatively high packing fractions required to get desired operating cycle length. Must be shut down periodically for refueling and refueling is relatively complicated. Fuel at hot spots remains at same location relatively long time. Relatively strong reactivity increase upon significant water ingress. _______________ Items in red font require methods R&D to demonstrate capability of tools.
8
Strengths & Weaknesses of Pebble-bed Design Relative to Prismatic Design—INL Perception
Very little excess reactivity is needed—thus (a) reactivity insertion accident essentially eliminated from consideration, (b) proliferation attempts easy to detect, and (c) significant reduction in reactivity insertion due to water ingress. Very effective fuel utilization. Few reactor shutdowns required (no refueling outages). Fuel enrichment is lower, easier to fabricate fuel pebbles, thus probably lower fuel costs. Peak fuel temperatures will probably be lower. Pebbles pass through high power region relatively rapidly—so fuel duty is milder and shared among many more elements. Weaknesses: Likely to be more difficult to calculate flow and temperature variations. More pebble withdraw tubes are needed for annular core than for a solid core; bridging and stuck pebbles are a possibility. Larger pressure drops across core (for 10 m high core). However, cross-flow design may eliminate this issue. Production of dust. AVR generated approximately 3 kg/year from control rod insertions, rubbing of fuel pebbles, and drag of fuel pebbles on vessel. Potentially may be more difficult to license. _______________ Items in red font require methods R&D to demonstrate capability of tools.
9
Portion of R&D Need Matrix…
Region of System Operational Conditions Depressurized Conduction Cooldown Pressurized Conduction Cooldown Inlet Plenum IP1: Validation of CFD mixing calculation during transient. Core CO1: Nuclear data measurements to reduce calculational uncertainty. CO2: Modification of cross-section generation code to treat low-energy resonances with upscattering. CO3: Development of improved method for computing Dancoff factors. CO4: Characterization of hot channel temperatures and fluid behavior at operational conditions. CO5: Validation using integral experimental data. CD1: Validation of systems analysis codes to demonstrate capability to predict thermal behavior. CD2: Validation of models that calculate fission product release from fuel. CD3: Validation and calculation of air ingress and potential water ingress behavior into reactor vessel and core region. CP1: Validation of systems analysis codes to demonstrate capability to predict thermal and hydraulic behavior. Outlet Plenum PO1: Validation of CFD mixing using mixed index refraction (MIR) facility data & data available in literature. Perform calculation of fluid behavior with validated code. PD1: Validation of CFD mixing during operational transients and effect on turbine operational characteristics. Perform calculation of fluid behavior. PP1: Validation of CFD mixing during operational transients and effect on turbine operational characteristics. Perform calculation of fluid behavior. RCCS RO1: Validation of natural convection characteristics in cavity at operational conditions. RO2: Characterization of natural convection characteristics in cavity at operational conditions. RD1: Validation of heat transfer & convection cooling phenomena present in reactor cavity and via RCCS. RP1: Validation of heat transfer & convection cooling phenomena present in reactor cavity and via RCCS.
10
VHTR Objectives Demonstrate a full-scale prototype VHTR that is commercially licensed by the U.S. Nuclear Regulatory Commission Demonstrate safe and economical nuclear production of hydrogen and electricity
11
Goal: Design Methods Development & Validation
Ensure the software tools are available to enable the VHTR to be designed and licensed to achieve Generation IV Program standards & objectives.
12
Effect of Goals on VHTR Software…
Because VHTR design goals are ambitious: High efficiencies High operating temperatures Exceptional safety margins And licensing the VHTR will be a challenge, Software tools for VHTR must have demonstrated capability and low calculational uncertainty
13
VHTR Design Methods Development & Validation
For each plant, the R&D Process is based on… Identifying the most demanding scenarios for candidate plant design Isolating key phenomena in scenarios Determining whether analysis tools can be used to confidently analyze plant behavior scenarios (Validation) Performing R&D to upgrade analysis tools where needed Scenario Identification: Operational and a ccident s c enario that require analysis are identified PIRT: Important phenomena are identified for each scenario (P henomena I dentification & R anking T ables ) Validation: Analysis tools are evaluated to determine whether important phenom ena can be calculated Development: If important phenomena cannot be calculated by analysis tools, then further development is undertaken Analysis: The operational and accident scenarios that require study are analyzed No Yes
14
The Calculation Process…
Consists of seven steps It’s assumed to be equally likely that the VHTR will be either pebble-bed or block-type reactor a. Material Cross Section Compilation and Evaluation b. Preparation of Homogenized Cross Sections c. Whole - Core Analysis (Diffusion or Transport), Detailed Heating Calculation, and Safety Parameter Determination d. Thermal Hydraulic and Thermal Mechanical Evaluation of System Behavior f. Fuel Behavior: Fission Gas Release Evaluation g. Fission Gas Transport e. Models for Balance of Plant Electrical Generation System and Hydrogen Production Plant Requires the analysis tools to have reasonable† agreement with data for key phenomena. † Reasonable agreement: calculated value sometimes lies within data uncertainty band and shows same trends as data.
15
Includes Software such as:
16
What exists: Thermal-Fluid Analysis…
Partially validated systems analysis codes that cannot predict localized hot spots. Unvalidated computational fluid dynamics codes that will calculate the presence of hot spots—but the results cannot be trusted. Identification of only some of the important phenomena. Large incomplete validation data base.
17
Many of the Phenomena that Must Be Quantified & Analyzed Require CFD
Forced convection: turbulent behavior and mixing Mixed convection & free convection Analysis of flow behavior in plena and chambers Isolation of local hot spots a special need Coupled calculations required: that is, calculations that require coupled CFD codes and systems analysis codes (such as RELAP5) are quite important.
18
Thermal-Hydraulic Phenomena: CFD
Normal operation at full or partial loads Coolant flow and temperature distributions through reactor core channels (“hot channel”) Mixing of hot jets in the reactor core lower plenum (“hot streaking”) Loss of Flow Accident (LOFA or “pressurized cooldown”) Mixing of hot plumes in the reactor core upper plenum Coolant flow and temperature distributions through reactor core channels (natural circulation) Rejection of heat by natural convection and thermal radiation at the vessel outer surface Loss of Coolant Accident (LOCA or “depressurized cooldown”) Prediction of reactor core depressurized cooldown - conduction and thermal radiation
19
Role of CFD & Systems Analysis Codes
Relevance: Key phenomena will greatly influence the material temperatures at operational conditions and accident conditions. Some validation data are available and are being used to validate CFD & one-dimensional systems analysis codes. Present focus is on maximum channel coolant exit temperature at operational conditions and turbulent mixing in lower plenum. Importance: The software must be shown capable of calculating the maximum material temperatures to enable the VHTR to be licensed. The calculational uncertainty must be acceptably low to demonstrate the VHTR achieves its design claims and is licensable.
20
CFD: Commercial vs. Non-Commercial…
There are ongoing studies to evaluate non-commercial CFD codes. Commercial CFD codes are user-friendly and have an extensive V&V matrix. They can be used to analyze a wide variety of problems. Quite often, the opposite is true for non-commercial CFD codes since their development is often aimed at specific problems. However, non-commercial CFD codes are frequently state-of-the art and unexcelled in selected areas.
21
Approach for Achieving CFD Validation Objectives…
The process begins by using what is available in commercial CFD codes and by using RANS. As deficiencies are identified in commercial CFD codes, the R&D path is chosen based on magnitude of deficiency. Use LES and DES as necessary Define development, including need for non-commercial codes, as required.
22
Example*: To Quantify Capabilities of Selected Commercial CFD Codes…
__________ * Similar approach proposed for systems analysis software
23
CFD Tool Certification: for Use in Gen IV
Industrial quality standards recommended, e.g., ISO 9001:2000 standards or equivalent (for software quality, construction, and certification)—particularly for design calculations CFD certification should be based on V&V matrix that contains dominant phenomena for key Gen IV system scenarios CFD software V&V required for key phenomena identified in matrix. Acceptance based on degree of agreement with qualified data Extent of commercial and non-commercial CFD capabilities & need for further development/R&D can be demonstrated by specifying a series of International Standard Problems—with results published in journals.
24
Proposal: To Quantify Capabilities of Selected Commercial CFD Codes & Systems Analysis Codes…
Form PMB-sponsored committees with charter to define CFD code V&V matrix and systems analysis code V&V matrix based on Gen IV needs (the Standard Problem Committee). Sponsor forums, similar to those hosted by Coordinating Group for Computational Fluid Dynamics (1993-4) or the Stanford Olympics (1968), to define International Standard Problems for CFD. Publish results in journals. Use validation results as basis for achieving validation objectives. In summary, base approach on that used by US Nuclear Regulatory Commission when certifying their software tools to be used for reviewing the AP600 design for LWRs.
25
Analysis Practices Should be Well-Defined
For CFD software: Suggest following policy defined by Journal of Fluids Engineering (Vol 115, 1993) Benchmark study requirements: suggest following lead of: C. J. Freitas, “Perspective: Selected Benchmarks from Commercial CFD Codes,” Journal of Fluid Mechanics, 117, 1995, pp. 208 to 218. For systems analysis software—use directly practices and procedures implemented by USNRC.
26
Summary: CFD Code Certification
The CFD V&V matrix must be defined. Proposed that: a series of CFD International Standard Problems be specified, commercial and non-commercial CFD organizations be solicited to participate in an ISP Forum the analyses be performed using rigorous standards the analyses be published in the literature.
27
The Punch Line This effort is critical to mission success.
Current software and methods are not ready to perform design and analysis to the standard that will be required by the VHTR–considerable validation, and probably development, are required. The above conclusion also applies to present software capabilities to perform VHTR licensing calculations. Practices and procedures acceptable to community must be defined and implemented. This effort is critical to mission success.
28
For Validation Purposes: Some Data Sets Already Exist & Others Are Already Planned
Examples: IAEA benchmark problems. Integral facilities: HTTR, HTR-10 Various other experimental data recorded in GIF member experimental facilities US DOE sponsored data: matched-index-of-refraction (INL) and NSTF (ANL)
29
Summary of INL Experimental R&D
Ongoing R&D Plans
30
Thermal hydraulic benchmark experiments
Objectives: Provide benchmark data for assessment and improvement of codes proposed for NGNP designs and safety studies and Obtain better understanding of related phenomena, behavior and needs Two emphases: Spatial variations in local fission rate and material behavior will cause "hot channels" which may cause "hot streaking" in lower plenum – and possible structural problems Started in August 2004 Plan view of lower plenum for prismatic NGNP concept
31
Overview of INL experimental tasks
Scaling Needs Heated experiment concepts Isothermal lower plenum flow experiment concepts Design Fabrication Measurements Documentation Coolant channels Lower plenum Selection of first heated experiment Pebble bed exits? Design Fabrication Measurements Documentation Second heated experiment Other important geometries
32
Effect of turbulence model in CFD codes ("hot channel")
Run 618, turbulent, moderate q w Twall (K) v f Turbulence models Shehata CFD predictions of Richards, Spall and McEligot [2004] Experiment of Shehata and McEligot [1998] Selection of an adequate turbulence model is critical for CFD predictions
33
“Hot channel” experiment
Example of operating conditions for which code predictions are needed: non-dimensional heat flux Tabulated benchmark data are available for these conditions Vacuum vessel Traversing table Pressure transducer Multi-sensor hot-wire probe Thermocouples Power supply Upflow or downflow Heat exchanger Gas circulator Flow meter Potential concept to obtain benchmark data for turbulence modeling – low pressure to reduce buoyancy effects For normal operation Coolant channels have dominant forced convection with slight property variation Parameters for buoyancy, streamwise acceleration and heat flux are low relative to thresholds for importance Benchmark data are available to assess correlations in systems codes
34
Preliminary scaling studies -- lower plenum ("hot streaking")
One typical concept Desired: T (or C), V, v as fns{x,y,z}, Nu for surfaces, f or L, Str Parameters: Geometry, Rep, Rej, Vj/Vp, Ri or a buoyancy parameter and Tj/Tp for heated jets Conditions: Normal, reduced power, transients Dj/Dp 0.7 Dp p/Dp 1.7
35
Lower plenum experiments
Parameter ranges in normal full power operation Geometry: Posts under active core – pitch/D 1.7, height/D 7 Plenum Reynolds number, Replenum 24,000 (away from outlet) 3x106 (near outlet) "Mixed flow" Turbulent Jet-to-crossflow velocity ratio, (Vj/Vp) 50 (away from outlet) 0.6 (near outlet) "penetrating" "bent" Qualitative flow visualization with dye injection [McCreery, 2004] Model configuration Example of flow away from outlet
36
Potential experiment concepts for lower plenum flows
Purpose: assessment of momentum equation solution, scalar mixing and turbulence models for typical plenum geometry in limiting case of negligible buoyancy and constant properties Conceptual model designs (plan view) Measurements Flow visualization Particle imaging velocimeter – flow patterns, mean velocities, mixing of passive scalars Laser Doppler velocimeter – time-resolved mean velocity components and turbulence quantities in flows Possible flow paths in NGNP lower plenum
37
How does refractive-index-matching help?
Optical techniques will be used with models in our large Matched-Index of Refraction (MIR) flow system How does refractive-index-matching help? Optical techniques avoid disturbing the flow to be measured Typical approaches are LDV, PIV, PTV, flow visualization, PLIF, etc. Laser Doppler Velocimetry Particle Image Velocimetry Snell’s Law Unless the refractive indices are matched, the view may be distorted or impossible even with "transparent" materials and position measurements may be incorrect Example of application of refractive-index-matching Refractive index not matched (Marking is on back of beaker) Plexiglas model Fluid Laser beam Laser transmission optics Signal collection Internal plexiglas rods Not matched Matched (Rod is resting on the bottom of the beaker)
38
Benefits of INL MIR flow system
Refractive index-matching and optical measuring techniques allow flow measurements when flow geometries are complicated Most previous MIR experiments have been cm-scale; INL test section is about 0.6 m x 0.6 m x 2.5 m Example of application of refractive-index-matching (Marking is on back of beaker) Not matched Matched (Rod is resting on the bottom of the beaker) Apparatus to study fluid physics phenomena in idealized SNF canister for EM Science project Advantages Versatile - basic/applied research, internal / external / coupled flows Non-intrusive, undistorted measurements of flow and transport m-scale to building scale experience Good spatial and temporal resolution Benchmark measurements
39
Initial INL MIR experiment on lower plenum flow for assessment of CFD tools
Some desired features Represent generic features of flows near LP outlet (crossflow) and away from outlet (no crossflow) Well-defined geometry; ratios as in NGNP point design Turbulent flow in jet entry ducts Crossflow in "mixed flow" or "turbulent" regime [Zukauskas, 1972] Jet velocity ratio: ½ < (Vjet/Vplenum) < "large" Limited domain to ease initial CFD modeling Measurements concentrating in important regions U, V, W, u, v , w , uv, etc. as function of x, y, z Flow visualization Mixing of passive scalars Inlet flow quantities for boundary conditions
40
Fabrication sketch of model
Jet inlet ducts Milestones/deliverables Fabrication sketches (for code developers) Initial measurements Proposed for later years* Measurements and comparisons Documented databases Technical papers and reports *Depending on funding from FY US Gen IV program Sep 2005 H Support posts Flow Dimensions: Dpost = 31.8 mm (1.25 in.), p/Dp = 1.7 Djet/Dp = 0.7, Hplenum/Dp = 6.85
41
Example of comparable PIV data from INL MIR flow system (idealized SCWR coolant channels in US/Korea I-NERI project [2005]) Flow Streamwise velocity field
42
Supporting PIV measurements by Prof. B.L. Smith at Utah State U.
Purpose = determine the minimum flow rate (Reynolds number) to achieve "mixed flow" in the crossflow of the INL lower plenum model Experimental configuration Geometry Measurements
43
PIV measurements Re 400 Deduced contours of stream "Raw" photograph
function, steady laminar flow Results include instantaneous and mean velocity fields, mean stream functions, Reynolds stresses and 2-D "turbulence kinetic energy"
44
Preliminary classification
Re > 2 x 105 [Zukauskas, 1972] Steady laminar flow Unsteady laminar flow Mixed partially-turbulent flow Mixed turbulent flow Turbulent flow Visualizations and tabulations on web site (
45
Potential experiment concept for heated flows in lower plenum
Objective: assessment of codes predicting thermal mixing under influence of buoyancy for typical plenum geometry Fluid: heated air or water with salt solution for density variation Measure temperature field with miniature multi- sensor probes of Vukoslavcevic and Wallace [U. Maryland, NERI project] and possibly velocity and turbulence fields Possible model
46
INL thermal-hydraulic experiments - plans
FY-05 Complete fabrication sketches for MIR LP experiment Complete fabrication and installation for first MIR LP experiment = jet inflow without imposed crossflow (i.e., away from LP outlet) Continue evaluation of experiment concepts for heated flows Submit deliverable (fabrication sketches) Initiate measurements with MIR LP experiments (milestone) FY-06 Complete measurements for MIR LP experiment without crossflow; document Design and cost system to provide crossflow in MIR LP model Fabricate and install system to provide crossflow in MIR LP experiment Initiate measurements with imposed crossflows Complete fabrication sketches for second MIR LP experiment Continue design and selection of experiments for Turbulence and stability data for vertical cooling channels Heated flows in lower plenums Initiate design of MIR experiments on exit flows in pebble beds (if funding permits)
47
Other recommended INL experiments for code validation
Core heat transfer experiments Turbulence and stability data from vertical cooling channels Bypass flow studies Exit flows in pebble beds Upper and lower plenum fluid behavior experiments Fluid dynamics of lower plenums Heated flows in lower plenums Interactions between hot plumes in an upper plenum and parallel flow instabilities Air ingress experiments: heat transfer and pressure drop of mixtures of air and helium Larger scale vessel experiments: to examine the behavior in the core, in the plenums and the interactions between them
48
VHTR - Scalability of the ANL Natural Convection Shutdown Heat Removal Test Facility (NSTF)
Identified major scaling parameters & phenomena and constructed a semi-analytical scaling model for air cooled RCCS Evaluated available RCCS designs, reviewed archival NSTF data, and identified needs for additional sets Constructed CFD models of available RCCS designs and NSTF and performed accident condition analyses which showed strong 3-D effects and heat transfer differences with existing 1-D correlations
49
R&D Plans are Based on a Preliminary Phenomena Identification & Ranking (PIRT)
Based on prioritization of scenarios and phenomena : Identified by experienced gas-cooled system designers Engineering judgment Aimed at requirements for performing reasonable calculations of plant behavior for: Operational conditions (rated power): neutronics and working fluid behavior Pressurized conduction cooldown transient scenario (PCCS) Depressurized conduction cooldown transient scenario (DCCS) including possible air/water ingress Focus is on maintaining integrity of fuel, core, and reactor vessel.
50
PIRT: Components PIRT based on following components:
51
Example: Depressurized Conduction Cooldown Scenario has 3 Phases
Blow down (depressurization), Molecular diffusion: following blowdown, there is not sufficient helium left in reactor vessel to have natural circulation—however, air from confinement moves into reactor vessel via diffusion. Heat removal dominated by radiation and conduction heat transfer. Natural convection: Sufficient air has moved into reactor vessel via diffusion that natural circulation is initiated and convective cooling becomes an important ingredient.
52
Blowdown—Sample of Phase Description…
Event initiated by rupture of largest system pipes System depressurizes Helium working fluid discharged into volume that surrounds reactor “Rapid” heat-up of core occurs by the loss of forced convection Graphite dust from reactor core is transported to volume (the reactor confinement building) that contains reactor The confinement relief valves lift and gas is discharged to environment. Filters minimize distribution of dust to environment. Relief valves close when pressure has been reduced sufficiently.
53
Phenomena are ranked… High if important.
Medium if phenomena may have supporting role in system behavior or may be important and thus should be carefully evaluated. Low if can be neglected.
54
A Sample PIRT Table (for core)…
Phenomena HPCC LPCC LC 1 2 3 flow distribution H M heat transfer (forced convection) heat transfer (mixed and free convection) pressure drop (forced convection) pressure drop (mixed and free convection) thermal mixing and stratification jet discharge thermal striping bulk CO reaction molecular diffusion Fluid properties (gas mixture) Graphite oxidation (PBR)
55
CFD Validations Required Based on PIRT…
Region of System Operational Conditions Depressurized Conduction Cooldown Pressurized Conduction Cooldown Inlet Plenum IP1: Validation of CFD mixing calculation during transient & action of plumes on pressure heat boundary.. Core CO4: Characterization of hot channel temp and fluid behavior at operational conditions. CD3: Validation and calculation of air ingress and potential water ingress behavior into reactor vessel and core region. CP1: Validation of systems analysis codes to demonstrate capability to predict thermal and hydraulic behavior. Outlet Plenum PO1: Validation of CFD mixing using mixed index refraction (MIR) facility data & data available in literature. Perform calculation of fluid behavior with validated code. PD1: Validation of CFD mixing during operational transients and effect on turbine operational characteristics. Perform calculation of fluid behavior. PP1: Validation of CFD mixing during operational transients and effect on turbine operational characteristics. Perform calculation of fluid behavior. Reactor Cavity Cooling System RO1: Validation of natural convection characteristics in cavity at operational conditions. RO2: Characterization of natural convection characteristics in cavity at operational conditions. RD1: Validation of heat transfer & convection cooling phenomena present in reactor cavity and via RCCS. RP1: Validation of heat transfer & convection cooling phenomena present in reactor cavity and via RCCS.
56
Preliminary PIRT Being Used to Define CFD Related NGNP R&D…
Most applications are at low Mach numbers—thus CFD techniques used for incompressible flows will be most used. A few applications for compressible, high-velocity CFD software are foreseen: Perhaps blowdown Pressure pulse propagation during turbomachinery transients, e.g. compressor surges.
57
Conclusions—Methods R&D
Passage of Energy Bill has authorized the project for siting the VHTR at INL, execution requires appropriations. Candidate VHTRs are both prismatic & pebble-bed designs. The VHTR will: Make possible the efficient and cost effective production of either or both electricity and hydrogen Excel in safety and reliability Discharge waste that is suitable for long term disposal Provide a bridge to a low-emissions economy based on water as our primary transportation fuel The current Generation IV/VHTR methods R&D is centered on ensuring the software tools are available to enable the VHTR to be designed and licensed to achieve Generation IV Program standards & objectives.
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.