A Few Thoughts on Computational Nuclear Science and the New Nuclear - - PowerPoint PPT Presentation

A Few Thoughts on Computational Nuclear Science and the New Nuclear Energy Era

Dr Cassiano R E de Oliveira Department of Chemical and Nuclear Engineering The University of New Mexico cassiano@unm.edu

II SEN – I ISNE Rio de Janeiro, August 13-17 2012

Your Speaker

• B.S. Physics (PUC-RJ, 1976), M.Sc. Nuclear Engineering (COPPE/UFRJ,

1981), Ph.D. Nuclear Engineering (University of London, 1987)

• Researcher at IEN/CNEN from 1979-1982
• Currently a Professor at the Department of Chemical and Nuclear

Engineering at the University of New Mexico.

• Prior to joining UNM worked for 14 years at Imperial College London, UK

and 4 years at the Georgia Institute of Technology, US

• His research interests concern computational methods development and

their applications. These have spanned nuclear engineering, engineering thermofluids flow, global ocean circulation, cloud radiative transfer, optical tomography, turbomolecular flow, optimization and data assimilation problems

• His research code EVENT has been adopted by Rolls Royce Marine Power

for shielding applications. He was a co-developer of the FETCH code for non-linear criticality excursion analysis and ICOM – the Imperial College Global Ocean Circulation Model

Obligatory Equation – but not just any Equation

Rate of change of neutrons population

• = source of neutrons

( ) leakage ( ) destruction ( )

Neutron Population Balance: Boltzmann Transport Equation

1 v (r,,E,t) t = d

• 4
• d

E

• s(
• ,

E E)(r,

• ,

E ,t) + s(r,,E,t) .(r,,E,t) t(r,E)(r,,E,t)

Making yourself useful….

New Mexico, USA

Beep, beep Land of Enchantment

Nuclear Energy and New Mexico

6

Uranium mining United Nuclear Los Alamos National Laboratory LES enrichment plant Palo Verde(AZ) 16% PNM electricity Waste Isolation Pilot Plan Trinity Sandia National Laboratories

Trinity

UNM NE Program

UNM has the only Nuclear Engineering Program in the State of New Mexico

• Program came into existence in 1960 as the Nuclear Laboratory
• Standalone graduate Department of Nuclear Engineering created in 1965

Combined with Chemical Engineering in 1972, following nuclear industry downturn

• Department of Chemical and Nuclear Engineering
• Undergraduate degree in NE added

AGN-201M 5W Homogeneous Thermal Nuclear Reactor (1966)

• Teaching and Experiments for UG and graduate program, training

Historically Strong Ties with Los Alamos and Sandia National Laboratories

• Education, Research and Outreach
• Dating to Inception of Program

Academic Program

• Degree Programs:

– B.S. Nuclear Engineering (4 years, 130 credits) – M.S. Nuclear Engineering (2 years, 30 credits) – Ph.D. Engineering (NE) (3-5 years, 48 credits) – M.S. NE Concentration in Medical Physics (2 years, 33 credits)

• 2010-11 Enrollments:

– 34 Undergraduates (Sophomores, Juniors and Seniors) – 19 Graduate M.S. – 21 Graduate Ph. D.

• Course Coverage (offered via live ITV and VOD across state):

Reactor Physics and Technology, Radiation Physics and Interactions, Fluid and Thermal Sciences, Instrumentation and Reactor Lab, Applied Mathematics, Radiation Transport, Numerical Methods, Special Topics Medical Physics: Joint With Department of Radiology (SOM) and UNM Cancer Center

9

NE Faculty

Full time:

• 3 Professors
• 1 Associate Professor
• 2 Assistant Professors
• 1 Lecturer

Part time:

• 2 Research Professors
• Multiple Adjunct Professors (Sandia and LANL)
• UNM-National Laboratory Professor
• 1 Emeritus Professor

Nationally-recognized Faculty with visible research programs

10

Faculty

Gary Cooper

Mohamed El Genk

Adam Hecht Anil Prinja

Cassiano de Oliveira

11

Ed Blandford Bob Busch Ed Arthur

Faculty Research Areas

Diverse research interests across Nuclear Engineering disciplines:

• Nuclear Reactor Technology and Safety, Space Nuclear Power,

Energy Conversion

• Nuclear Instrumentation and Detection for Nonproliferation and

Safeguards, Fusion Plasmas

• Stochastic and Deterministic Radiation Transport Methods for Active

and Passive Interrogation, Reactor Physics, Nuclear Criticality Safety, Weapons Safety, Space Radiation

• Uncertainty Quantification Techniques in Single- and Multi-physics

Nuclear Applications Two organized Research Units:

• Center for Nuclear Nonproliferation Science and Technology

(CNNST)

• Institute for Space and Nuclear Power Studies (ISNPS)

Strategic Planning Framework

13

applied radiation science for a healthier and more secure world radiation interaction with materials

UNM School of Engineering

ENERGY — MATERIALS — HEALTH

safe and sustainable energy for the earth and beyond

Nuclear Engineering at UNM

reactor system design and safety nuclear non- proliferation and safeguards

areas of strength major research problem areas societal mission

nuclear fuel cycle nuclear medicine

Safety and Risk Assessment Nuclear Instrumentation and Detection Computational Radiation Transport Advanced Computational Science Thermal Hydraulics Nuclear Reactor Technology Material Engineering Science Radio- chemistry

Outline

• Motivation for Computational Nuclear Science
• Examples of Computational Material Science

Research

• Building confidence on Computational Nuclear

Science

• Afterthoughts

Food for Thought (I)

“A computer lets you make more mistakes faster than any invention in human history — with the possible exceptions of handguns and tequila.” Mitch Ratliffe, Technology Review, April, 1992

Food for Thought (II)

“The paper-fueled, ink-moderated, hot-air cooled reactor is built on time, at the specified budget, operates at 100% efficiency and produces no nuclear waste” Enrico Sartori (NEA/Nuclear Data Bank)

Food for Thought (III)

“No tool is so clever that it cannot be used by an idiot” Confucius

The Mantra

• Duty of methods developers is to give

back physics to physicists/engineers

• Expectation from numerical simulations is

not only to describe, but also help to prescribe and understand

• One should have confidence in the

methods and results

• Never throw away information/data

Motivation

• Maturity of simulation tools with hardware

advances means that solutions to hitherto intractable problems in nuclear science and related areas are now affordable

• However needs have also changed:

increasingly higher-fidelity simulations are now required and in larger quantities

• Multiphysics and multiscale simulations are

also becoming the order of the day

Time and accuracy is of essence

• S i m u l a t i o n m e t h o d s h a v e t o b e

computationally efficient ie use available resources efficiently and expediently

• Thus the general quest is for practical

solution strategies for very large sets linear and non-linear algebraic system of equations

• But confidence is also needed on simulation

tools →V&V and UQ

• Nuclear energy has always been in the

forefront of computational science since the Manhattan project

• Unique blend of physics, engineering and

computer science

• Major contributor to advances in mathematical

and numerical computer techniques

• Current reactors designed with well-thought

but heavily approximated numerical procedures

Computational Science and Nuclear Energy

Need for Advanced Modelling and Simulation

• New programs and advanced reactor

concepts require smaller operational margins and detailed safety analyses which can only be accomplished via numerical simulations

• Advanced numerical simulations the most

efficient way to achieve this

• Thus need for high-fidelity, first principles

analyses

Hardware Infrastructure

A R C H I T E C T U R E S

Applications

Run-up to the era of simulation

scientific models numerical algorithms computer architecture scientific software engineering

(dates are symbolic) 1686 1947 1976 1992

Enrico Fermi

• Helped create modern physics
• Invented, but of course not named,

the present Monte Carlo method when he was studying the moderation of neutrons in Rome.

• He did not publish anything on the

subject, but he used the method to solve many problems with whatever calculating facilities he had, chiefly a small mechanical adding machine.

The Monte Carlo Method

FERMIAC

• The Monte Carlo trolley, or

FERMIAC, was an analog computer invented by physicist Enrico Fermi to aid in his studies of neutron transport.

• The FERMIAC employed the Monte

Carlo method to model neutron transport in various types of nuclear systems.

Computational Trends: Hardware

• Moore’s Law (1:2 Every 18 Months, 1:10 Every 5

Years)

• → Today’s Supercomputer Problem Will Reach PC

in 15 Years

• Current PCs: 107 Elements/Cells → 108 in 5 Years
• Current SCs: 1010 Elements/Cells → 1011 in 5

Years

• Programmable GPU/FPGA

Product Development Cycle

Preliminary Design Product Definition Detailed Design Prototype Testing Production Time

Possibility Of Change Information Content Cost of Change

Effect of Computational Science

New Aeroplane: $20B, New Car: $4B, New Reactor: $?

(courtesy of Kord Smith, Studsvik Scandpower, Inc.)

(courtesy of Kord Smith, Studsvik Scandpower, Inc.)

Multiphysics and Reactor Analysis

• Non-linear multiphysics coupling in reactor

analysis recognized from the onset

• Coupled analyses performed with simpler

physical models

• Current trend/need is to make use of existing

and future computational software and hardware resources to enable tighter coupling

• Main physics to be coupled: neutronics,

thermo-fluid flow, solid heat transfer, structural mechanics

(courtesy of Kord Smith, Studsvik Scandpower, Inc.)

Trials and tribulations of nuclear fuel

Issues and Challenges in Material Science

Material Issues Challenges Near-term goals

Structural materials Hardening/embrittlement Phase stability/RIS/IASCC Creep/swelling/fatigue He embrittlement Microstructural evolution under radiation in order to predict properties Fe-C interatomic potential Radiation-induced hardening via DD models Assess Gen. III alloys Fuels Modeling actinides Complex chemistry Fission products/damage Fuel performance code Physics-based fuel performance code with thermodynamics and thermophysical props. MOX phase stability including MA/fission products Model UO2 microstructural evolution (incl. FP transport) Waste Forms Structure-property Impurities/stoichiometry Disorder Bonding Phase stability under looooooong-term irradiation Develop scientific basis for I, Tc sequestration (more tractable than current once- thru fuel cycle waste immobilization) Functional materials Electrical resistivity Optical properties Embrittlement Suitable interatomic potentials

• Comput. framework for

multiscale

Drawbacks of Current Fuel Performance Codes

• Design
• Not object oriented
• Do not run parallel
• Models
• Empirical correlations, unreliable

extrapolations

• Too material specific
• No uncertainty evaluation
• Input/Output
• No user friendly interface
• Rudimentary post processing
• Impossible to interconnect

Advanced Models, Simulations, and Fuel Performance Code Advanced Models and Simulations Advanced Fuel Performance Code

• Continuum-scale: Thermo-mechanical properties
• Meso-scale: Microstructural evolution, Species mobility
• Atomic-scale: Defect formation free energy, Irradiation effects
• Electronic Structure: Structural stability, elastic constants
• Fission products kinetics and concentration
• Heat transfer simulations
• Diffusion of species (gas and fission products) simulations
• Chemical reactions simulations

Incorporation of the Advanced Models

Ideal Advanced Fuel Performance Code Design Template and Modules

Output (XML) Input (GUI) Main Driver Users (people or other codes)

Simulations

External Databases Models

• Continuum
• Atomistic
• Electronic

Structure Internal Database

Heat Transfer Thermo- Mechanics Fluid Flow Neutronics Mass Transport

Solvers Users Other Codes

AFPC
 AFPC


Users only see the red boxes

Summary

• Current fuel performance codes limited to

experimental data.

• Codes are currently too material specific,

not of use to advanced fuel cycles programs for most part.

• Codes need broader applications to fuel

processing, in addition to fuel performance.

• Can build upon success of ASCI program.

Diffusion of innovation is useful to understand how ideas advance

The Gap! De Developer elopers of s of new T new Tec echnolog hnology y User Users of s of (w (was) as) new tec new technolog hnology y

“So easy, even a caveman could do it” - Geico

(courtesy of Kord Smith, Studsvik Scandpower, Inc.)

(courtesy of Kord Smith, Studsvik Scandpower, Inc.)

Building Confidence on Simulations

Main thrust areas in computational science:

• Novel, advanced numerical discretization and

solution techniques

• Multiphysics coupling
• Validation and Verification (V&V)
• Uncertainty Quantification

Validation and Verification

• How should credibility of Modeling and

Simulation be critically assessed?

– Verification: assessment of accuracy of simulation by comparison with known solution (mathematical issue) – Validation: assessment of accuracy of simulation by comparison with real data (physics issue)

Role of V&V

Verification

Verification

Food for Thought (IV)

“Reports that say that something hasn't happened are always interesting to me, because as we know, there are known knowns; there are things we know we know. We also know there are known unknowns; that is to say we know there are some things we do not know. But there are also unknown unknowns — the ones we don't know we don't know.” Former US Secretary of Defence, D. Rumsfeld

Food for Thought (V) “I know one thing, that I know nothing” (Socrates)

Uncertainty Quantification

• Uncertainty Quantification (UQ) is the process of

identifying, characterizing, and quantifying those factors in an analysis that could affect the accuracy

• f the simulation results. Sources of uncertainties

may arise during:

• Construction of the computational model
• Formulation of the mathematical model
• Computation of the simulation results
• Uncertainties propagate from their various sources

through computational models to the simulated results

Uncertainty Quantification

Total Uncertaintity Quantification

Where do I start?

• Ordinary differential equations
• Partial differential equations
• Statistics
• Numerical Analysis

– Root finding – Systems of linear equations – Interpolation – Numerical Integration – Ordinary Systems of Equations – Optimization

Where do I start?

• MATLAB→ numerical computing environment

and fourth-generation programming language

• COMSOL→ finite element analysis, solver and

simulation software/ FEA Software package for various physics and engineering applications, especially coupled phenomena, or multiphysics

• Programming Languages

– C++ – Java – Fortran

Final Thoughts

• Present and future science requirements will demand

ever more complex non-linear simulations

• Multiphysics strategies will have to be well-thought
• Brute force solution will always outpace hardware

development

• Intelligent simulation tools are the only way forward due

their optimum use of computational effort and user transparency

• Interdisciplinarity is the way forward
• Verification and Validation and Uncertainty Quantification

will become an integral part of computational science

• Industry involvement essential

Final Thoughts

• The failure to develop a

unified experimental & computational program could be a key limiting factor in developing advanced simulation tools

Nature Simulation Theory Experiment

Verification Measurement Models Validation Achilles’ Heel Holy Grail

Words of Wisdom

• Please join LAS - Latin American Society
• f ANS.
• Consider also the ANS and the Institute of

Nuclear Materials Management (INMM)

• Don’t forget about ABEN
• Tap into the past to learn and preserve the

knowledge

• Think big: see the world!

Personal Thoughts

• Nuclear Energy suffers from the “nobody likes me” syndrome
• Most of it is justified and self inflicted:

– Costs – Industry inevitable association to nuclear weapons – Three Mile Island, Chernobyl – Japan: Monju (1993), Tokaimura (1999) and Fukushima (2011)

• Brazil’s hydroelectric lakes produce more methane than UK

greenhouse gases emissions

• Transmissions lines lose significant amount of energy along the way
• Biofuels were a bad idea way back in the 70’s and they still are
• Coal releases more radioactivity into the environment that anything

else (and mercury etc)

• Nuclear will never produce the bulk of electricity

Final, final thought

“Students: apply yourself, work hard and get good grades. Otherwise you will end up like me” (Cassiano de Oliveira)