2

For Accelerating Technology Development

National Labs Academia Industry Advisory Board

Rapidly synthesize optimized processes to

identify promising concepts

Better understand internal behavior to reduce time for

troubleshooting

Quantify sources and effects of uncertainty to guide testing &

reach larger scales fasterStabilize the cost during commercial deployment

3

Develop new computational tools and models to enable industry to more rapidly develop and deploy new advanced energy technologies Base development on industry needs/constraints Emphasis on supporting post-combustion capture technologies

• Limited work on oxycombustion system• General tools capable of supporting all types of carbon capture technology development

Demonstrate the capabilities of the CCSI Toolset on non-proprietary case studies Examples of how new capabilities improve ability to develop capture technology

Deploy the CCSI Toolset to industry

Collaborate with industry to use the tools to support technology development & scale up

Goals & Objectives of CCSI (2011-2016)

4

Maximize the learning at each stage of technology development

Early stage R&D Screening concepts Identify conditions to focus development Prioritize data collection & test conditions

Pilot scale Ensure the right data is collected Support scale-up design

Demo scale Design the right process Support deployment with reduced risk

CCSI Toolset: New Capabilities for Modeling & Simulation

4

2016 R&D 100 Award

5

CCSI Toolset Module Connectivity

Process Models

Solid In

Solid Out

Gas In

Gas Out

Utility In

Utility Out

Basic Data Submodels

Carbon Capture Process

GHX-001CPR-001

ADS-001

RGN-001

SHX-001

SHX-002

CPR-002

CPP-002ELE-002

ELE-001

Flue GasClean Gas

Rich Sorbent

LP/IP SteamHX Fluid

Legend

Rich CO2 Gas

Lean Sorbent

Parallel ADS Units

GHX-002

Injected Steam

Cooling Water

CPT-001

1

2

4

7

8

5 3

6

9

10

11

S1

S2

S3

S4

S5

S6

12

13

14

15

16

17

18

19

21

24

2022

23

CYC-001

Process Synthesis, Design & Optimization

CFD Device Models

Process Dynamics and Control

6

High Fidelity Process Models for Carbon CaptureBubbling Fluidized Bed (BFB) Model Variable solids inlet and outlet location Modular for multiple bed configurations

Moving Bed (MB) Model Unified steady-state and dynamic Heat recovery system

Compression System Model Integral-gear and inline compressors Determines stage required stages, intercoolers Based on impeller speed limitations Estimates stage efficiency Off-design and surge control

Solvent System Model Predictive, rate-based models

Membrane System Model Hollow fiber Supports multiple configurations

0 0.2 0.4 0.6 0.8 10.04

0.06

0.08

0.1

0.12

0.14

z/L

yb (k

mol

/km

ol)

CO2H2O

0 0.5 1200

400

600

800

1000

kmol

/hr

z/L

Solid Component Flow Profile of MB Regenerator

Bicarb. Carb. H2O

7

Framework for Optimization, Quantification of Uncertainty and Surrogates

D. C. Miller, B. Ng, J. C. Eslick, C. Tong and Y. Chen, 2014, Advanced Computational Tools for Optimization and Uncertainty Quantification of Carbon Capture Processes. In Proceedings of the 8th Foundations of Computer Aided Process Design Conference – FOCAPD 2014. M. R. Eden, J. D. Siirola and G. P. Towler Elsevier.

SimSinterStandardized interface for simulation software

Steady state & dynamic

SimulationAspen

gPROMSExcel

SimSinter Config GUI

Res

ults

FOQUSFramework for Optimization Quantification of Uncertainty and Surrogates

Meta-flowsheet: Links simulations, parallel execution, heat integration

Sam

ples

SimulationBased

OptimizationUQ

ALAMO Surrogate

Models

TurbineParallel simulation execution management

systemDesktop – Cloud – Cluster

iREVEALSurrogate

ModelsOptimization

Under UncertaintyD-RM

BuilderHeat Integration Data ManagementFramework

8

Discrete decisions: How many units? Parallel trains? What technology used for each reactor?

Continuous decisions: Unit geometries Operating conditions: Vessel temperature and pressure, flow rates,

compositions

Carbon Capture System Superstructure Using Surrogate Models

Surrogate models for each reactor and technology used

9

Sim

ulta

neou

s Pr

oces

s O

ptim

izat

ion

& H

eat I

nteg

ratio

n

w/o heat integration Sequential Simultaneous

Net power efficiency (%) 31.0 32.7 35.7Net power output (MWe) 479.7 505.4 552.4Electricity consumption b (MWe) 67.0 67.0 80.4Base case w/o CCS: 650 MWe, 42.1 %Chen, Y., J. C. Eslick, I. E. Grossmann and D. C. Miller (2015). "Simultaneous Process Optimization and Heat Integration Based on Rigorous Process Simulations." Computers & Chemical Engineering. doi:10.1016/j.compchemeng.2015.04.033

• Industrial Collaborations• 7 CO2 Capture Program projects $40MM+ in total project value (TRL 3-7)• 6 additional external industrial agreements (executed or in progress)

- Cooperative R&D Agreement: GE, ADA-ES, Ion, TCM, SINTEF- Contributed Funds Agreement: COSIA ($500k)

• Includes enabling capture technology support:- Aerosol, dynamic characterization, turndown, advanced process control

• Optimal Design of Experiments• Improves model while optimizing experimental data generation• Applicable to lab through large pilot scale

• Solvent modeling framework• Fundamental characterization of solvent, device and system• Collaboration with International Test Center Network (ITCN) & SINTEF

CCSI2 Industrial Collaboration & Contributions

10

11

What was missing in the previous DoE? Mainly designed using a space-filling approach without considering the

output space When designed considering the output space, feedback from the

experimental data are not leveraged to update the DOE How to solve these issues? Develop DOE by taking into consideration the output space by using a

preliminary process model Use a sequential approach to improve DOE (through improvement of

process model, etc.-more on this later) as experimental data are obtained Being employed at NCCC and TCM for Test Campaigns in 2017

CCSI using Optimal Design of Experiments to inform test plans

11

12

Typical Steady-State DOE

Abso

rbe

rR

egen

erat

or

CCSI DOE Approach and Comparison with CCSI Model

13

Developing Detailed, Predictive Models of Solvent-Based Capture Processes

Steady-State and DynamicProcess Model

Process Sub-Models

Kinetic model

Hydrodynamic Models

Mass Transfer Models

Properties Package

Chemistry Model

Thermodynamic Models

Transport Models

UQ

Pilot/Commercia

l Scale Data

WWC/Bench/Pilot Scale Data

Lab Scale Data

UQ

Process UQ

Measurement Uncertainty

Measurement Uncertainty

Measurement Uncertainty

14

CCSI Solvent Model is More Predictive Because It Utilizes Better Multiscale Models Developed with a Holistic Approach

20

40

60

80

100

120

0 0.2 0.4 0.6 0.8

Neg

ativ

e H

eat o

f A

bsor

ptio

n (k

J/m

ol

CO

2)

CO2 Loading (mol CO2/mol MEA)

Heat of Absorption Data* (30 wt% MEA and 40,80, and 120°C)

40°C, 30 wt% MEA

0.0001

0.01

1

100

0 0.2 0.4 0.6CO2

Part

ial P

ress

ure

(kPa

)

CO2 Loading (mol CO2/mol MEA)

80°C, 30 wt% MEA

0.0001

0.01

1

100

0 0.2 0.4 0.6

CO

2Pa

rtia

l Pr

essu

re (k

Pa)

CO2 Loading (mol CO2/mol …

Deterministic ModelStochastic Model

(Prior Parameter Distribution)Posterior Parameter

Distribution

Bayesian inference

15

In conjunction with a dynamic model, a greater amount of data can be collected within the same amount of time during pilot plant testing

Demonstrated at NCCC Dynamic data reconciliation

enables model validation

Increasing Learning at Pilot Scale with Dynamic Data Collection

4900

5300

5700

6100

6500

0.0 0.5 1.0 1.5 2.0

Lean

Sol

vent

to

abso

rber

(kg/

h)Time (h)

65.0

70.0

75.0

80.0

85.0

90.0

0.0 0.5 1.0 1.5 2.0

CO

2ca

ptur

ed

(%)

Time (h)

1900

2100

2300

2500

2700

0.0 0.5 1.0

Flue

gas

flow

rate

to

abso

rber

(kg/

h)

Time (h)

86889092949698

0 0.5 1CO

2ca

ptur

ed (%

)

Time (h)

Absorber Stripper

16

CCSI Toolset Support Advanced Process Control For Carbon Capture

Initial Flue Gas Ramp

Double-Pole Result (New)

Fast Response Slow Response

Flue

Gas

Flo

w(k

mol

/hr)

Time (sec)

17

CCSI Toolset supports oxy-combustion processes & optimization

4. Pollution Controls5. CO2 Compression Train

1 2

3

4

5

Dowling, A. W.; Eason, J. P.; Ma, J.; Miller, D. C.; Biegler, L. T., Equation-Based Design, Integration, and Optimization of Oxycombustion Power Systems. In Alternative Energy Sources and Technologies: Process Design and Operation, Martín, M., Ed. Springer International Publishing: Switzerland, 2016; pp 119-158.Ma, J.; Eason, J.; Dowling, A. W.; Biegler, L. T.; Miller, D. C., Development of a First-Principles Hybrid Boiler Model for Oxy-Combustion Power Generation System. International Journal of Greenhouse Gas Control 2016, 46, 136-157.

Zone 1

Zone 2

Zone 3

Zone 4

Zone 5

Zone 6

Zone 7

Zone 8

Zone 9

Flue GasExit Plane

18

CFD Models

Superstructure Optimization

Oxycombustion System Optimization Framework

Solvent Blend Model

CCSI TOOLSET PRODUCTSAVAILABLE AS OPEN SOURCE SUMMER 2017

19

CCSI2 employs a multi-scale modeling framework (materials through systems) based on fundamental, scientific principles, providing “glass-box” understanding of capture technology

Interconnectivity of scale, physics and chemistry permits well-informed modeling framework with full quantification of uncertainty

UQ leveraged to improve model prediction and data generation for lower risk CCS scale up, demo and commercialization

High throughput, intelligent computational screening informs most effective R&D pathways

Multiple active collaborations with world-class industrial partners and test centers

CCSI2 supports the full commercialization pathway for the following CCS technology platforms: Post combustion solvents, sorbents & membranes; oxycombustion; & compression

Open source suite of tools and methodologies can help all capture technology development projects maximize their learning and decrease technical risk

CCSI2 and Toolset Summary

For more informationhttps://www.acceleratecarboncapture.org/

David C. Miller, Ph.D., Technical DirectorDavid.Miller@netl.doe.gov

Michael S. Matuszewski, Associate Technical DirectorMichael.Matuszewski@netl.doe.gov

21

Legend: 5 (Highest capability), 1 (Low capability), - (No capability) Competition

Description CCSI Aspen ANSYS Mode Frontier

CCSI Advantage

Cos

t Annual cost for a single user within a company. $10k $25k+ $25k+ $10k+ The CCSI Toolset is available for a flat fee of $10k/year, whereas the competition is licensed per concurrent user. Competition cost is estimated based on market research. The CCSI Toolset licensing structure is designed to help companies get the maximum benefits across the entire organization.

Annual cost for unlimited use within a company. $10k $1,000k+ $1,000k+ $500k+

Key

Fe

atur

es

Specifically designed to support development and scale up of new technology 5 2 2 -The CCSI Toolset is the only fully integrated and validated suite of computational tools and models that are specifically designed to support the development and scale up of carbon capture technologies.

Specifically designed for carbon capture 5 - - -Advanced machine learning tool for determining simple algebraic functional form and parameters of data 5 - - -

Applicable to full range of chemical processes 4 3 3 3

Dat

a M

anag

emen

t Graphical workflows for setting up and analyzing results 5 3 3 3

The CCSI Toolset has a fully integrated data management system to enable provenance tracking (i.e, dependencies) from experimental data to fully optimized process.

Data management framework that includes provenance relationships among models and experimental data 5 3 2 -

Simultaneous property and process parameter estimation 5 - - -Property estimation 5 3 - 2Solvent blends properties estimation tool that minimizes data requirements for new blends 5 3 - -Sorbent properties model estimation/fitting tool 5 - - -

Proc

ess

and

Dev

ice-

Scal

e M

odel

s

Dynamic moving bed reactor process model for gas/solid contacting 5 * - -All solid-gas process models are dynamic, non-isothermal, rigorous, flexible, capable of simulating embedded heat exchangers, and readily configurable for general absorption/ adsorption/ desorption processes. Highly accurate yet computationally efficient process models available in the CCSI Toolset can be readily used for UQ, process control, optimization and scale-up studies. Industry is already teaming with CCSI to utilize the Toolset.

#Aspen supports dynamic simulation of solvent-based capture systems, but does directly not support dynamic rate-based simulation.

*A user can develop these models using this software, but the software does not have pre-built, validated models available as part of its commercial offerings.

Dynamic bubbling fluidized bed process model for gas/solid contacting 5 * - -Dynamic compression process models tailored for CO2 5 * - -Membrane models 5 - - -Dynamic solvent models with rate-based mass transfer 5 # - -High resolution "filtered" models for hydrodynamics and heat transfer in a fluidized bed 5 - * -Hierarchically validated device-scale fluidized bed model at multiple scales 4 - * -

Oxycombustion boiler model 5 - * -

Adv

ance

d Pr

oces

s C

ontr

ol

Nonlinear Model Predictive Control 5 2 - -

The CCSI Toolset includes the ability to develop fast and accurate nonlinear dynamic reduced models from generic process models (not limited to specific simulation software) and include dynamic uncertainty quantification. Supports advanced process control using computationally efficient algorithms with the flexibility to use open-source nonlinear programming solvers and rigorous disturbance estimation methods. Aspen’s nonlinear control algorithms are non-generic and based on dated and inefficient sub-space identification methods.

Capabilities of the CCSI Toolset

22

Legend: 5 (Highest capability), 1 (Low capability), - (No capability) Competition

Description CCSI Aspen ANSYS Mode Frontier

CCSI Advantage

Adva

nced

Pro

cess

C

ontr

ol

Nonlinear Model Predictive Control 5 2 - - The CCSI Toolset includes the ability to develop fast and accurate nonlinear dynamic reduced models from generic process models (not limited to specific simulation software) and include dynamic uncertainty quantification. Supports advanced process control using computationally efficient algorithms with the flexibility to use open-source nonlinear programming solvers and rigorous disturbance estimation methods. Aspen’s nonlinear control algorithms are non-generic and based on dated and inefficient sub-space identification methods.

Multiple Model-based Predictive Control 5 4 - -Efficient State/Disturbance Estimation 5 2 - -

Automated tool for developing fast dynamic, nonlinear models 5 4 - -

Opt

imiz

atio

n

Derivative free optimization 5 - - 4

The CCSI Toolset includes the most advanced and flexible set of optimization tools. Aspen’s optimization solvers are dated, and modeFrontier does not natively link with process simulation software.

Process optimization 5 5 - 2Incorporates code from the world's most advanced algebraic deterministic MINLP optimization solver (BARON) 5 - - -

Support for optimization under uncertainty (OUU) 4 - - 4Support for external optimization solvers 5 - - 4Support for multi-objective optimization 4 - - 4

UQ

Simultaneous property and process parameter estimation with UQ 5 - - -Only the CCSI Toolset was created from the ground up

with uncertainty quantification as a central goal. As such, it is the only tool with integrated Bayesian statistics and uncertainty quantification that goes well beyond simple sensitivity analysis, enabling propagation of uncertainty across scales and enabling Bayesian calibration for unprecedented predictive model capability.

Support for Sensitivity analysis 5 4 4 4Support for Bayesian statistics to estimate uncertainty and/or parameters 5 - - -Propagation of uncertainty across scales from properties to device to equipment 5 - - -Hierarchical validation framework for CFD models to predict scale up performance with quantitative confidence 5 - 4 -

Wor

kflo

w

Integrated multiscale modeling framework that supports models from particle, to device, to process 4 - - -

Only the CCSI Toolset includes native support for validation using rigorous statistical methods.

The CCSI Toolset includes various tools to enable physics-based scale –bridging of multi-physics models across different length scales to allow model-based optimization and uncertainty propagation.

Workflow for developing & validating particle scale models 4 - - -Workflow for developing & validating device scale models 5 - 3 -Workflow for developing & validating process scale models 5 3 - -Ability to automatically manage 1000's of simulations in parallel on cloud-based computers 5 - - -Ability to exploit parallel computing architectures 5 - 5 3Support for interconnecting multiple software and simulation tools (Aspen, gPROMS, Thermoflow, Excel, MFIX, Fluent) 5 - - 3

Automated workflow to enable superstructure optimization using detailed models 4 - - -Automated workflow to enable detailed 2-D and 3-D models to be incorporated into systems models 4 - - -