Simulation and Design of Green Hydrogen Production

Systems

Ramanuja Jagannathan
Application Engineer – Control Design
MathWorks, Bangalore

© 2024 The MathWorks, Inc.

1
Agenda

▪ Green hydrogen production through electrolysis

– energy conversion: wind to H2
– multi-domain system simulation
– suitable fidelity for development models

▪ Techno-economic analysis based on physical models

– user case: photovoltaic to H2
– enabling 8760 simulations (24h x 365days)

▪ Empowering collaborative R&D

– model deployment (FMU)
– MATLAB applications

▪ Summary & Follow-up opportunities

2
Green hydrogen production - Electrolysis

[1] - Water Electrolysis - an overview | ScienceDirect Topics

[2] - Electrolysis of Water. (2021, May 31). Retrieved from https://chem.libretexts.org/@go/page/53976
[3] - Hydrogen Production: Electrolysis | Department of Energy

3
Green hydrogen production – Motivation / Feasibility

Advantages Deltas

▪ 100% sustainable ▪ high energy consumption

▪ storable ▪ safety (managing H2)
▪ versatile ▪ high cost
▪ transportable

“If this electricity (for electrolysis) is obtained Simulation

from renewable sources we will, therefore, based R&D
produce energy without emitting carbon dioxide
into the atmosphere”

Iberdrola > Sustainability > Green Hydrogen

4
Green hydrogen production – Microgrid / Solar

Multi-domain simulation

5
Green hydrogen production – Microgrid / Wind

Multi-domain simulation

6
Green hydrogen production – Import field data with MATLAB

Wind speed (mph)

Inverness, United Kingdom Weather History | Weather Underground (wunderground.com)

7
Green hydrogen production – Fidelity in simulation models

milliseconds (ms) seconds months

microseconds (us) minutes years

Embedded Key performance

assessment Techno-economic
development analyses
(component) (system)

High Medium Low

fidelity fidelity fidelity

8
Green Hydrogen production – Electrolysis unit (component)
detailed representation
High
capturing thermodynamic behaviour
fidelity in the electrolysis process
(Released in 2022a)

openExample('simscape/PEMElectrolysisSystemExample') 9
Green hydrogen production – Electrolysis unit (system)
system-level representation
Medium enabling agile simulations
fidelity and multi-domain integration
(Release 2022a)

openExample('simscapeelectrical/GreenHydrogenMicrogridExample') 10
 Medium
fidelity

https://in.mathworks.com/matlabcentral/fileexchange/53428-green-hydrogen-wind-solar-from-alkaline-electrolysis 11
Green hydrogen production – Energy conversion (wind-to-gas)

Power electronics are key in converting mechanical energy into H2

– Generator (mechanic > electric)

▪ DC or AC type
▪ Induction vs Synchronous

– DC-DC converter (electric DC > DC)

▪ Electrolyzer
▪ Battery

Power
– AC-DC converter (electric AC < > DC) Electronics
▪ Three-phase AC generator
▪ Energy storage e.g. flywheel

12
Embedded development
Component analysis

High
fidelity

13
Green hydrogen production – Challenges at physical unit level

Component design: Digital twin development:

– electrolyzer – degradation & anomaly detection
▪ thermodynamic (gas-fluid) ▪ synthesis of data
▪ safety e.g. pump and valve control ▪ feature identification
– energy storage – prognostics development
▪ Battery Management System (BMS) ▪ predictive maintenance
▪ battery characteristics ▪ remaining useful life (RUL) predictions
– power converter unit
▪ fault management
▪ converter control (grid connection)
– generator
▪ electromagnetic design
▪ Finite Element description

14
Green hydrogen production – Energy conversion (electrical machine)

High
fidelity

High (+)
fidelity

15
Green hydrogen production – Energy conversion (DC-DC converter)

A detailed converter model is built

with discrete semiconductors.

High
fidelity

N-channel MOSFET
(thermal option available)

16
Green hydrogen production – Energy conversion (AC-DC converter)

High
fidelity

17
Green hydrogen production – AC wind generation

Various fidelity levels for

Medium different components High
fidelity may co-exist. fidelity

Fidelity is often added

gradually per component.

https://in.mathworks.com/matlabcentral/fileexchange/53428-green-hydrogen-wind-solar-from-alkaline-electrolysis 18
Green hydrogen production – Stand-alone electrolyzer (high fidelity)

19
Green hydrogen production – Stand-alone electrolyzer

PM Generator
Control

20
Green hydrogen production – From concept to code

Automatic Code Generation

21
Green hydrogen production – Grid-connected electrolyzer (high fidelity)

22
Green hydrogen production – Grid-connected Electrolyzer

Control Design
& Grid Connection
(Phase Locked Loop) Detected grid
frequency

23
Green hydrogen production – Stand-alone (AC) electrolyzer

FFT
analysis
(currents)

Energy optimization
Prognostics development
24
 Key performance
system assessment
Medium
fidelity

25
Green hydrogen production – Challenges at System Level

Plant design: Algorithmic design:

– what is the best concept? – how can I develop supervisory logic?

▪ AC vs DC generation ▪ energy storage contribution
▪ grid-connected vs remote ▪ asset management
▪ wind vs solar (or both) ▪ energy source selection

– how do I meet requirements? – how do I define relevant setpoints?

▪ component sizing & integration ▪ active unit(s)
▪ modularity & scalability ▪ meteorological conditions
▪ history
– do I fully understand energy balances?
▪ multi-domain considerations
▪ electricity and water consumption
▪ H2 production estimation

27
Green hydrogen production – Energy conversion (Electrical machine)

Focus on energy flow rather than machine control.

Medium
fidelity

Simscape Language

28
Green hydrogen production – Energy conversion (DC-DC converter)

Converter transfers energy

based on the operating mode (buck, boost).
Losses are accounted for.

Medium
fidelity

29
Green hydrogen production – System Assessment

31
Green hydrogen production – System Assessment

Selected results obtained from the simulated case (1day)

32
Green hydrogen production – System Assessment Medium
fidelity
o expected H2 production & water consumption
o suitable control algorithm (conditions, use of physical assets)
o energy storage (dimensioning, expected duty regime)
o planning of operations (collect – replace - maintain)

33
Techno Economic analysis

Low
fidelity

34
Solar Powered DC Microgrid

▪ System analysis of a localized

DC network with loads and
energy generation units

▪ The DC microgrid consists of

– Solar Array
– Energy Storage
– Electrolyzer

35
36
System Integration from Verified Components

▪ Identify standalone components

▪ Choose the right fidelity and

incrementally build the models

▪ Use simulation to understand design

of components

▪ Test components independently to

verify requirements

37
 Battery modeling
Start with equivalent circuit model and parametrize using measurement data
Cell Physical Model: Parameter Estimation
𝑅1
𝑅0

𝐸𝑚 𝐶1
Thevenin Model

Lithium Cell Characteristic :

𝐸𝑚 SOC
1
SOC
0.9
SOC
0.8
…
SOC
0

5°C 4.20 V 4.12 V 4.05 V … 3.09 V

20°C 4.18 V 4.09 V 4.01 V … 3.05 V

40°C 4.15 V 4.02 V 3.97 V … 3.01 V

https://www.mathworks.com/videos/modeling-batteries-using-simulink-and-simscape-1562930245321.html 38
Testing the Solar Array

39
Solar Arrays and MPPT

▪ Drive the solar panel to operate at

the maximum power point

▪ Continuous maximum power point PMPP

tracking (MPPT) needed to deal with
variations in ambient conditions

▪ MPPT operates on the principle of

impedance matching.

40
Impedance Matching

▪ Maximum power is delivered to a load when the load impedance matches

the source impedance.

I ZS

If ZL = ZS then PL = max P

VS ZL If ZL ≠ ZS then PL < max P

Source Load

41
Impedance matching

42
Evaluating System Operation

43
Evaluating System Operation

44
Testing the Electrolyzer

Connected to an ideal DC voltage source at 240V.

45
What does techno-economic mean?
From Wikipedia

Techno-economic assessment or techno-economic analysis (abbreviated

TEA) is a method of analyzing the economic performance of an industrial
process, product, or service. It typically uses software modeling to estimate
capital cost, operating cost, and revenue based on technical and financial
input parameters.

At present, TEA is most commonly used to analyze technologies in the

chemical, bioprocess, petroleum, energy, and similar industries

46
Considerations for Techno-Economic Analysis

▪ Conducted across long durations of time at regular intervals ranging from 1 minute to 1 hour

▪ Relies on steady-state operational conditions at each time-interval instead of system dynamics

▪ Uses a quasi-steady state simulation

47
Quasi-Steady State Simulation using Reduced Order Modeling

▪ Use Look-Up Tables to simulate time based

quasi-steady state behavior

▪ For Example: solar panel use source voltage

and source impedance for a given power level.

▪ for a solar array that is operating under

maximum power point tracking (MPPT)

48
Reduced Order Modeling

49
Microgrid Example

▪ Objective: Analyze economical aspect of setting up the system

– Highest and lowest grid cost
– Highest and lowest solar resources

50
51
Irradiance Data

▪ The irradiance data is 8760 TMY3 data available from NREL. There are 242
locations in the data set.

52
Electricity Price Data

▪ Electricity price data is one day of data from ISO-NE and simply repeated
365 times

53
Run scenarios using parsim

>> parsimScenarios

242 years in 500 seconds

i.e.
1 year every 2 seconds

54
Empowering Collaborative R & D

55
Collaborative R&D - MathWorks’ provides the broadest range of
deployment options
Simulink
Compiler

Simulink

MATLAB
Coder MATLAB
Compiler
Products Compiler
MATLAB SDK

Embedded Hardware Enterprise Systems

Excel Hadoop/ Java MATLAB

Add-in
C/C++ ++ Python .NET Production
C, C++ HDL PLC CUDA .exe Spark DLL JAR py DLL Server
Deployed
MATLAB
Web App Archive

MATLAB
MATLAB Web Production
App Server Server
56
Collaborative R&D - Scenario 1: Standalone Desktop App

Standalone Apps

▪ Runs on PC

▪ Can use App Designer GUI

▪ Needs local installation

57
Collaborative R&D - Scenario 2: (MATLAB) Web App

Web Apps

Cropped picture
suggesting
integration into
custom app

▪ Runs on MATLAB Web App

Server

▪ Uses App Designer GUI

▪ Browser-based access, no
local installation needed

58
Collaborative R&D - Scenario 3: Service API

Service APIs
Cropped picture
suggesting
scaling
production
(JS/HTML-based
app?)

▪ Runs on MATLAB
Production Server*

▪ Supports customer
developed client-server App
and web apps (e.g.
HTML/JavaScript)

▪ Centrally hosted, no local

installation needed http://insidelabs-mpsdemos/simulink/massspring.html 59
Collaborative R&D - Scenario 4: Standalone FMU

Standalone FMUs

▪ Runs in 3rd party simulation

tools (which support FMI
import)

▪ V1 has no dependency on
MATLAB Runtime

▪ V1 supports standalone co-

simulation FMU with fixed-
step solver only 60
Conclusions & Follow-up Opportunities

61
Conclusions – Green hydrogen
Embedded Converter control
(Digital Twin) Generator control
High
fidelity Development Prognosis

Medium
fidelity System Concept evaluation
Performance Energy regulation
Assessment Component sizing

Multi-domain physical
modeling & simulation Low
Long term predictions
fidelity Techno-
platform Commercial feasibility
economic
analyses (RoI calculation)

62
Conclusions – Green hydrogen
Embedded Converter control
(Digital Twin) Generator control
Development Prognosis
Collaborative
Research & Development

Deployment System Concept evaluation

(web applications, Performance Energy regulation
FMUs, Assessment Component sizing
stand-alone apps)

Techno- Long term predictions

economic Commercial feasibility
analyses (RoI calculation)

Strong adoption: https://www.mathworks.com/services/training.html

Successful implementation: https://www.mathworks.com/services/consulting.html
63
 https://in.mathworks.com/learn/training.html
Enable Your Team For Electrification

Motor Control with

Simulink and Simscape

64
Enable Your Team For Electrification

https://matlabacademy.mathworks.com/ 65
Additional resources at www.mathworks.com
▪ Solutions
– Battery Systems - MATLAB & Simulink
– Power Conversion Control - MATLAB & Simulink
– Renewable Energy and Energy Storage - MATLAB & Simulink
– Microgrid, Smart Grid, and Charging Infrastructure - MATLAB & Simulink
– Generation, Transmission, and Distribution - MATLAB & Simulink
– Simulink for Fuel Cells and Electrolyzers - MATLAB & Simulink

▪ Community Page
– Power Electronics Control Community - MATLAB & Simulink

▪ User Testimonies
– Fuel Cells
– Batteries
– Power Conversion
– Renewable Energy & Storage
– Microgrid, Smart Grid & Charging Infrastructure
– Generation, Transmission & Disribution

66
Thank you

▪ Please send your questions to

Ramanuja Jagannathan at

▪ E-mail:
[email protected]

▪ LinkedIn:
www.linkedin.com/in/jrvinayak

67