Simulation Structure for Simulation Model of MCFC–GT Hybrid

System

Jarosław Milewski

, Jakub Skibinski

and Piotr Biczel

Institute of Heat Engineering, Warsaw University of Technology, 21/25 Nowowiejska Street, Warsaw, Poland

Institute of Electrical Power Engineering, Warsaw University of Technology, 75 Koszykowa Street, Warsaw, Poland

Keywords: Fuel Cells, Hydrogen, Control System, Hybrid System.

Abstract: The control system of MCFC coupled with a gas turbine should be based on the multi–layer structure, (two

or three–layers), wherein the third layer relates to the power output from the system and can be considered

separately. Simulation model of MCFC–GT hybrid system was built. The simulator is based on a zero-

dimensional modelling of the individual elements of the system. The simulator was used for mapping the

main components behaviour (MCFC and GT separately). Based on the obtained maps of the performances

and adopted restrictions on technical–operational nature the operation line for the first line of the control

strategy was obtained. The presented results indicate that the analysed MCFC–GT Hybrid System possesses

a high operation and control flexibility while at the same time maintaining stable thermal efficiency. Operation

of the system is possible over a wide range of parameter changes.

1 INTRODUCTION

Presently, most energy is produced by large baseload

power plants (Kawabata et al., 2012) and distributed

to customers via a grid. In the future the energy

distribution system will probably take a radically

different form—it will be composed of many small

units connected to a network called distribution

generation (DG).

DG is a system of energy distribution where the

energy is produced locally. A connection to the grid

allows energy to be bought from and sold to other

customers. Energy sources for DG will have to meet

certain requirements: appropriate range of power

output, electrical efficiency—which are higher than

presently obtained by large power plants, acceptable

costs of installation, possibility of using standard fuels.

Most of the requirements are met by fuel cell

hybrid systems (HS), which are a combination of

a fuel cell module and gas turbine system and utilized

for power generation, combined heat, and power, and

even triple–generation. Hydrogen is the most

promising energy carrier for future applications.

There are many types of fuel cells, two of them are

high temperature: Molten Carbonate Fuel Cell

https://orcid.org/0000-0003-1215-1802

(Ramandi et al., 2014; Razbani and Assadi, 2014;

Sheng et al., 2006; Wee, 2014)—MCFC—and Solid

Oxide Fuel Cell (Ding et al., 2014; Razbani and

Assadi, 2014; Xu et al., 2014) —SOFC. Solid Oxide

Fuel Cells (SOFCs) are potential sources for this

system of energy conversion due to their high

efficiency and possibility of direct use of

hydrocarbons. Moreover, their high working

temperature allows for the possibility of using lower-

cost catalysts (Ni vs. Pt), standard fuels (even

biofuels) and the possibility of adding a gas turbine

subsystem to increase total efficiency.

Control strategy is an important element in

designing any system of this kind and it constitutes

a significant part of the modeling work done. Off-

design (part-load) analysis is an important issue for

any type of system involving MCFC–HS and should

be considered when designing and defining the

operational characteristics. A proper off-design map

of performance underscores control strategy design.

Results drawn from system behavior analysis at part

load conditions should aid in defining the system

structure and its nominal parameters, as well as the

constructional solution and characteristics of a given

subsystem.

Paper published in Journal of Fuel Cell Science and

Technology 12(4) DOI:10.1115/1.4031169

202

Milewski, J., Skibinski, J. and Biczel, P.

Simulation Structure for Simulation Model of MCFCâ

A ¸SGT Hybrid System.

DOI: 10.5220/0012052100003546

In Proceedings of the 13th International Conference on Simulation and Modeling Methodologies, Technologies and Applications (SIMULTECH 2023), pages 202-209

ISBN: 978-989-758-668-2; ISSN: 2184-2841

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

Bedont et al. (Bedont et al., 2003) investigated off-

design performance of a hybrid system based on an

existing 100 kW MCFC stack. Chen et al. (Chen et

al., 2006) show that the generating efficiency of the

MCFC-HS ( ≈ 13 kW) is close to 60% at the design

point and over 56% at part load conditions. A few

dynamic response investigations of the MCFC can be

found in (He, 1998; Kang et al., 2001; Sheng et al.,

2006; Yang et al., 2019). The paper (Iora et al., 2010)

presents a model for the off-design analysis of

a hybrid plant based on a MCFC and a gas-turbine.

The model is used to define a possible regulation

strategy for the power plant, minimizing the

performance decay at partial load and allowing

investigation of the interaction issues among the

different plant components. The results indicate the

possibility to effectively regulate the plant power

output acting on the turbine shaft speed, the air-to-

fuel ratio, the bypass of cathode air, and the fuel

utilization, achieving very high part-load efficiency

and respecting constraints on the admitted operating

range for the plant components. In (J Milewski and

Miller, 2012) the results of mathematical modeling

and numerical simulations of the off-design (part-

load) operation of the molten carbonate fuel cell

hybrid system (MCFC-HS) are set out. The governing

equations of modeling are given and an adequate

simulator of the MCFC stack was made and

described. The performance of the MCFC-HS with

part- and over-load operation is shown, and adequate

maps are given and described.

Figure 1: The configuration of MCFC–GT Hybrid System.

In this control strategy study for MCFC-HS, use was

made of methodology and experience of the Institute

of Heat Engineering (Warsaw University of

Technology). Institute of Heat Engineering

methodology was utilized in the mathematical

modeling of the “classic” system elements (e.g.,

compressor, turbine, heat exchanger). HYSYS.Plant

(“HYSYS.Plant Steady State Modelling,” 1998)

software was used for modeling and simulations.

The results presented in this paper concern

a larger system (3 MW), which can be utilized for

office building applications. An axial turbine can

replace the radial turbine for this range of power.

2 MOLTEN CARBONATE FUEL

CELL–GAS TURBINE HYBRID

SYSTEM

The planar MCFC has pre-commercial applications

(100 kW to 2.8 MW) which include the methane–

fueled MCFC Module (MCFC-M) as the main object.

The fuel cell stack, heat exchangers, mixing chamber,

blower, pre-reforming plenum and re-cycle plenum

Table 1: Nominal parameters of the MCFC-GT hybrid

system (Milewski et al., 2010; Jarosław Milewski and

Miller, 2012).

Paramete

Value

Overall Efficienc

(

LHV

)

, % 58

Re-c

cle Facto

Excess Air Facto

1.68

Avera

e Cell Volta

e, V 0.66

DC/AC inverter efficienc

, % 95

Electric

enerator efficienc

, % 99

Mechanical efficienc

of the GT, % 99

Electric motor efficienc

, % 95

Turbine Inlet Temperature

(TIT), ○ C

650

Compressor pressure ratio 8.35

Fuel cell electrolyte material Li/Na

Electrolyte matrix thickness, mm 1

are placed inside the MCFC-M. The Molten

Carbonate Fuel Cell—Hybrid System consists of the

following elements:

Simulation Structure for Simulation Model of MCFCâ

A¸SGT Hybrid System

203

• Air Compressor

• Fuel Compressor

• Gas Turbine

• Air Heater

• Fuel Heater

• MCFC Module

The stack consists of parallel, and series

connected cells. The singular cell consists of three

main layers: anode, electrolyte, and cathode. The

electrolyte is kept by a matrix, which provides cell

support. Proper and efficient operation of the MCFC

module requires additional devices.

Process air is delivered to the MCFC at an

elevated temperature and flows through the heat

exchanger. Air then flows through the stack and

escapes on the other side. Finally, it enters the

combustion plenum where the remaining non-

oxidized components are utilized.

Excess power from the compressor turbine

subsystem is converted into electricity. HS efficiency

is given by the equation:

𝜂





𝑃



⋅𝜂



𝑃



𝑃

,

⋅𝜂



⋅𝜂



𝑃

,

𝑛



⋅𝐻𝐻𝑉



where: η —efficiency; g —electric generator; m —

mechanical; C —compressor; n —molar flow,

kmol/s; HHV —Higher Heating Value,

kJ/kmol; DC-AC —DC/AC inverter.

Based on 0 D mathematical modeling the design

point parameters of the system presented in

Fig. 1 were estimated; they were obtained by the

researchers’ own calculations based on an adequate

mathematical model (Milewski et al., 2013). During

the simulation the electrolyte matrix thickness and

electrolyte materials assumed were 1 mm and Li/Na,

respectively. The main system parameters are

presented in Table 1.

3 THE CONTROL STRATEGY

Mathematical modeling is now the basic method for

analyzing systems incorporating fuel cells. A zero-

dimensional approach is used for the modeling of

system elements. Mathematical models of MCFC,

and other system elements as well as the control

strategy for MCFC–GT based on triple–layer control

system is presented in our previous works (Milewski

et al., 2010; Jarosław Milewski and Miller, 2012). As

the results of the previous works adequate maps of

performances were obtained with indicated the

control line of the system. In general, it should be

underlined that in the case of a system with

a pressurized MCFC and gas turbine set there is

a possibility of changing the system power output by

changing not only the amount of fuel but also the

voltage and the MCFC current at variable rotational

speeds of the compressor-turbine unit. This is

accompanied by varying system efficiencies. Hence

there is a need to formulate an appropriate control

concept (control strategy logic) and approach for

technical realization.

Figure 2: Triple-layer control system.

A triple layer control system is proposed for

MCFC-HS operation control. The system consists of

three layers: Control Strategy, Adaptation, and

Regulation (see Fig. 2).

The first layer is responsible for safe and efficient

operation of the whole hybrid system. Adequate

functional relationships between all controlled

parameters and constraints should be applied. The

adaptation layer is responsible for making corrections

to first layer characteristics due to the degradation of

system elements. The last layer of the control system

acts in dynamic mode to realize the control strategy.

The control strategy is based on three functional

relationships:

𝑚



𝑓𝑃





𝑛𝑓𝑃





𝐼



𝑓𝑃





where: P

—is power demanded by an external load.

Approximately, 16,000 system operation points

(state points) were found. Every state point is defined

by three independent parameters: delivered fuel flow,

rotational speed of compressor-turbine subsystem,

and stack current. The other flow and electric

parameters were collected for these three parameters.

This data set was analyzed, and it was found that the

best system performances are obtained at the highest

values of fuel utilization factor. The highest possible

fuel utilization factor was found to be 90%. This

means that fuel flow can be correlated with used

SIMULTECH 2023 - 13th International Conference on Simulation and Modeling Methodologies, Technologies and Applications

204

electric current from the stack by the following

relationships:

𝐼

𝑚



𝑐𝑜𝑛𝑠𝑡







,

𝑎⋅𝑃



𝑏 (1)







,

𝑎⋅𝑃



𝑏 (2)

where: a and b —linear regression factors. The linear

regression factors a and b are: 2.1 ⋅ 10

-4

and 8.69 ⋅ 10

The maps presented below include the values of

parameters for constant fuel utilization factor of 90%.

Adequate maps of performance were generated by

a MCFC–GT Hybrid System simulator using 0 D

mathematical modeling. The operation line of

MCFC–GT Hybrid System was determined based on

in-depth analysis of the maps obtained. The best

performance and safe operation of the system can be

achieved with constant fuel utilization factor (at the

highest possible value). Setting the constant fuel

utilization factor at 90% appears acceptable and

adequate relationships are proposed (1 and 2). Thus,

the obtained relationships are used in the first layer of

the triple layer control system.

The second layer is needed for adaptation of the

control strategy to changes in external environments

(fuel quality, single element degradations, etc.). The

long operation of the MCFC–GT Hybrid System is

not the subject of the paper; thus, this layer remains

empty.

4 THE CONTROL SYSTEM

The third layer of the control system is responsible for

dynamic operation of the system and requires to be

built as the control system of the unit. The control

strategy can be realized based on various

architectures, in this paper we propose to use internal

DC micro network to balance the power between

MCFC stack and gas turbine subsystem.

Figure 3: The general structure of DC/DC converter.

The control system is based on DC/DC

converters, which can be classified in various ways,

but generally they possess the similar structure as

shown in Fig. 3. Two major subsystems can be

distinguished here the power circuit and control

circuit. The heart of the control circuit regulators is

selected signal (voltage, current or power, input or

output, or any combination of these), depending on

the application, and the modulator. Power circuit

consists of an input filter, the encoder system

transformer and rectifier (in some types of converters,

for example in the breastbone) and output filter.

Figure 4: Block scheme of single unit controller.

Figure 5: Texas Instruments TMS320F2812 controller.

Currently control systems for complex circuitry,

such as hybrid systems that use different sources of

energy, the building blocks of the special signal

processors DSP, commercially available controller

(Texas Instruments TMS320F2812—see Fig. 5) was

used. The main tasks of the used controller in the

hybrid system control are:

• measurements of currents and voltages of each

local devices

Simulation Structure for Simulation Model of MCFCâ

A¸SGT Hybrid System

205

• if necessary, modifying the control signal to

ensure constant current–voltage conditions at

the output resulting from the setpoints coming

from the main controller.

• transmission of information on the operation

of the inverter by using external interfaces

• possibility to change the setting operation of

the inverter via the external interface.

Figure 6: Block scheme of the algorithm implemented in

the single unit controller.

The designed control system was achieved on the

processor TMS320F281 calling the timer interrupt

every 50 microseconds (see Fig. 6). The

discontinuation of this measurement is made,

integration and correction of the width of the

rectangular waveform. This is sufficient because the

frequent measurement gives the ability to quickly

respond to changing output parameters. PWM signal

generation takes place outside the main program. It is

independent of the computational process. Thus,

when conducting the calculation of new settings

waveforms are still generated from a predetermined

frequency—typically 10 .. 30 kHz.

A good idea is also to use multiple interfaces, e.g.

ISP for communication and RJ-45 interface to make

simple changes or reading an information from

master (external) controller, as implemented in

MCFC–GT hybrid system—Fig. 7. The used

microcontroller (TMS320F2812) cannot be

connected in a simple way to Ethernet interface, and

since it is necessary to realize the communication

between CPU to the local actuators, it was decided to

apply additional microcontroller for a web server

function. Fig. 8 presents the view motherboard with

integrated microcontroller ATMEGA 128 carrying

out the functions of the web server.

Figure 7: The structure of the communication between the

MCFC module and gas turbine sub-system by the control

system.

Figure 8: A view of motherboard which integrating

a microcontroller ATMEGA 128 16 MHz with Ethernet

controller RTL1819AS IEEE 802.3 10Mb/s for carrying out

the function of a web server.

The main assumptions designed control system,

using several communication interfaces, standards,

SIMULTECH 2023 - 13th International Conference on Simulation and Modeling Methodologies, Technologies and Applications

206

and protocols (Ethernet, SPI, ModBus) for

communication between components of MCFC-GT

hybrid system are as follows:

• Power electronics inverter control for

simulating MCFC behavior based on signals

from the CPU based on two cooperating

microcontrollers ATMEGA 128 and

TMS320F2812, one of whom one acts as

server and the other is responsible for the

stable operation of the inverter.

• Classic serial interface was used for the

communication between TMS320F2812

and a microprocessor (server). Interface is

relatively simple to use, the amount of

information sent by it will not be large, and

the distance between the two processors will

be the order of centimeters, so there is no

significant risk of interference between the

control signals.

• Gas Turbine subsystem controller is

a microcontroller Danfoss drive, see Fig. 9

• Drivers of power electronic converters

enable parallel operation of both

components (MCFC and gas turbine set) are

based on two cooperating microcontrollers,

one of which acts as a server and the other is

responsible for the stable operation of the

inverter.

• Communication between the central control

unit and the controller of the MCFC and the

driver circuits enable parallel operation is

realized by Ethernet.

• Communication between the central control

unit and the Gas Turbine set controller (drive

inverter) is implemented using the ModBus

protocol.

Figure 9: Danfoss drive inverter.

The control system was implemented into software–

hardware simulator of the MCFC–GT hybrid system

which is composed by the following elements:

• two power electronics inverters AC/DC with

half-bridge topology with power 2 kW each

• one power electronics inverter DC/DC with

half-bridge topology with a power of 2 kW

• hardware model of Gas Turbine set

consisting of a three-phase synchronous

motor and three-phase synchronous

generator mechanically connected.

Figure 10: Ga Turbine set hardware–based model with an

inverter drive by Danfoss.

For mapping the MCFC behavior the two power

electronic converters are used, whereby one (AC/DC

converter) simulates the fuel cell and the second

(DC/DC) is responsible for cooperation with the

turbomachinery. The simulating of the air

compressor–gas turbine–generator required the

statement of both machines in the system gas turbine

Figure 11: Scheme of the electric system of the simulator.

Figure 12: The simulator with connected measurements and

data acquisition instrumentation.

set (engine shaft connected to the shaft of the

generator by using a coupling). For powering a model

Simulation Structure for Simulation Model of MCFCâ

A¸SGT Hybrid System

207

of gas turbine set, the Danfoss drive inverter with

a power of 2.2 kW is used—see Fig. Turbine with an

inverter drive the company Danfoss.

The output synchronous generator connected with

the AC/DC inverter is responsible for cooperation

with MCFC. Schematic diagram of the electrical part

of the simulator is shown in Fig. 11. Initially it was

assumed that the system would operate with an

external electric power system, but ultimately limited

to a DC power receiver. Fig. 12 shows the view of the

built simulator with measurements and data

acquisition instrumentation.

Figure 13: The measured power generated by the system

and the corresponding equipment obtained using the control

system.

The control system and hardware–software

simulator was developed from a series-manufactured

components. The simulator was controlled by the

control system with implemented the control strategy.

Fig. 13 presents the power as a function of time

recorded during the tests under load by the assembled

system.

Figure 14: MCFC and Gas Turbine set powers as functions

of total system power (MCFC–GT).

Fig. 14 presents the dependence of the particular

components that make up the MCFC-GT hybrid

system as a function of output power (compare

against Eqs (1) and (2)). The curves in

Figs 13 and 14 confirm the correctness and validity

of the control strategy carried out, simulation and

constructed mathematical models.

5 CONCLUSIONS

The control system of MCFC coupled with a gas

turbine should be based on the multi–layer structure,

(two or three-layers), wherein the third layer relates

to the power output from the system and can be

considered separately.

Simulation model of MCFC–GT hybrid system

was built. The simulator is based on a zero-

dimensional modeling of the individual elements of

the system. The simulator was used for mapping the

main components behavior (MCFC and GT

separately).

Based on the obtained maps of the performances

and adopted restrictions on technical–operational

nature the operation line for the first line of the control

strategy was obtained. The highest system efficiency

is achieved by working at the highest possible fuel

utilization in the MCFC stack. It was assumed that the

limit value of this parameter is 90%. Adequate

relationships are proposed for achieving this

assumption (see Eqs (1) and (2))—the fuel amount

delivered to the system is in proportion to the amount

of electric current drawn from the MCFC stack.

Maintaining the safe and efficient operation of the

system is realized by gas turbine set in this case, the

gas turbine subsystem, what is obtained by

appropriate changes in the rotational speed of the

shaft.

The control system which realizes the obtained

control strategy was built. Then, hardware-based

models of the main elements were created based on

the electric equipment. The gas turbine was simulated

by coupled electric engine and electric motor

simulated by programmed micro-controller for giving

current–voltage characteristics. The hardware–

software model was connected to the control system

and adequate simulations were performed.

The presented results indicate that the analyzed

MCFC–GT Hybrid System possesses a high

operation and control flexibility while at the same

time maintaining stable thermal efficiency. Operation

of the system is possible over a wide range of

parameter changes.

The presented control strategy offers some

practical advantages, which include maintaining the

operating point at a possibly high level of efficiency

even under changed operating conditions.

Furthermore, it is possible to separate the overriding

control layer (layer III) from the global strategy of

such a system.

SIMULTECH 2023 - 13th International Conference on Simulation and Modeling Methodologies, Technologies and Applications

208

ACKNOWLEDGEMENTS

This paper was supported by the program Lider grant

nr 0005/L-11/2019 by the National Centre for

Research and Development, Poland.

REFERENCES

Bedont, P., Grillo, O., Massardo, A.F., 2003. Off-design

performance analysis of a hybrid system based on an

existing molten fuel cell stack. J. Eng. Gas Turbines

Power 125.

Chen, Q., Weng, Y., Zhu, X., Weng, S., 2006. Design and

Partial Load Performance of a Hybrid System Based on

a Molten Carbonate Fuel Cell and a Gas Turbine. Fuel

Cells 6.

Ding, J., Li, X., Cao, J., Sheng, L., Yin, L., Xu, X., 2014.

New sensor for gases dissolved in transformer oil based

on solid oxide fuel cell. Sensors Actuators, B Chem.

202, 232–239.

He, W., 1998. Dynamic Model for Molten Carbonate Fuel-

Cell Power-Generation Systems. Energy Convert.

Manag. 39, 775–783.

HYSYS.Plant Steady State Modelling, 1998.

Iora, P., Campanari, S., Salogni, A., 2010. Off-design

analysis of a MCFC-gas turbine hybrid plant.

Kang, B.S., Koh, J.-H., Lim, H.C., 2001. Experimental

study on the dynamic characteristics of {kW}-scale

molten carbonate fuel cell systems. J. Power Sources

94, 51–62.

Kawabata, M., Kurata, O., Iki, N., Tsutsumi, A., Furutani,

H., 2012. Advanced integrated gasification combined

cycle {(A-IGCC)} by exergy recuperation---Technical

challenges for future generations. J. Power Technol. 2,

90–100.

Milewski, J, Miller, A., 2012. Off-design analysis of MCFC

hybrid system. Rynek Energii 151–160.

Milewski, Jarosław, Miller, A., 2012. Triple-layer based

control strategy for molten carbonate fuel cell--hybrid

system. Chem. Process Eng. 445–461.

Milewski, J., Świercz, T., Badyda, K., Miller, A.,

Dmowski, A., Biczel, P., 2010. The control strategy for

a molten carbonate fuel cell hybrid system. Int. J.

Hydrogen Energy 35, 2997–3000.

Milewski, J., Wołowicz, M., Miller, A., Bernat, R.,

RafałBernat, 2013. A reduced order model of molten

carbonate fuel cell: A proposal. Int. J. Hydrogen Energy

38, 11565–11575. https://doi.org/10.1016/j.ijhydene.

2013.06.002

Ramandi, M.Y., Dincer, I., Berg, P., 2014. A transient

analysis of three-dimensional heat and mass transfer in

a molten carbonate fuel cell at start-up. Int. J. Hydrogen

Energy 39, 8034–8047.

Razbani, O., Assadi, M., 2014. Artificial neural network

model of a short stack solid oxide fuel cell based on

experimental data. J. Power Sources 246, 581–586.

https://doi.org/10.1016/j.jpowsour.2013.08.018

Sheng, M., Mangold, M., Kienle, A., 2006. A strategy for

the spatial temperature control of a molten carbonate

fuel cell system. J. Power Sources 162, 1213–1219.

Wee, J.-H., 2014. Carbon dioxide emission reduction using

molten carbonate fuel cell systems. Renew. Sustain.

Energy Rev. 32, 178–191.

Xu, H., Dang, Z., Bai, B.-F., 2014. Electrochemical

performance study of solid oxide fuel cell using lattice

Boltzmann method. Energy 67, 575–583.

Yang, C., Deng, K., He, H., Wu, H., Yao, K., Fan, Y., 2019.

Real-Time Interface Model Investigation for MCFC-

MGT HILS Hybrid Power System. ENERGIES 12.

https://doi.org/10.3390/en12112192.

Simulation Structure for Simulation Model of MCFCâ

A¸SGT Hybrid System

209