Modeling and Simulation of Cooling System for Fuel Cell Vehicle

(1)

UPTEC ES 17032

Examensarbete 30 hp

Juni 2017

Modeling and Simulation of Cooling

System for Fuel Cell Vehicle

(2)

Teknisk- naturvetenskaplig fakultet UTH-enheten Besöksadress: Ångströmlaboratoriet Lägerhyddsvägen 1 Hus 4, Plan 0 Postadress: Box 536 751 21 Uppsala Telefon: 018 – 471 30 03 Telefax: 018 – 471 30 00 Hemsida: http://www.teknat.uu.se/student

Abstract

Modeling and Simulation of Cooling System

for Fuel Cell Vehicle

Samuel Swedenborg

This report is the result of a master’s thesis project which covers the cooling system in Volvo Cars’ fuel cell test vehicle. The purpose is to investigate if the existing cooling system in the fuel cell test vehicle works with the current fuel cell system of the vehicle, in terms of sufficient heat rejection and thus sustaining acceptable temperature levels for the fuel cell system. The project also aims to investigate if it is possible to implement a more powerful fuel cell system in the vehicle and keep the existing cooling system, with only a few necessary modifications. If improvements in the cooling system are needed, the goal is to suggest improvements on how a suitable cooling system can be accomplished. This was carried out by modeling the cooling system in the simulation software GT-Suite. Then both steady state and transient simulations were performed. It was found that the cooling system is capable of providing sufficient heat rejection for the current fuel cell system, even at demanding driving conditions up to ambient temperatures of at least 45°C. Further, for the more powerful fuel cell system the cooling system can only sustain sufficient heat rejection for less demanding driving conditions, hence it was concluded that improvements were needed. The following improvements are suggested: Increase air mass flow rate through the radiator, increase pump performance and remove the heat exchanger in the cooling system. If these improvements were combined it was found that the cooling system could sustain sufficient heat rejection, for the more powerful fuel cell system, up to the ambient temperature of 32°C.

ISSN: 1650-8300, UPTEC ES 17032 Examinator: Petra Jönsson

(3)

Populärvetenskaplig sammanfattning

Dagens samhälle är beroende av transport. Vidare är dagens transportsektor beroende av oljebaserade drivmedel. Dessa drivmedel anses varken vara hållbara ur ett miljömässigt eller ekonomiskt perspektiv, dessutom är olja en ändlig resurs. Att hitta hållbara transportlösningar är en nyckelfråga för dagens regeringar, företag och privatpersoner. Vad är då lösningen på detta problem? En teknik som får mer och mer uppmärksamhet i fordonsindustrin är bränsleceller. Dessa använder vanligtvis vätgas för att producera el och då är det enda utsläppet vatten. Den producerade elen används för att driva fordonet framåt via en elmotor. Dessutom har bränsleceller högre verkningsgrad än konventionella förbränningsmotorer som används i fordon idag. Emellertid är inte verkningsgraden 100 %, vilket innebär att en hel del värme bildas. För att bränslecellerna inte ska överhettas behöver värmen kylas bort på något sätt. Det är där som den här rapporten och det här projektet kommer in. Projektet handlar om kylsystem i bränslecellsfordon. Mer specifikt behandlas kylsystemet i Volvo Cars bränslecellstestfordon. Syftet med projektet är att undersöka om det befintliga kylsystemet i bränslecellstestfordonet fungerar med det nuvarande bränslecellssystemet. Vidare syftar projektet även till att undersöka om ett kraftfullare bränslecellssystem kan implementeras och samtidigt behålla det befintliga kylsystemet, med endast några nödvändiga förändringar. Om kylsystemet behöver förbättras är målet med projektet att ge förslag på förbättringar som kan bidra till att kylsystemet passar för de respektive bränslecellssystemen. För att ta reda på detta analyserades om kylsystemet kunde avge tillräckligt mycket värme för att behålla temperaturen för bränslecellssystemet inom acceptabla nivåer.

Projektet genomfördes genom att modellera kylsystemet i simuleringsprogrammet GT-Suite. För att göra detta samlades data, specifikationer och information in för de olika kylsystemskomponenterna. Därefter utfördes så kallade stationärt tillstånd (eng. steady state) och transienta simuleringar för kylsystemet. Stationärt tillstånd simuleringarna gav inblick i hur kylsystemet fungerar med avseende på verkningsgrad och värmeavgivning för bränslecellsystemet. De transienta simuleringarna representerade olika körcykler för att se hur kylsystemet hanterar dessa. Dessutom kördes de transienta simuleringarna vid olika omgivningstemperaturer vilket är intressant för att förstå kylsystemets prestanda.

(4)

bränslecellssystemet för alla föreslagna förbättringar och därmed kyls mer värme bort jämfört med det ursprungliga kylsystemet. När förbättringsförslagen kombinerades kunde kylsystemet kyla bort tillräckligt värme, upp till en omgivningstemperatur om 32°C.

Slutsatsen är att kylsystemet i bränslecellstestfordonet kan kyla bort värmen som uppstår vid körning av ett mindre kraftfullt bränslecellssystem. Emellertid vid kraftfullare bränslecellssystem behöver kylsystemet förbättras och förslagen i det här projektet kan leda till det. Hur dessa förbättringsförslag ska implementeras på bästa sätt får framtida arbete utvisa. Det finns även andra aspekter att undersöka vidare som tas upp i detta projekt. Till exempel: Hur fungerar bränslecellssystemet och kylsystemet vid kallstart? Hur kan optimal drifttemperatur uppnås så snabbt som möjligt?

(5)

Executive summary

(6)

Acknowledgments

I would like to acknowledge the support of the people who have helped me to complete this project. First of all, I would like to recognize the support of my supervisor, Signy Tryggvadottir, at the Outer Cooling System group of Volvo Cars. Thank you for your many hours of helping me with finding the errors of the simulation models and for your input in this project. Secondly, thanks to my subject reviewer, Cecilia Boström at the Department of Engineering Sciences of Uppsala University, for your support through the whole project. Also, a big thank you to the personnel at PowerCell, especially Felix Haberl who has answered my many questions. Finally, I want to express my sincerest gratitude to all Volvo Cars personnel who have helped me. A special thanks to the CAE engineers of the Outer Cooling System group who have supported me during the whole project.

(7)

Nomenclature

𝑐 Specific heat

𝑐𝐴𝐶 Specific heat of the coolant at the air compressor 𝑐𝐹𝐶 Specific heat of the coolant at the fuel cell stack 𝑒− _Electron

𝐻+ _Proton

𝐻2 Hydrogen gas 𝐻₂𝑂 Water

𝑚̇ Mass flow rate of the medium

𝑚̇𝐴𝐶 Mass flow rate of the coolant at the air compressor 𝑚̇𝐹𝐶 Mass flow rate of the coolant at the fuel cell stack 𝑂₂ Oxygen gas

𝑃_𝐴𝐶 Power consumption of AC 𝑃𝑏𝑎𝑡𝑡𝑒𝑟𝑦 Battery output electric power

𝑃_𝐹𝐶 Fuel cell stack output electric power 𝑃_𝑖𝑛 Power supplied to the fuel cell stack 𝑃𝑛𝑒𝑡 Fuel cell system net power

𝑄̇ Heat input rate to the medium

𝑄_𝐴𝐶 Heat generation rate from air compressor 𝑄𝐹𝐶 Heat generation rate from fuel cell stack

𝑇_{𝐴𝐶,𝑖𝑛} Inlet coolant temperature of the air compressor 𝑇𝐴𝐶,𝑜𝑢𝑡 Outlet coolant temperature of the air compressor 𝑇_𝑎𝑚𝑏 Ambient temperature

𝑇𝐹𝐶,𝑖𝑛 Inlet coolant temperature of the fuel cell stack 𝑇_{𝐹𝐶,𝑜𝑢𝑡} Outlet coolant temperature of the fuel cell stack

𝑇_{𝐹𝐶,𝑚𝑒𝑎𝑛} Mean temperature of the inlet and outlet temperature of the coolant ∆𝑇 Temperature difference of the medium

∆𝑇𝐴𝐶 Temperature difference of the inlet and outlet coolant of the air compressor ∆𝑇_𝐹𝐶 Temperature difference of the inlet and outlet coolant of the fuel cell stack 𝑡_{𝑝𝑢𝑚𝑝 𝑑𝑒𝑙𝑎𝑦} Time of start-up delay for Inner Pump

(10)

Abbreviations

AC Air Compressor BC Berlin Cycle DC Direct Current EM Electric Machine/Motor FC Fuel Cell HX Heat Exchanger

HVAC Heat Ventilation and Air Conditioning HVCH High Voltage Coolant Heater

ID Inner Diameter

PEMFC Proton Exchange Membrane Fuel Cell PID Proportional-Integral-Derivative SOC State Of Charge

VP Vehicle Performance

(11)

1

1 Introduction

The search for alternative energy sources, replacing the conventional fossil fuels, is today a demand due to the growing concern about the environmental impacts and the depletion of fossil fuels [1]. This search is subjected to both governments and industries [2]. For the automobile industry the fossil fuel dependence is significant [3] and therefore this industry are pushing the limits of existing technology, introducing new improvements and investing in research and development of new technologies [4]. Hydrogen gas (henceforth called simply hydrogen) fuel cells have received more attention during the past decade and could have potential to become the power source of the future, due to their reduced emissions and high efficiency characteristics [3], [4]. However, the lack of a hydrogen infrastructure, cost and durability of the fuel cell (FC) stack are considered the biggest issues to the introduction of FC vehicles [3]. According to [5] the proton exchange membrane fuel cell (PEMFC) is considered to be the most promising candidate for the next generation of power source for transportation applications, due to its high power density, rapid startup and low operating temperature. Even though PEMFCs have a relatively high energy conversion efficiency, there is still a significant amount of heat generated when the hydrogen is converted into electric power. To avoid overheating of the FC, especially its membrane, the generated heat must be effectively removed. The cooling system of FCs have exclusively this purpose, to remove heat from the FC. This report will focus on cooling systems for FC vehicle applications. More specifically the report will immerse in a specific cooling system for a FC car at Volvo Cars and whether it is possible to use conventional cooling system technology and structure in order to fulfill the cooling demands for FC systems.

The report is a result of the Degree Project of Energy Systems Engineering at Uppsala University. Firstly, the project’s background, purpose, goals and definitions are formulated. This is followed by an extensive theoretical background chapter which include all the necessary theory and background needed in order to understand the rest of the report. The answer to how the project’s purpose and goals were accomplished can be found under the Method section, which also include the limitations and assumptions of the project. The results of the project are compiled in Results, expressed in tables and charts. Further, the results are discussed in Discussion and the conclusions are expressed in Conclusions. Finally, there is a Future work section which contains suggestions of work which should be done in order to get better results. This section also comprises ideas of interesting parameters which should be taken into consideration to get a more complete understanding of the area and better prerequisites to develop an optimized cooling system.

1.1 Background

(12)

2

down the combustion engine. This cooling system is connected to the cooling system of the FC system, which is designed and manufactured by the FC system supplier, PowerCell. In the future, the plan is to implement a 100 kW FC stack in the car. The question is how that will affect the cooling system and what is needed to be done to the cooling system in order for the implementation to work.

1.2 Purpose and goals

The purpose of the project is to investigate if the existing cooling system in the FC test vehicle works with the current FC system in terms of sufficient heat rejection and thus sustaining acceptable temperature levels for the FC system. If the cooling system does not meet these demands the goal is to suggest improvements in order to fulfill the required performance. The project will also investigate the possibility to implement a more powerful FC system with the existing cooling system, including a few necessary modifications. If improvements are needed the goal is to suggest improvements on how a suitable cooling system can be accomplished.

1.3 Definitions

(13)

3

2 Theoretical background

2.1 Fuel cells

FCs convert chemical energy in a fuel into electrical energy [6]. Further, described by [6], FCs consist of an anode, a cathode and an electrolyte. At the anode the oxidation, which is a chemical reaction where electrons (e-_{) are delivered, occurs, for a PEMFC see the anode} reaction below. Similarly, at the cathode the reduction, which is a chemical reaction where e -are gained, occurs, for a PEMFC see the cathode reaction below. In the electrolyte ions carry the current between the electrodes (anode and cathode). For PEMFC typical anode and cathode materials are porous carbon and platinum (used as a catalyst). Characteristically for PEMFCs the electrolyte consists of a proton (H+_{)-conducting polymer membrane. The (anode) fuel is} generally pure hydrogen (H2) and the cathode supply is air or oxygen (O2). At the anode, in the presence of the catalyst, the H2 molecules are dissociated into H+ and free e- [7]. Further, explained by [7], the e- are conducted as usable electric current through an external circuit, while the H+ migrate with water (H2O) through the membrane electrolyte to the cathode. On the cathode side O2, H+ and e- combine into H2O. The only byproducts are H2O and heat, which will be covered later in the report. The operation of a PEMFC is schematically presented in Figure 1 [8]. An important component of the PEMFC is the bipolar plate [9]. Further explained by [9], the bipolar plate is a multi-functional component, whose primarily roles are to supply the reactants to the electrodes and provide electrical connection between adjacent cells.

Anode reaction: 2𝐻2 → 4𝐻++ 4𝑒−

Cathode reaction: 𝑂₂+ 4𝐻++ 4𝑒− → 2𝐻₂𝑂

(14)

4

2.2 Fuel cell vehicles

FC vehicles can in general be divided into two categories: Pure FC vehicles and hybrid FC vehicles [2]. In pure FC vehicles the only energy source is the hydrogen storage and the power demand is solely supplied by the FC system. For the hybrid FC vehicles there is additionally an electric energy storage and also more power electronic components. Further, according to [2], there are both conventional hybrids and plug-in hybrids. In a conventional hybrid the electric energy storage system, batteries or super-/ultracapacitors, stores recuperation braking energy and supplies peaking power. In a plug-in hybrid the electric energy storage system, typically batteries, can be charged from the FC system, braking energy or through connection to the electric grid. This makes it possible to drive some moderate distance on grid-supplied electric power. According to [7] and [2] the main components in a typical (hybrid) FC vehicle are the following:

 FC stack

 Electric machines/motors (EMs)  Power electronics

 Electric energy storage system  Hydrogen storage system

 Air management system: Compressor or blower.  Water management system: Humidifiers.

 Heat management system (called cooling system in this report)

The FC stack consists of FCs connected in series to obtain higher voltages [2]. According to [7] the EMs in FC vehicles normally use alternating current to enable propulsion of the vehicle. Two EM technologies are commonly used: permanent magnet or induction. The power electronics include a motor controller, DC/DC converter and inverter. Their responsibility is to process and control the electric power between the FC stack, battery and EMs. The DC/DC converters convert high direct current (DC) voltage to low DC voltage, which is used to power the auxiliary loads of the vehicle, for example the lightning, windshield wipers and radio. The inverter converts DC power from the FC stack or battery to alternating current power for the EMs, or vice versa when the braking energy is recuperated to the battery. Finally, the motor controller regulates the power to the EM. The current electric energy storage technologies used in FC vehicles are lithium-ion, nickel-metal hydride and lead-acid batteries, as well as super-/ultracapacitors. Regarding the onboard hydrogen storage the most common system is high pressure hydrogen gas tanks, however there are research and development of additional technologies as well.

(15)

5

management system (henceforth called cooling system) is described in 2.4 Cooling systems for fuel cell vehicles.

2.3 Cooling system components

2.3.1 Coolant

According to [10], coolant is a general term that applies to any liquid that removes heat and provides corrosion inhibition. The general requirements of the coolant are:

 Good heat conductor  Low freeze point  High boiling point  Non-corrosive to metals  Non-foaming

 Compatible with other commercial coolants

The typical composition of the coolant consists of [10]:  Water  Glycol  Corrosion inhibitor  De-foamer  Dye (colorant) 2.3.2 Pump

The most commonly used pump in automotive cooling system applications is the electric centrifugal pump according to [11]. Further, [12] describes, that the centrifugal pump is powered by a so called impeller. The rotating impeller creates a pressure rise by transferring mechanical energy from an electric motor to the coolant, which flows from the pump inlet to the impeller center and out along the impeller blades. The centrifugal force increases the coolant velocity and consequently the kinetic energy is transformed to pressure.

2.3.3 Radiator

(16)

6

2.3.4 Condenser

The condenser is not directly connected to the cooling system, but is a component in the Heat Ventilation and Air Conditioning (HVAC) system of the vehicle. Without going into any details about the HVAC system, the purpose of the condenser is to condense the refrigerant, used in the HVAC system. This is done by letting the heat of the high temperature and pressure refrigerant transfer to the ambient air [14]. By doing this the HVAC system transfers the heat from the cabin air to the ambient air.

2.3.5 Fan

The fan consists of an electric motor with a flange-mounted fan wheel [13]. It is used to provide more air flow through radiator (and condenser) when necessary, either at lower vehicle speeds or if there is a high heat rejection demand [11].

2.3.6 3-way Valve

According to [11] a 3-way valve enables controlling of the direction of the coolant flow. For example this can be used to bypass a component, e.g. the radiator, if cooling is not needed in order to recirculate the coolant to increase its temperature. Unlike a conventional thermostat, which controls the coolant flow direction based on coolant temperature, the 3-way valve is electrically controlled by using specified control variables.

2.3.7 Expansion Tank

The expansion tank is used to accumulate expanding coolant from the cooling circuit, which occurs at increasing temperature and pressure [13]. Also, according to [13], the expansion tank cap valves reduce the pressure if it reaches a certain level. When the coolant temperature decreases a partial vacuum forms in the cooling circuit, which is compensated by the expansion tank cap valves by letting ambient air flow into the tank. An additional function of the expansion tank is to remove air bubbles from the cooling circuit [11].

2.3.8 Heat Exchanger

(17)

7

2.3.9 High Voltage Coolant Heater

According to [11], the High Voltage Coolant Heater (HVCH) is used in plug-in hybrid and battery electric vehicles as a heating system for the coolant. It converts electric power, with DC voltages from 250 to 450 V, into heat without loss. This is used to warm up components at low ambient temperature (𝑇_𝑎𝑚𝑏) conditions.

2.4 Cooling systems for fuel cell vehicles

The purpose of the cooling system in FC vehicles is mainly to keep the FC stack’s operating temperature at its optimal, usually in the temperature range from 60 to 80°C [5]. There are several different types of cooling techniques for FC stacks as [5] reviews. However, for high power (>5 kW) FC stacks, hence for automotive applications, liquid cooling systems are more suitable and commonly used [5]. This is due to that the heat transfer coefficients for liquid flow are much higher compared to air flow, [5] explains. The coolant flows in the cooling channels of the FC stack, which are usually integrated in the bipolar plates [5].

According to [5], the liquid coolant is either deionized water or an antifreeze coolant for operation during subzero conditions. However, conventional coolant with a water and glycol mixture becomes normally too electrically conductive, either due to ion contamination from the bipolar plates or ionic production of the glycol. Electrically conductive coolant could cause leakage currents, which reduces FC stack efficiency, lead to coolant electrolysis and possibly cause degradation of bipolar plates. Therefore, the electrical conductivity of the coolant is monitored by using a so called deionizer, which contains a large amount of ion-exchange resin, to remove the ions from the coolant. Also, by adding antioxidants to the coolant a low conductivity is maintained.

For cooling systems in FC vehicles the heat is always transferred from the cooling circuit to the ambient air via a radiator [5]. According to [5], the radiators in FC vehicles are currently large, due to the relatively small temperature difference between the FC stack and the ambient air.

2.5 Heat generation and temperature sensitivity of fuel cells

The following four sources of heat generation can be found in a FC [5]:  Entropic heat of reactions

 Irreversible heat of electrochemical reactions  Heat from ohmic resistances

 Heat from condensation of water vapor

(18)

non-8

uniform over the FC and varies depending on local current density, temperature, reactant concentration and water content.

In general, FCs have better performance at higher temperatures, due to the lower activation energy in the reaction kinetics [3]. However, there is an upper temperature limit for FCs which should not be exceeded, according to [3]. Further explained by [3], as the temperature increases, the humidification water mass flow should increase in order to avoid membrane dehydration. The ohmic losses increases at high temperatures if the membrane is dehydrated and consequently the efficiency of the FC decreases.

2.6 Heat flow rate formula

The heat input rate to the cooling system from the FC stack and the Air Compressor in this project can be expressed using Equation 1. As can be seen the formula consists of the phisical quantaty specific heat (also known as specific heat capacity), c. The specific heat is the amount of heat per unit mass required to raise the temperature by one degree Celsius.

(19)

9

3 Overview of the cooling system

As previous mentioned the cooling system of Volvo Cars’ FC test vehicle consists of two separate cooling systems; the outer and inner cooling system. A schematic cooling system layout, including these two cooling systems and their cooling system components, can be seen in Figure 2.

(20)

10

3.1 Outer cooling system

The outer cooling system comprises cooling system components from the previous cooling system used for a combustion engine, i.e. these components are Volvo Cars parts. The outer cooling system layout is similar to the one used for a combustion engine, however some arrangements have been done in order for it to be compatible with the FC system.

The outer cooling system consists of the coolant pump which pumps the coolant into the HX, where heat is transferred from the inner cooling system to the outer cooling system. After the HX the coolant can flow two ways. Either to the radiator, where heat is rejected to the ambient air. Or the coolant can flow to the Climate Circuit, where the heat is used to heat up the air in the cabin. The heat which can be rejected in the radiator depends on the air mass flow rate through the radiator. The air mass flow rate depends on vehicle speed and also fan speed. The Climate Circuit consists of the coolant pump which pumps the coolant into the HVCH. The HVCH is an electric water heater which is used to heat up the coolant even more if there is a high demand of heat to the cabin of the car. The next component is the cabin heater itself, where the heat is transferred from the coolant to the cabin air. As long as there is a heat demand to the cabin heater the coolant is bypassed via the 3-way valve, i.e. the coolant is circulated in the Climate Circuit. If the heat demand to the cabin heater is decreased the 3-way valve is opened and the coolant flows to the radiator.

Adjacent to the radiator there is the condenser, whose purpose is to dissipate heat from the air, using a refrigerant, in order to have cool cabin air. After the radiator the coolant flows either back to the pump or to the expansion tank. The coolant flows to the expansion tank if it is expanded due to rising temperature and pressure. It flows back to the cooling circuit at decreasing temperature and pressure. In the outer cooling system a conventional coolant is used, containing 50 % water and 50 % ethylene glycol.

3.2 Inner cooling system

(21)

11

3.3 Overview of the fuel cell test vehicle

Volvo Cars’ FC test vehicle is based on a plug-in hybrid vehicle, whose internal combustion engine is replaced with the FC system. Its cooling system is described in 3.1 Overview of the cooling system. The FC system consists of an air management system, e.g. an AC, and water management system, e.g. a humidifier. The vehicle’s powertrain components and the energy storage systems are specified bellow:

 FC stack: 25 kW, 125-250 V  EM: 60 kW, 240 Nm

 Battery: 8.8 kWh, 400 V  Inverter

 Converter (for auxiliary loads): 400 V to 12 V  Converter (for FC): 125-250 V to 400 V  Hydrogen tank: 1.43 kg, 700 bar

(22)

12

4 Method

A model of the existing cooling system was built in the CAE (Computer-Aided Engineering) simulation software GT-Suite. The model of the existing cooling system and the current FC system is called Model 1 in this report. Both steady state and transient 1D CFD (One-Dimensional Computational Fluid Dynamics) simulations were performed, representing various demanding driving cases, for the system. Then a new model, referred to as Model 2 in this report, for the more powerful FC was built, by keeping the same cooling system but replacing the current FC system with the more powerful FC system. Similarly to Model 1, steady state and transient simulations for Model 2 were performed in order to examine if the system works within acceptable temperature limits. Improvement proposals were based on the analysis of the simulation results, interviewing Volvo Cars and PowerCell personnel and finally improving the performance for components of interest in the models and performing simulations. All accomplishments of the project were compiled in this report.

4.1 Modeling of the cooling system

The modeling of the cooling system was accomplished by several different procedures. Firstly, documents about the different cooling systems in the cars of Volvo Cars was read. Then, a literature review was conducted regarding FC vehicles and their cooling systems. Cooling and thermal management tutorials in the simulation software were performed. The FC test vehicle was viewed at Volvo Cars’ prototype workshop in order to understand how the outer cooling system works and analyzing the layout of the cooling system. Models for all of the Volvo Cars’ cooling system components were available for the simulation software. However, all of the connections, consisting of hoses, between the components were not available for the current outer cooling system layout and consequently their dimensions were measured and then entered into the model.

The FC supplier, PowerCell, was visited and personnel was interviewed in order to understand how the inner cooling system works and its layout. Additionally, during the visit, information about the FC stack was comprehended, mainly with focus on the thermal aspects of the FC stack. The suppliers for the different cooling system components were contacted to obtain the data which is necessary to build a model of the component in question. The resulting models of the different cooling system components are explained and can be seen in section 4.4 Models.

4.2 Simulation of the cooling system

(23)

13

stage of the FC test vehicle project. The transient simulations were performed for different ambient temperatures. More detailed information about the simulations can be found in section 4.5 Simulations. Similarly, these simulations were performed for Model 2 as well, however the simulations were performed for different 𝑃_𝐹𝐶which are more applicable for the system. For all performed simulations the results were analyzed by creating plots, using Excel, consisting of different parameters of interest.

4.3 Limitations and Assumptions

Because the modeling of the cooling system is dependent on data, specifications and information about the different components, a lot of work was made to collect as much of this data as possible in order to have a representative model. However, this was not available for all components and then assumptions were made based on the knowledge of how a realistic component or system would be designed. Both for Model 1 and Model 2 there were several assumptions made, which are described in the section 4.4 Models. One general assumption for the Outer Cooling System is that some of the hoses are assumed to be straight, due to the unavailability to estimate the bends of these hoses.

As the simulations were performed there were no heat input contribution to the cooling system from other components than the FC stack and AC. This means for example that the heat input of the condenser and HVCH is not taken into consideration in this project. Further, any heat rejection from the Cabin Heater is excluded as well. The Climate Pump is assumed to always perform at its maximum speed. Also, the Climate 3-way valve is assumed to always be open. Air pressure is assumed to be the average sea-level pressure which is equal to 101.325 kPa for all simulations. Hence, the eventual deviations in altitude and weather are not accounted for. The simulation scenario parameter values, determined in GT-Suite’s Case Setup, are either sent directly to the affected component in question or sent to a Control System model, which receives the parameter values from Case Setup and sends them to the respective component. This Control System model is excluded in this report.

4.4 Models

(24)

14

The data used to model the components is provided by the supplier, if nothing else is stated in this report. However, as it is difficult to obtain all the required information, assumptions were made. Especially the modeling of the control systems for the components were assumed, based on the limited information of how they worked, interviews with Volvo Cars personnel and analysis of simulation results which were obtained during the modeling process.

4.4.1 Cooling Package (radiator, fan & condenser)

The Cooling Package consists of the radiator, fan and condenser. The schematic layout for the model of the Cooling Package can be seen in Figure 3. The coolant enters either via a hose from the HX or from the Climate Circuit. It then goes through the master side of the radiator and exists to the Outer Pump and to a certain degree to the Outer Expansion Tank. The master side of the radiator is representing the heat transfer between the coolant and the radiator walls. Further, the air first goes through the condenser before going through the slave side of the radiator and finally exists to the End Environment. The slave side of the radiator is representing the heat transfer between the radiator walls and the ambient air. The fan is not modeled explicitly in this model, due to the unavailability of data. Its air flowing capabilities is modeled in a map together with the air mass flow rates through the radiator for different vehicle speeds, this can be seen in Figure 4. The fan is assumed to be 100 % ON (i.e. rotating at its maximum speed) at 0 km/h and to contribute less significantly to the air mass flow rate through the radiator with increased vehicle speed.

(25)

15

Figure 4 Air mass flow rate through the radiator for different vehicle speeds.

4.4.2 Outer Pump

The schematic layout for the model of the outer pump can be seen in Figure 5. The hose from the radiator and the hose from the outer expansion tank are connected via a T-split connection which is directly connected to the outer pump. The pump performance is modeled by using data of the pressure rise for different volume flow rates, pump speeds and isentropic efficiencies. The consequential values of the pressure rise as a function of volume flow rate for different pump speeds can be seen in the Pump Map in Figure 6. The values of these parameters vary depending on the cooling demand. When a test of the FC test vehicle was performed a maximum coolant flow rate of around 52 L/min was measured and this is modeled by decreasing the inner diameter of the hose before the pump from 16 mm to 11 mm.

Figure 5 Schematic layout of the Outer Pump model.

(26)

16

Figure 6 Pump performance map of the Outer Pump.

Due to the lack of information of how the control of the outer pump speed will be designed, an assumption of a likely control design is used in this project. It is modeled by receiving a signal from the inner pump speed and multiplying it with 1.3, due to their different maximum pump speeds (𝜔_{𝑜𝑢𝑡𝑒𝑟,𝑚𝑎𝑥} and 𝜔_{𝑖𝑛𝑛𝑒𝑟,𝑚𝑎𝑥}), see Equation 2, and sending it to the Pump Speed Control Unit. This means that the outer pump speed (𝜔𝑜𝑢𝑡𝑒𝑟 ) will always follow the inner pump speed (𝜔_{𝑖𝑛𝑛𝑒𝑟}), though 1.3 times higher.

𝜔 𝑜𝑢𝑡𝑒𝑟,𝑚𝑎𝑥 𝜔𝑖𝑛𝑛𝑒𝑟,𝑚𝑎𝑥 =

5900 rpm

4600 rpm≈ 1.3 (2)

4.4.3 Outer Expansion Tank

(27)

17

Figure 7 Schematic layout of the Outer Expansion Tank model.

4.4.4 Climate Circuit

In Figure 8 the schematic layout of the Climate Circuit can be seen. The coolant flows via a hose from the HX to the Climate Pump. The Climate Pump speed is set to its maximum speed in this project in order to have coolant flowing in the Climate Circuit, which includes the pressure drops occurring as the coolant flows though the Climate Circuit. After the Climate Pump the coolant flows via a hose to the HVCH and then to the Cabin Heater. Returning from the Cabin Heater, the coolant flows to the Climate 3-way Valve, which is always open in this project, i.e. the coolant is never circulated in the Climate Circuit. The control signals for the Climate Pump speed, HVCH heat input and Climate 3-way Valve position are received in the Receive Signal template and sent to each component respectively.

(28)

18

4.4.5 Cabin Heater

The schematic layout for the Cabin Heater can be seen in Figure 9. The coolant is flowing from the Climate Circuit and enters the Cabin Heater in order to provide heat to the cabin. This is done by letting the cabin air flow through the Cabin Heater as well.

Figure 9 Schematic layout of the Cabin Heater model.

4.4.6 Heat Exchanger

The model of the HX can be seen in the schematic layout in Figure 10. The HX master side is representing the part of the HX where the heat transfer between the coolant (of the Inner Cooling System circuit) and the HX pipe walls occurs. Similarly, the HX slave side is representing the part of the HX where the heat transfer between the HX pipe walls and the coolant (of the Outer Cooling System circuit) occurs. It is in the HX Master object where the HX geometry, heat transfer properties and master and slave pressure drops are defined.

Figure 10 Schematic layout of the Heat Exchanger model.

4.4.7 Inner Pump

(29)

19

Figure 11 Schematic layout of the Inner Pump model.

Figure 12 Pump performance map of the Inner Pump.

(30)

20

Table 1 Inner Pump speed as a function of temperature difference of the inlet and outlet coolant of the FC stack.

∆𝑻𝑭𝑪 [°C] 𝝎𝒊𝒏𝒏𝒆𝒓[rpm]

0 500

5 2500

10 4600

Table 2 Inner Pump start-up time delay for different ambient temperatures.

𝑻_𝒂𝒎𝒃 [°C] 𝒕_{𝒑𝒖𝒎𝒑 𝒅𝒆𝒍𝒂𝒚} [s]

-15 60

20 30

45 10

4.4.8 Inner Expansion Tank

Due to the lack of information about the Inner Expansion Tank, it is assumed to be the same as the Outer Expansion Tank. It is modeled exactly the same (see 4.4.3 Outer Expansion Tank), however both the hose to the inlet manifold and the hose from the outlet manifold are found in the model for the Inner Cooling System (see 4.4.12 Inner Cooling System).

4.4.9 Fuel Cell Stack

The schematic layout for the FC stack can be seen in Figure 13. The FC Stack is modeled with a so called Heat Addition template. To model it the needed data are the volume of coolant inside the component, pressure drop for different volume flow rate and heat input rate, referred to as 𝑄_𝐹𝐶 in this report. The 𝑄_𝐹𝐶 is provided via a so called Lookup Table, where the input parameter FC stack output power (𝑃_𝐹𝐶) (from the Signal Generator template) is used to determine the 𝑄𝐹𝐶. The 𝑄𝐹𝐶 is calculated by using Equation 3. For Model 1 the data for the FC stack efficiency (𝜂_𝐹𝐶) was provided by PowerCell for different 𝑃_𝐹𝐶. For Model 2 the 𝑄_𝐹𝐶 was provided for different 𝑃_𝐹𝐶, i.e. Equation 3 was not needed. As can be seen in Figure 13 the inlet and outlet coolant temperature of the FC stack (𝑇_{𝐹𝐶,𝑖𝑛} and 𝑇_{𝐹𝐶,𝑜𝑢𝑡}) are measured. These measurements are used to calculate the mean temperature of the inlet and outlet temperature of the coolant (𝑇_{𝐹𝐶,𝑚𝑒𝑎𝑛}), see Equation 4, which is sent to the Inner 3-way valve. Also, the temperature measurements are used to calculate the ∆𝑇𝐹𝐶, see Equation 5, which is sent to the Inner Pump.

𝑄_𝐹𝐶 = 𝑃_𝑖𝑛− 𝑃_𝐹𝐶 = [𝑃_𝐹𝐶 = 𝜂_𝐹𝐶 𝑃_𝑖𝑛] = (1 − 𝜂𝐹𝐶)𝑃𝐹𝐶_𝜂 𝐹𝐶 = (

1

𝜂𝐹𝐶− 1) 𝑃𝐹𝐶 (3) Where 𝑃_𝑖𝑛 is the input power supplied to the FC stack, i.e. the hydrogen supply.

𝑇_{𝐹𝐶,𝑚𝑒𝑎𝑛} = 𝑇𝐹𝐶,𝑜𝑢𝑡+𝑇𝐹𝐶,𝑖𝑛

2 (4)

(31)

21

Figure 13 Schematic layout of the Fuel Cell Stack model.

4.4.10 Air Compressor

(32)

22

Figure 14 Schematic layout of the Air Compressor model.

4.4.11 Inner 3-way Valve

The schematic layout for the model of the Inner 3-way valve can be seen in Figure 15. The Inner 3-way valve is modeled by using the Pressure Loss Connection, which uses data for the pressure drop for different volume flow rates. The valve position is controlled by the 𝑇𝐹𝐶,𝑚𝑒𝑎𝑛 (defined in 4.4.9 Fuel Cell Stack). The valve is closed at 𝑇𝐹𝐶,𝑚𝑒𝑎𝑛 below 75±1°C, letting the coolant bypass the HX. At 𝑇_{𝐹𝐶,𝑚𝑒𝑎𝑛} above 75±1°C the valve is open, allowing the coolant to pass through the HX. This is controlled in the Hysteresis Switch template, whose output signal (y) decides the inner diameter (ID) of the pipes of the Inner 3-way valve. If the valve is closed the output signal is 0 mm, which results in a closed HX branch and an open bypass branch, due to the Math Equation template, which is calculating: ID – y. If the valve is open the output signal is ID mm, resulting in the opposite scenario compared to the closed valve case.

(33)

23

4.4.12 Inner Cooling System

The model of the Inner Cooling System can be seen in the schematic layout in Figure 16. As can be seen, the different components are connected via hoses and T-split connections. The length and bends of the hoses are estimated based on a 3D system layout. Also, the control systems are connected via connection lines. The Restrictor valve (see Figure 2) is not modeled explicitly, but by modeling the inlet hose to the Inner Expansion Tank with a relatively small ID, defined as 2 mm. Note that the hoses and T-split connection connected to the left (i.e. slave) side of the HX belong to the Outer Cooling System but are part of the Inner Cooling System model.

Figure 16 Schematic layout of the Inner Cooling System model.

4.4.13 Modifications from Model 1 to Model 2

(34)

24

Realistic Modifications

 FC stack specifications:

o Lookup Table for 𝑄_𝐹𝐶 versus 𝑃_𝐹𝐶.

o Table for pressure drop versus volume flow rate. o Volume of coolant inside component.

 More powerful Inner Pump:

o Volume flow rate multiplied with 2. o Pressure rise multiplied with 2.

 Larger ID for hoses in the Inner Cooling System: o Generally: From 19 mm to 40 mm.

o Inlet hose for Inner Expansion Tank: From 2 mm to 4 mm. o Hoses to/from AC: From 6 mm to 12 mm.

Simulation Modifications

The Inner Expansion Tank was flooded when simulations for Model 2 were performed, which resulted in simulation failure. Therefore, the following specifications were modified:

 Inner Expansion Tank specifications:

o Volume multiplied with 2: From 2.56 L to 5.12 L.

o Expansion tank cap opens at pressure above 4 bar, compared to 2.45 bar.

Simplifying Modifications

 The 𝑃_𝐹𝐶 used to determine 𝑄_𝐴𝐶 is multiplied with 0.25, based on the difference of 𝑃_𝐹𝐶 for Model 1 and Model 2. This results in the same amount of 𝑄_𝐴𝐶 is generated for both models.  The hoses between the components in the Inner Cooling System are all 10 cm long and

have no bends.

4.5 Simulations

This section describes the relevant input parameters for the simulations performed for Model 1 and Model 2.

4.5.1 Steady State Simulations

(35)

25

4.5.2 Transient Simulations

By performing transient simulations it was possible to analyze how the cooling system responds to different cases of operation over time. For the simulations three different driving cycles were simulated: The Berlin Cycle (BC), Worldwide harmonized Light vehicles Test Procedure (WLTP) and Vehicle Performance (VP). The BC is a driving cycle recorded in Berlin with urban, interurban and city highway sections. The WLTP driving cycle is a global harmonized standard used to determine the performance of light-duty vehicles (passenger cars and light commercial vans). The VP driving cycle tests the acceleration and top speed of the vehicle by accelerating from standstill to top speed as fast as possible. For the BC and VP the battery initial state of charge (SOC) was 95 % and for the WLTP the initial SOC was 60 %. The SOC became 0 % for VP after 12 min and 42 s (762 s) and the car was solely driven by power from the FC stack from that point. Hence, there is a decrease of (top) speed at that point of time, but the 𝑃𝐹𝐶 is unchanged. For Model 1 the corresponding vehicle speeds and 𝑃𝐹𝐶 for the different driving cycles can be seen in Figure 17, Figure 18 and Figure 19.

Figure 17 FC Stack Output Power and vehicle speed for the BC driving cycle for Model 1.

Figure 18 FC Stack Output Power and vehicle speed for the WLTP driving cycle for Model 1.

0 20 40 60 80 100 120 140 0 5 10 15 20 25 0 1000 2000 3000 4000 5000 Sp ee d [ km/h ] Po w er [kW ] Time [s]

FC Stack Output Power [kW] Vehicle Speed [km/h]

0 20 40 60 80 100 120 140 0 2 4 6 8 10 12 14 16 0 200 400 600 800 1000 1200 1400 1600 1800 Sp ee d [ km/h ] Po w er [kW ] Time [s]

(36)

26

Figure 19 FC Stack Output Power and vehicle speed for the VP driving cycle for Model 1.

As can be seen in Figure 17, Figure 18 and Figure 19 the 𝑃_𝐹𝐶 is constantly 3.9 kW even when the vehicle speed is 0 km/h. This demonstrates the use of the FC system as a range extender, i.e. the battery is constantly charged by the FC system. The driving cycles were simulated at the following 𝑇𝑎𝑚𝑏: 20°C, 45°C and -15°C.

The same driving cycles were simulated for Model 2 as for Model 1. However, for Model 2 the transient simulations were performed for 𝑃𝐹𝐶 more applicable for the more powerful FC system. This was done by adding the power supplied by the battery (𝑃_{𝑏𝑎𝑡𝑡𝑒𝑟𝑦}) to the 𝑃_𝐹𝐶 used in the simulations of Model 1. The total maximum power supplied both from the FC stack and the battery is about 60 kW, due to the maximum power of the EM. Furthermore, the sum of these powers were scaled up by a factor of 5₃ (motivated by Equation 6) over time. This power is defined to be the 𝑃_𝐹𝐶 for the simulations of Model 2 and is considered to be more applicable for the 100 kW FC stack.

𝑃𝐹𝐶,𝑚𝑎𝑥 (for Model 2) 𝑃𝐹𝐶,𝑚𝑎𝑥+𝑃𝑏𝑎𝑡𝑡𝑒𝑟𝑦,𝑚𝑎𝑥 (for Model 1)= 100 kW 60 kW = 5 3 (6)

The 𝑃_𝐹𝐶 of Model 2 for the different driving cycles can be seen in Figure 20, Figure 21 and Figure 22. For simplicity, the vehicle speed is assumed to be the same as for the simulations of Model 1. 0 20 40 60 80 100 120 140 160 180 0 5 10 15 20 25 0 200 400 600 800 1000 1200 Sp ee d [ km/h ] Po w er [kW ] Time [s]

(37)

27

Figure 21 FC Stack Output Power and vehicle speed for the WLTP driving cycle for Model 2.

Figure 22 FC Stack Output Power and vehicle speed for the VP driving cycle for Model 2.

0 20 40 60 80 100 120 140 160 180 0 10 20 30 40 50 60 70 80 90 100 110 120 0 200 400 600 800 1000 1200 Sp ee d [ km/h ] Po w er [kW ] Time [s]

0 20 40 60 80 100 120 140 0 10 20 30 40 50 60 70 80 90 100 110 120 0 1000 2000 3000 4000 5000 Sp ee d [ km/h ] Po w er [kW ] Time [s]

Figure 20 FC Stack Output Power and vehicle speed for the BC driving cycle for Model 2.

0 20 40 60 80 100 120 140 0 10 20 30 40 50 60 70 80 90 100 0 200 400 600 800 1000 1200 1400 1600 1800 Sp ee d [ km/h ] Po w er [kW ] Time [s]

(38)

28

5 Results

5.1 Simulation results for the current fuel cell system (Model 1)

In this section the simulation results for the existing cooling system with the current FC system, which has the FC stack with 25 kW of maximum 𝑃_𝐹𝐶.

5.1.1 Steady state results

In Figure 23 the 𝑄𝐹𝐶 and 𝑄𝐴𝐶 can be seen for the different simulated 𝑃𝐹𝐶 for Model 1. It can be seen that the 𝑄𝐹𝐶 is higher than the 𝑃𝐹𝐶 for 𝑃𝐹𝐶 >20 kW, thus the 𝜂𝐹𝐶 is lower than 50% at higher 𝑃_𝐹𝐶. Note that the 𝑄_𝐴𝐶 is significantly smaller than 𝑄_𝐹𝐶 and does not contribute to any substantial (only maximum 1.25 %) heat addition.

Figure 23 Steady state simulation results for Model 1: Heat Generation from the FC stack and AC as functions of FC Stack Output Power.

The simulation results for Model 1 can be validated by using Equation 1 to calculate 𝑄𝐹𝐶 and 𝑄_𝐴𝐶. In Table 3 the steady state values of 𝑇_{𝐹𝐶,𝑖𝑛}, 𝑇_{𝐹𝐶,𝑜𝑢𝑡} and the corresponding ∆𝑇_𝐹𝐶 are presented. Also, the mass flow rate of the coolant at the FC (𝑚̇𝐹𝐶) and the specific heat for the coolant at the FC (𝑐𝐹𝐶) are presented. Finally, by using these values in Equation 1 the calculated 𝑄_𝐹𝐶 is obtained. Similarly for the AC, the values of the inlet and outlet coolant temperature (𝑇𝐴𝐶,𝑖𝑛 and 𝑇𝐴𝐶,𝑜𝑢𝑡) and the corresponding coolant temperature difference (∆𝑇𝐴𝐶) are presented in Table 4. Also, the mass flow rate of the coolant at the AC (𝑚̇_𝐴𝐶) and the specific heat for the coolant at the AC (𝑐_𝐴𝐶) are presented. By using these values in Equation 1 the calculated 𝑄_𝐴𝐶 is obtained.

As can be seen in Table 3 and Table 4 the calculated 𝑄_𝐹𝐶 and 𝑄_𝐴𝐶 correspond (with only minor deviations) to the simulated 𝑄_𝐹𝐶 and 𝑄_𝐴𝐶 in Figure 23. This means that the (thermal) simulation results of Model 1 are valid according to the physical relationship of Equation 1.

0 0,05 0,1 0,15 0,2 0,25 0,3 0,35 0,4 0,45 0 5 10 15 20 25 30 35 0 5 10 15 20 25 He at Flow Rate , Q AC [kW] H eat Flo w Rat e, QFC [kW ]

FC Stack Output Power, PFC[kW]

(39)

29

Table 3 Calculated Heat Generation from FC stack for Model 1 by using the mass flow rate, specific heat and temperature difference for the coolant at the FC stack.

5.1.2 Transient results

In Figure 24 the 𝑃𝐹𝐶 and 𝑄𝐹𝐶 can be seen for the BC driving cycle. By analyzing Figure 24 it is possible to comprehend the resulting coolant temperatures. The 𝑇_{𝐹𝐶,𝑖𝑛} and 𝑇_{𝐹𝐶,𝑜𝑢𝑡} for the different 𝑇_𝑎𝑚𝑏 (20°C, 45°C and -15°C) can be seen in Figure 25, Figure 26 and Figure 27. As can be seen the 𝑇_{𝐹𝐶,𝑜𝑢𝑡} is maximum 82-84°C in this driving cycle, which is below the maximum acceptable 𝑇_{𝐹𝐶,𝑜𝑢𝑡} of 90°C. This means that the cooling system sustains sufficient heat rejection. Note that with increased 𝑇𝑎𝑚𝑏 the maximum 𝑇𝐹𝐶,𝑜𝑢𝑡 is only increased slightly. The reason for the large drops in 𝑇𝐹𝐶,𝑖𝑛 is discussed in section 5, this applies for all plots of 𝑇𝐹𝐶,𝑖𝑛 in this report. 𝑷𝑭𝑪 [kW] 1 5 10 15 20 22.5 25 𝑻_{𝑭𝑪,𝒊𝒏} [°C] 22.8 32.8 41.2 50.2 61.0 68.0 76.2 𝑻_{𝑭𝑪,𝒐𝒖𝒕} [°C] 23.8 35.3 45.2 55.4 67.7 75.4 84.4 𝒎̇_𝑭𝑪 [kg/s] _0.16 _0.32 _0.48 _0.62 _0.78 _0.87 _0.97 𝒄𝑭𝑪 [J/kg,°C] 3740 3780 3820 3850 3900 3920 3960 ∆𝑻_𝑭𝑪 [°C] _1.0 _2.5 _4.0 _5.2 _6.7 _7.4 _8.2 𝑸𝑭𝑪 [kW] (calculated) 0.6 3.0 7.3 12.4 20.4 25.2 31.5

Table 4 Calculated Heat Generation from AC for Model 1 by using the mass flow rate, specific heat and temperature difference for the coolant at the AC.

(40)

30

Figure 24 Heat Generation from the FC stack and FC Stack Output Power for the simulated BC driving cycle for Model 1.

Figure 25 Inlet and outlet coolant temperature of the FC stack for the simulated BC driving cycle at the ambient temperature of 20°C for Model 1.

0 5 10 15 20 25 30 0 1000 2000 3000 4000 5000 Po w er / H eat Flo w Rat e [kW] Time [s]

Heat Generation from FC Stack [kW] FC Stack Output Power [kW]

0 10 20 30 40 50 60 70 80 90 0 1000 2000 3000 4000 5000 Te m p era tu re [ °C] Time [s]

(41)

31

Figure 27 Inlet and outlet coolant temperature of the FC stack for the simulated BC driving cycle at the ambient temperature of -15°C for Model 1.

In Figure 28 the 𝑃𝐹𝐶 and 𝑄𝐹𝐶 can be seen for the WLTP driving cycle. By analyzing Figure 28 it is possible to comprehend the resulting coolant temperatures. The 𝑇_{𝐹𝐶,𝑖𝑛} and 𝑇_{𝐹𝐶,𝑜𝑢𝑡} for the different 𝑇_𝑎𝑚𝑏 (20°C, 45°C and -15°C) can be seen in Figure 29, Figure 30 and Figure 31. The 𝑇_{𝐹𝐶,𝑜𝑢𝑡} is maximum 77-78°C in this driving cycle, which is below the maximum acceptable 𝑇_{𝐹𝐶,𝑜𝑢𝑡} of 90°C. This means that the cooling system sustains sufficient heat rejection. Note that with increased 𝑇𝑎𝑚𝑏 the maximum 𝑇𝐹𝐶,𝑜𝑢𝑡 is only increased slightly.

0 10 20 30 40 50 60 70 80 90 0 1000 2000 3000 4000 5000 Te m p era tu re [ °C] Time [s]

Inlet Temperature FC Stack [°C] Outlet Temperature FC Stack [°C]

-20 0 20 40 60 80 100 0 1000 2000 3000 4000 5000 Te m p era tu re [ °C] Time [s]

(42)

32

Figure 28 Heat Generation from the FC stack and FC Stack Output Power for the simulated WLTP driving cycle for Model 1.

Figure 29 Inlet and outlet coolant temperature of the FC stack for the simulated WLTP driving cycle at the ambient temperature of 20°C for Model 1.

0 2 4 6 8 10 12 14 16 0 200 400 600 800 1000 1200 1400 1600 1800 Po w er / H eat Flo w Rat e [kW] Time [s]

Heat Generation from FC Stack [kW] FC Stack Output Power [kW]

0 10 20 30 40 50 60 70 80 90 0 200 400 600 800 1000 1200 1400 1600 1800 Te m p era tu re [ °C] Time [s]

(43)

33

Figure 31 Inlet and outlet coolant temperature of the FC stack for the simulated WLTP driving cycle at the ambient temperature of -15°C for Model 1.

In Figure 32 the 𝑃𝐹𝐶 and 𝑄𝐹𝐶 can be seen for the VP driving cycle. By analyzing Figure 32 it is possible to comprehend the resulting coolant temperatures. The 𝑇_{𝐹𝐶,𝑖𝑛} and 𝑇_{𝐹𝐶,𝑜𝑢𝑡} for the different 𝑇_𝑎𝑚𝑏 (20°C, 45°C and -15°C) can be seen in Figure 33, Figure 34 and Figure 35. The 𝑇_{𝐹𝐶,𝑜𝑢𝑡} is maximum 78-87°C in this driving cycle, which is below the maximum acceptable 𝑇_{𝐹𝐶,𝑜𝑢𝑡} of 90°C. This means that the cooling system sustains sufficient heat rejection. Note that with increased 𝑇𝑎𝑚𝑏 the maximum 𝑇𝐹𝐶,𝑜𝑢𝑡 is increased.

0 10 20 30 40 50 60 70 80 90 0 200 400 600 800 1000 1200 1400 1600 1800 Te m p era tu re [ °C] Time [s]

-20 -10 0 10 20 30 40 50 60 70 80 90 0 200 400 600 800 1000 1200 1400 1600 1800 Te m p era tu re [ °C] Time [s]

(44)

34

Figure 32 Heat Generation from the FC stack and FC Stack Output Power for the simulated VP driving cycle for Model 1.

Figure 33 Inlet and outlet coolant temperature of the FC stack for the simulated VP driving cycle at the ambient temperature of 20°C for Model 1. 0 5 10 15 20 25 30 0 200 400 600 800 1000 1200 Po w er / H eat Flo w Rat e [kW] Time [s]

FC Stack Output Power [kW] Heat Generation from FC Stack [kW]

0 10 20 30 40 50 60 70 80 90 0 200 400 600 800 1000 1200 Te m p era tu re [ °C] Time [s]

(45)

35

Figure 34 Inlet and outlet coolant temperature of the FC stack for the simulated VP driving cycle at the ambient temperature of 45°C for Model 1.

Figure 35 Inlet and outlet coolant temperature of the FC stack for the simulated VP driving cycle at the ambient temperature of -15°C for Model 1. 0 10 20 30 40 50 60 70 80 90 0 200 400 600 800 1000 1200 Te m p era tu re [ °C] Time [s]

-20 -10 0 10 20 30 40 50 60 70 80 90 0 200 400 600 800 1000 1200 Te m p era tu re [ °C] Time [s]

(46)

36

The simulation results for Model 2 can be validated by using Equation 1 to calculate 𝑄_𝐹𝐶 and 𝑄𝐴𝐶. In Table 5 the steady state values of 𝑇𝐹𝐶,𝑖𝑛, 𝑇𝐹𝐶,𝑜𝑢𝑡 and the corresponding ∆𝑇𝐹𝐶 are presented. Also, the 𝑚̇_𝐹𝐶 and 𝑐_𝐹𝐶 are presented. Finally, by using these values in Equation 1 the calculated 𝑄𝐹𝐶 is obtained. Similarly for the AC, the values of 𝑇𝐴𝐶,𝑖𝑛 and 𝑇𝐴𝐶,𝑜𝑢𝑡 and the corresponding ∆𝑇_𝐴𝐶 are presented in Table 6. Also, the 𝑚̇_𝐴𝐶 and 𝑐_𝐴𝐶 are presented. By using these values in Equation 1 the calculated 𝑄_𝐴𝐶 is obtained.

As can be seen in Table 5 and Table 6 the calculated 𝑄_𝐹𝐶 and 𝑄_𝐴𝐶 correspond (with only minor deviations) to the simulated 𝑄_𝐹𝐶 and 𝑄_𝐴𝐶 in Figure 36. This means that the (thermal) simulation results of Model 2 are valid according to the physical relationship of Equation 1.

5.2 Simulation results for the more powerful fuel cell system (Model 2)

In this section the simulation results for the cooling system with the more powerful FC system, which has the FC stack with 100 kW of maximum 𝑃_𝐹𝐶.

5.2.1 Steady state results

In Figure 36 the 𝑄𝐹𝐶 and 𝑄𝐴𝐶 can be seen for the different simulated 𝑃𝐹𝐶 for Model 2. It can be seen that the 𝑃_𝐹𝐶 is continuously higher than the 𝑄_𝐹𝐶, thus the 𝜂_𝐹𝐶 is constantly higher than 50%. Similarly as for Model 1, the 𝑄_𝐴𝐶 is significantly smaller than 𝑄_𝐹𝐶 and does not contribute to any substantial (only maximum 0.45 %) heat addition.

Figure 36 Steady state simulation results for Model 2: Heat Generation from the FC stack and AC as functions of FC Stack Output Power. 0 0,05 0,1 0,15 0,2 0,25 0,3 0,35 0,4 0,45 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100 H eat Flo w Rat e, QAC [kW ] H eat Flo w Rat e, QFC [kW ]

FC Stack Output Power, PFC[kW]

(47)

37

Table 5 Calculated Heat Generation from FC stack for Model 2 by using the mass flow rate, specific heat and temperature difference for the coolant at the FC stack.

Table 6 Calculated Heat Generation from AC for Model 2 by using the mass flow rate, specific heat and temperature difference for the coolant at the AC.

5.2.2 Transient results

In Figure 37 the 𝑃𝐹𝐶 and 𝑄𝐹𝐶 can be seen for the BC driving cycle. By analyzing Figure 37 it is possible to comprehend the resulting coolant temperatures. The 𝑇_{𝐹𝐶,𝑖𝑛} and 𝑇_{𝐹𝐶,𝑜𝑢𝑡} for the different 𝑇_𝑎𝑚𝑏 (20°C, 45°C and -15°C) can be seen in Figure 38, Figure 39 and Figure 40. The 𝑇_{𝐹𝐶,𝑜𝑢𝑡} is maximum 90-121°C in this driving cycle, which is above the maximum acceptable 𝑇_{𝐹𝐶,𝑜𝑢𝑡} of 90°C. This means that the cooling system does not sustain sufficient heat rejection. Note that with increased 𝑇𝑎𝑚𝑏 the maximum 𝑇𝐹𝐶,𝑜𝑢𝑡 is increased significantly.

(48)

38

Figure 37 Heat Generation from the FC stack and FC Stack Output Power for the simulated BC driving cycle for Model 2.

0 10 20 30 40 50 60 70 80 90 100 110 120 0 1000 2000 3000 4000 5000 Po w er / H eat Flo w Rat e [kW] Time [s]

0 10 20 30 40 50 60 70 80 90 100 110 0 1000 2000 3000 4000 5000 Te m p era tu re [ °C ] Time [s]

(49)

39

Figure 40 Inlet and outlet coolant temperature of the FC stack for the simulated BC driving cycle at the ambient temperature of -15°C for Model 2.

In Figure 41 the 𝑃_𝐹𝐶 and 𝑄_𝐹𝐶 can be seen for the WLTP driving cycle. By analyzing Figure 41 it is possible to comprehend the resulting coolant temperatures. The 𝑇𝐹𝐶,𝑖𝑛 and 𝑇𝐹𝐶,𝑜𝑢𝑡 for the different 𝑇_𝑎𝑚𝑏 (20°C, 45°C and -15°C) can be seen in Figure 42, Figure 43 and Figure 44. The 𝑇𝐹𝐶,𝑜𝑢𝑡 is maximum 86°C, 95°C and 84°C in this driving cycle, which means that the cooling system sustains sufficient heat rejection for 𝑇𝑎𝑚𝑏 up to around 20°C. Note that with increased 𝑇𝑎𝑚𝑏 the maximum 𝑇𝐹𝐶,𝑜𝑢𝑡 is increased.

0 10 20 30 40 50 60 70 80 90 100 110 120 130 0 1000 2000 3000 4000 5000 Te m p era tu re [ °C ] Time [s]

-20 -10 0 10 20 30 40 50 60 70 80 90 0 1000 2000 3000 4000 5000 Te m p era tu re [ °C ] Time [s]

(50)

40

Figure 41 Heat Generation from the FC stack and FC Stack Output Power for the simulated WLTP driving cycle for Model 2.

0 10 20 30 40 50 60 70 80 90 100 0 200 400 600 800 1000 1200 1400 1600 1800 Po w er / H eat Flo w Rat e [kW] Time [s]

0 10 20 30 40 50 60 70 80 90 100 0 200 400 600 800 1000 1200 1400 1600 1800 Te m p era tu re [ °C] Time [s]

(51)

41

Figure 44 Inlet and outlet coolant temperature of the FC stack for the simulated WLTP driving cycle at the ambient temperature of -15°C for Model 2.

In Figure 45 the 𝑃𝐹𝐶 and 𝑄𝐹𝐶 can be seen for the VP driving cycle. By analyzing Figure 45 it is possible to comprehend the resulting coolant temperatures. The resulting 𝑇_{𝐹𝐶,𝑖𝑛} and 𝑇_{𝐹𝐶,𝑜𝑢𝑡} for the different 𝑇_𝑎𝑚𝑏 (20°C, 45°C and -15°C) can be seen in Figure 46, Figure 47 and Figure 48. The 𝑇_{𝐹𝐶,𝑜𝑢𝑡} is maximum 99-141°C in this driving cycle, which is above the maximum acceptable 𝑇_{𝐹𝐶,𝑜𝑢𝑡} of 90°C. This means that the cooling system does not sustain sufficient heat rejection. Note that with increased 𝑇𝑎𝑚𝑏 the maximum 𝑇𝐹𝐶,𝑜𝑢𝑡 is increased significantly.

-20 -10 0 10 20 30 40 50 60 70 80 90 0 200 400 600 800 1000 1200 1400 1600 1800 Te m p era tu re [ °C] Time [s]

(52)

42

Figure 45 Heat Generation from the FC stack and FC Stack Output Power for the simulated VP driving cycle for Model 2.

Figure 46 Inlet and outlet coolant temperature of the FC stack for the simulated VP driving cycle at the ambient temperature of 20°C for Model 2. 0 10 20 30 40 50 60 70 80 90 100 110 0 200 400 600 800 1000 1200 Po w er / H eat Flo w Rat e [kW] Time [s]

(53)

43

Figure 47 Inlet and outlet coolant temperature of the FC stack for the simulated VP driving cycle at the ambient temperature of 45°C for Model 2.

Figure 48 Inlet and outlet coolant temperature of the FC stack for the simulated VP driving cycle at the ambient temperature of -15°C for Model 2. 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 0 200 400 600 800 1000 1200 Te m p era tu re [ °C] Time [s]

-20 -10 0 10 20 30 40 50 60 70 80 90 100 0 200 400 600 800 1000 1200 Te m p era tu re [ °C] Time [s]

(54)

44

5.3 Suggestion of cooling system improvements

5.3.1 Improvements for Model 1

According to 5.1.2 Transient results the cooling system in Model 1 is capable to sustain sufficient heat rejection for all driving cycles and ambient temperature conditions which were simulated according to 4.5.2 Transient Simulations. Hence, no improvements are considered necessary for Model 1.

5.3.2 Improvements for Model 2

According to 5.2.2 Transient results the cooling system in Model 2 is only capable to sustain sufficient heat rejection for the less demanding WLTP driving cycle up to 𝑇𝑎𝑚𝑏 of 20°C. In order to reject enough heat for more demanding driving cycles and be able to do this at higher 𝑇𝑎𝑚𝑏 than 20°C improvements for Model 2 are considered necessary. These improvements are implemented by changing component performance or removing components in Model 2. Three types of improvements are suggested and described below. The improvements are evaluated by simulating the most demanding driving cycle, the VP, at 20°C of 𝑇𝑎𝑚𝑏.

Increasing air mass flow rate through the radiator

The air mass flow rate through the radiator is increased with 25 and 50 % by increasing the values in the map mentioned in 4.4.1 Cooling Package (radiator, fan & condenser) and presented in Figure 4. The increase of the air mass flow rate can be seen in Figure 49. This improvement can be implemented in a vehicle by increasing the air flow intake at the grille of the vehicle or possibly by increasing fan performance or usage.

Figure 49 Air mass flow rate through the radiator for the original model and when it is increased with 25 and 50 %.

0 1 2 3 4 5 6 0 50 100 150 200 250 Air Ma ss Flow Rat e [kg/s ] Vehicle Speed [km/h]