Research on Economic and Operating Characteristics of Hydrogen Fuel Cell Cars Based on Real Vehicle Tests

Abstract

With the increase of the requirement for the economy of vehicles and the strengthening of the concept of environmental protection, the development of future vehicles will develop in the direction of high efficiency and cleanliness, and the current power system of vehicles based on traditional fossil fuels will gradually transition to hybrid power. As an essential technological direction for new energy vehicles, the development of fuel cell passenger vehicles is of great significance in reducing transportation carbon emissions, stabilizing energy supply, and maintaining the sustainable development of the automotive industry. To study the fuel economy of a passenger car with the proton exchange membrane fuel cell (PEMFC) during the operating phase, two typical PEMFC passenger cars, test vehicles A and B, were compared and analyzed. The hydrogen consumption and hydrogen emission under two operating conditions, namely the different steady-state power and the Chinese Vehicle Driving Conditions-Passenger Car cycle, were tested. The test results show the actual hydrogen consumption rates of vehicle A and vehicle B are 9.77 g/kM and 8.28 g/kM, respectively. The average hydrogen emission rates for vehicle A and vehicle B are 1.56 g/(kW·h) and 5.40 g/(kW·h), respectively. By comparing the hydrogen purge valve opening time ratio, the differences between test vehicles A and B in control strategy, hydrogen consumption, and emission rate are analyzed. This study will provide reference data for China to study the economics of the operational phase of PEMFC vehicles.

1. Introduction

The advent of the automobile has boosted economic development and improved people’s lives. The automotive industry has become a pillar industry in the world’s major industrialized countries and is one of the critical indicators of modernization. With the increase in vehicle production and ownership, emissions from traditional fossil-fuel-based vehicles have become an essential factor affecting the climate and urban environment, given the rise in the requirement for the economy of vehicles and the strengthening of the concept of environmental protection. The future automobiles will move in high efficiency and cleanliness, and the power system of traditional fossil-fuel-based vehicles will gradually transition to hybrid power [1,2]. Electric vehicles are showing an accelerated development trend as environmentally-friendly and energy-efficient vehicles with low carbon emissions and diversified energy use. They will have a profound impact on high-tech, emerging industries, and economic development. As a key technical solution of new energy vehicles, the fuel cell electric vehicle (FCEV) is significant in reducing traffic and stabilizing energy supply and sustainable development of the automotive industry.

New energy vehicles mainly consist of hybrid electric vehicles, pure electric vehicles, and FCEV. The only emission of the FCEV is water without pollutant emission, and it also has the characteristics of low noise, fast fuel replenishment, long driving range, and high energy conversion efficiency. Hence, FCEV has become a hot spot and one of the most promising development directions. However, the FCEV still faces a long way to go in terms of cost for commercialization. However, the FCEV has unparalleled performance as one kind of electric vehicle [3]. A comparison of different types of FCs is indicated in

Table 1. Comparison of the operating characteristics of different types of FCs [4].

Technical Parameters	Toyota			Honda	Hyundai
Product type	TL Power 700	First- generation fuel cells	Mirai fuel cell	Clarity fuel cell	NEXO fuel cell	ix35 fuel cell
Voltage range (V)	/	/	/	/	250–440	255–440
Number of cells	/	400	370	/	440	/
Cold start temperature (°C)	270.15 K	−30	−30	−30	−30	/
Peak power (kW)	/	114	128	100	135	/
Rated power (kW)	101	/	/	/	/	/
Power density (kW/L)	4.9	3500	5.4	3.1	3.1	/
Protection class	IP67	/	/	/	/	/
Maximum efficiency	56%	/	/	/	60%	/
Battery weight (kg)	142	108	56	/	142	/

.

Research by Demirdöven and Deutch [5] and the EU JRC [6] has shown that the performance of current FC powertrains is comparable to that of parallel hybrid powertrains and that FC powertrains have greater potential to reduce emissions and lower pollution. Not considered in these studies is the tandem hybrid drivetrain, in which the internal combustion engine is treated as a power generator only, and the electric motor drives the wheels. The tandem hybrid system can be an alternative to both the regular car and the parallel hybrid vehicle. It avoids the limited range and charging problems of electric vehicles, the expensive fuel cells, and the lack of infrastructure to refuel hydrogen vehicles. It is an intermediate stage in the move towards fully electric or fuel cell vehicles. The percentage of the subitem cost of the FC is shown in

Table 2. Percentage of the subitem cost of fuel cell [7].

Subitem	Percentage (%)
Membrane and catalyst	45
Bipolar plate	20
Accessory	19
Proton exchange membrane	10
Gas diffuse layer	6

.

Although the emission of FCEVs during driving is zero, the emission of hydrogen production and FCEV manufacturing is not able to be ignored. Martin et al. [8] found that the production of platinum catalyst and bipolar plate in the fuel cell manufacturing process had an impact on the environment. Hussain et al. [9] found that the energy consumption and greenhouse gas emissions in the life cycle of manufacturing fuel cells were about 2.3 times and 2.6 times those of traditional internal combustion engines, respectively. Sara et al. [10] have shown that the environmental impact of FCEVs is higher than that of internal combustion engine vehicles and pure electric vehicles. Zhu et al. [11] found that hydrogen production from coke oven gas in factories currently has the lowest greenhouse gas emissions and energy consumption. Other studies have also shown that the energy consumption of fuel cells comes mainly from the production of platinum and hydrogen. Meanwhile, greenhouse gas emissions primarily result from fossil energy consumption in hydrogen production [12].

The high cost of FCEV remains a significant constraint for its large-scale development. A study demonstrated that when the annual production of fuel cell systems is 500,000 units, the system manufacturing cost is $44.93/kW [13]. For the exact weight of the vehicle, the price of a FCEV and a purely electric vehicle will be $34,800 and $15,700 in 2020, respectively and will fall to $11,500 and $10,500 [14,15,16,17,18]. In terms of hydrogen cost, Wen et al. [19] proposed the most economical route to hydrogen production is onboard methanol-reforming. Kong et al. [20] concluded the cost for future FCEVs would undoubtedly be lower than that for internal combustion engine vehicles. The comparison of prices and carbon emissions for different hydrogen production methods is listed in

Table 3. Comparison of costs and carbon emissions of different hydrogen production methods.

Hydrogen Production Process	Hydrogen Production Cost (RMB/kg)	Carbon Emission Intensity (kg/kg)
Hydrogen from coal	10.67	22.00
Hydrogen from natural gas	13.44	4.80
Hydrogen from methanol	19.45	8.25
Hydrogen production from electrolytic water (thermal power)	22.9	51.50
Hydrogen production from renewable energy electrolysis	13.07	0.00

.

Fuel cell vehicles have many vital technologies, such as the basic structure of fuel cell vehicles, economic analysis, fuel cell stacks (FCS), health state diagnosis, energy management, characterization, motor control, electronic control, testing, and system optimization [21,22,23,24,25,26,27,28]. The increasing link between the on-site generation of renewable energy and electric mobility, in particular, maximizes the advantages of hydrogen as a carrier and a means of energy storage to meet hydrogen demand. The residential electricity demand through new clean microgrids will significantly leverage the techno-economic viability of renewable hydrogen generation [29,30]. The analysis of the economics of fuel cell vehicles has thus become particularly important [31]. The fuel cell industry in China is in its initial stages, and research on the life cycle evaluation of FCEVs is still immature [32,33,34,35]. This paper tests and analyses the differences in control strategies and fuel economy of two fuel cell vehicles, test vehicles A and B, to provide reference data for studying the economics of the operational phase of an FCEV, which consists of proton exchange membrane fuel cell (PEMFC) in China.

As the world’s first manufacturer to mass-produce FCEVs, test vehicle A was loaded with the first-generation fuel cells (with 114 kW power of electric reactor, 122.4 L hydrogen storage bottle capacity, 1.586 kWh power battery capacity, and 650 KM NEDC cycle) in 2014 and the second generation with higher performance in 2020 [36,37]. The power of the electric reactor is 128 kW, and it enables the second-generation test vehicle A to reach a range of 700 km on a 142.2 L hydrogen charge (World Light Vehicle Test Procedure, WLTP). Test vehicle B, to be released in 2019, has a system power of 95 kW and a power cell charge of 1.56 kWh, enabling a performance of 800 km (New European Driving Cycle, NEDC) with a range of 156.6 L of hydrogen. Hydrogen consumption per 100 km is an essential dimension in evaluating the performance of FCEVs, a reflection of the economic performance of the vehicle during operation, and one of the most critical technical priorities in the advancement of FCEV industrialization.

The hydrogen system piping arrangement at the fuel cell system level of the first-generation test vehicles A and B is analyzed in this study, particularly the hydrogen drain valve piping position. Additionally, it tests the Chinese Vehicle Driving Conditions-Passenger Car (CLTC-P) cycle, hydrogen consumption at different steady-state powers, and hydrogen emission rates from the FCS.

2. Subjects of Study

2.1. Subsection

Test vehicle A competes directly with electric cars. However, test vehicle A is not built like a conventional petrol car or a pure electric car. Test vehicle A’s powertrain is called TFSC and is a hybrid powertrain with a fuel cell at its core. It has no conventional fuel cell engine nor transmission, and inside the cabin is the control unit for the electric motor and motor. The FCS, arranged underneath the vehicle, is the heart of the system.

Test vehicle B is also an electro-hybrid in which a combined power of fuel cell and lithium battery is used. The high-voltage battery is located at the vehicle’s rear and warm up the car and recover energy. Test vehicle B also features a hydrogen storage system consisting of three identical-sized tanks with an innovative design that allows for a storage capacity of 156.6 L.

2.2. Introduction to Test Vehicles A and B’s Onboard Hydrogen Storage Systems

A hydrogen FCEV’s onboard hydrogen storage system consists of a hydrogen storage section and a hydrogen supply system. The hydrogen storage section contains hydrogen filling, storage, pressure reduction, drain protection, detection and feedback of pressure, temperature and concentration of the hydrogen storage system, and control of hydrogen delivery and shutdown. The hydrogen supply section contains pressure regulation, circulation application. It drains gas and mainly includes an injector (proportional valve), a hydrogen circulation pump, and a water-gas separator (with flush valve).

As shown in Figure 1a, test vehicle A uses a hydrogen tank with a storage pressure of 70 Mpa. The new high-pressure hydrogen storage tank enables the storage capacity of test vehicle A to reach 142 L. The high-pressure hydrogen is delivered to the power reactor through a high-pressure regulator and an ejector to produce the reduced pressure hydrogen as a power output.

As shown in Figure 1b, test vehicle B also uses a 70 Mpa hydrogen storage tank and three identical 52 L hydrogen fuel tanks placed at the rear of the car to ensure the driver’s safety in the event of a frontal collision and to store up to 156.6 L of hydrogen. In the event of an explosion, the tank is equipped with an exhaust system.

2.3. Onboard Hydrogen Storage Systems of the Test Vehicles

2.3.1. Hydrogen Supply System in Test Vehicle A

The onboard hydrogen storage system of an FCEV consists of a hydrogen storage section and a hydrogen supply system. The hydrogen storage section contains hydrogen filling, storage, pressure reduction, drain protection, detection and feedback of pressure, temperature and concentration of the hydrogen storage system, and hydrogen delivery and shutdown control. The hydrogen supply section contains pressure regulation and circulation application. It drains gas and mainly includes an injector (proportional valve), a hydrogen circulation pump, and a water-gas separator (with flushing valve). This article is concerned with the differences in the hydrogen circulation and discharge sections of the hydrogen supply system.

The piping arrangement of the hydrogen supply system is as follows: a hydrogen pressure regulator valve, three hydrogen injector valves, a hydrogen circulation pump, and a drain and discharge valve. The primary function of each part is as follows: the hydrogen pressure regulator valve reduces the hydrogen tank pressure from 70 MPa to 1–1.6 MPa to meet the demand of the FCS for hydrogen pressure. At the upstream end of the hydrogen circulation circuit, three hydrogen injectors are designed to control the hydrogen pressure and flow into the stack. Besides, the hydrogen circulation pump is designed to improve the hydrogen utilization and the hydrogen flow rate in the anode runners and to prevent water accumulation and re-circulate the hydrogen at the outlet of the stack anode (that is not involved in the internal electrochemical reaction) to the stack. The inverter of the hydrogen circulation pump and the inverter of the water pump are integrated into one unit, so that cooling water channels and hydrogen supply can be carried out simultaneously. A drain valve is used to discharge liquid water and nitrogen from the anode channel. The hydrogen venting valves are designed as a single unit to save on component layout but at the same time increase the difficulty of machining and controlling the components. The control strategy for the hydrogen drain valve needs to be calibrated to take into account the variation in fuel cell voltage output and the effect of the water generation from the fuel cell system, which is an integration of a coupled control system for both parameters. Test vehicle A hydrogen system piping is shown in the red line in Figure 2.

Due to the use of the hydrogen circulation pump, the parasitic power is also increased. Based on the steady-state power test data, the maximum power of the hydrogen pump is 450 W when the reactor power is 70 kW (this is the average value of the steady-state power, the trend of the hydrogen pump at each steady-state power-point is shown in Figure 3). To prevent flooding of the reactor, the hydrogen circulation pump and the hydrogen drain valve are positioned underneath the reactor to allow for the timely extraction and discharge of excess water. The hydrogen vent valve is for pulse venting, and its primary function is to remove the water and nitrogen gas collected on the anode side. At the moment of pulse venting, the pressure on the anode side will decrease. The hydrogen pump will increase its power appropriately to boost the circulating gas pressure, thus effectively circumventing flooding.

2.3.2. Hydrogen Supply System in Test Vehicle B

The hydrogen system piping of test vehicle B’s hydrogen fuel cell system consists of a hydrogen shut-off valve, a supply valve, an injection valve, an ejector, a hydrogen drains valve, a dehydrator, and a drain valve. Test vehicle B’s hydrogen system piping is indicated in Figure 4.

In contrast, to test vehicle A, test vehicle B replaces the hydrogen circulation pump injector approach of the previous generation of fuel cell vehicles by eliminating the hydrogen circulation pump and adopting the injector model. Compared to the hydrogen circulation pump, this device is an easy-to-fix component with no parasitic power, ideal for fuel cell hydrogen recycling. However, due to the structural limitations of the ejector, there is only one optimum operating point for a fixed size ejector. As a result, the fuel cell system only injects skillfully in specific ranges but performs poorly when running the fuel cell system at low power.

The hydrogen drain valve and the drain valve are designed independently, making machining less complicated than the test vehicle A model. The piping arrangement of the hydrogen drain valve and the drain valve are both at the front of the airline backpressure valve. Changes significantly influence the exhaust volume of the hydrogen drain valve in airline pressure, so the hydrogen drain valve opening strategy may be more complex than the MIRIA.

3. Test Method

3.1. Test Setup

In this article, two tests are carried out on test vehicles A and B, two benchmark vehicles, respectively. One is an FCS steady-state power test, and the other is a dynamic economy assessment of the entire vehicle. The test device and process are demonstrated in Figure 5 and Figure 6, respectively.

The FCS steady-state power test is used to evaluate the economy of both vehicles at single FCS steady-state power. The main parameters monitored and calculated are as follows: FCS power, FCS voltage, fuel stack current, hydrogen tank temperature, hydrogen tank pressure, hydrogen flow consumption, hydrogen flow consumption per unit power, hydrogen emission rate, hydrogen discharge valve opening time percentage at steady-state power, and steady-state power collection time. The test points are selected concerning the national standard GB/T 24554-2009 fuel engine performance test method. Vehicle A is only tested for the 30 s at 70 kW and 80 kW due to excessive heat dissipation.

The specific test procedure for steady-state power is as follows:

(1)

Mount the vehicle on the drum stand;

(2)

Pull up the vehicle speed to 60 km/h, warm up the engine for half an hour, and observe the stable water temperature at the outlet of the electric stack (fluctuation ±0.5 °C);

(3)

Stop the vehicle, switch the chassis dynamometer to constant speed mode within 3 s, and set the cooling fan airflow to the maximum;

(4)

Pull the vehicle speed up to 80 km/h via chassis dynamometer;

(5)

Depress the pedal and observe the FCS output power. After stabilizing at the set power ±0.5 kW range, start the timer for 5 min. After the timer is finished, switch to the following working condition and continue the test until the test is complete;

(6)

Observe the FCS outlet water temperature during the test. When the FCS outlet temperature exceeds 80 °C, the FCS power should be immediately reduced to 10 kW for cooling until the FCS outlet water temperature drops to 60 °C, and then continue to complete the test.

The main parameters to be monitored and calculated are FCS power, FCS voltage, FCS current, hydrogen tank temperature, hydrogen tank pressure, hydrogen consumption per unit mileage, and the percentage of time the hydrogen drain valve is open during a CLTC-P cycle. The test procedure is partly based on the industry-standard “Test Method for Fuel Cell Vehicle Energy Consumption and Range”, which is currently under development. The proposed method gives six CLTC-P cycle conditions. At the same time, test vehicle B cannot be connected to the laboratory’s hydrogen flow measurement system on the test rig. Thirteen CLTC-P cycle conditions are tested for both models to accumulate test samples and fully observe more comprehensive test results. In the CLTC-P cycling condition, the test method requires adjusting the power cell state of charge (SOC) value; the power cell SOC is not monitored during this test.

The typical dynamic economy test method is specified below:

(1)

Fix the vehicle to the chassis dynamometer, and immerse the car for 30 min;

(2)

Warm up the vehicle and drive the car on the chassis dynamometer for one completed CLTC-P cycle, after which switch off the car and preset the car for 15 min;

(3)

Start recording the CAN signals of the whole vehicle in a frequency of 10 Hz;

(4)

Carry out a continuous cycle work test according to the CLTC-P work table, during which parameters such as hydrogen consumption and chassis dynamometer mileage are recorded for each cycle;

(5)

Stop after the completion of 13 cycles of testing;

(6)

Stop data logging and analyze the data.

3.2. The Steady-State Test

3.2.1. GB/T24554-2009

GB/T24554-2009 “fuel cell engine performance test method” specifies the fuel cell engine starting characteristics test, steady-state characteristics, dynamic response characteristics, gas tightness testing, insulation resistance testing, and other experimental methods applicable to automotive PEMFC engines.

The cold engine state refers to the internal temperature of the fuel cell engine (coolant outlet temperature). It is the same as the ambient temperature, and the hot engine temperature is the engine’s internal temperature in the normal operating temperature range (the manufacturer specifies average operating temperature). The measurement instrument requirements are shown in

Table 4. Instrument accuracy requirements for tests.

Name	Specified Accuracy	Note
Voltage Sensor	≤0.5% FS	/
Current sensor	≤0.5% FS	/
Thermometer	±1 C	/
Humidity meter	±3%	/
Hydrogen flow meter	≤1%	According to the relative error
Coolant flow meter	≤1% FS	/
Weighing scales	≤0.5% FS	/

.

The performance test includes cold start characteristic test, hot start characteristic test, rated power test, dynamic response test, loaded dynamic response test, unloaded dynamic test, steady-state characteristic test, emergency stop function side face fuel cell engine airtightness test, insulation resistance test, and quality test. Among them, the steady-state characteristics test method is as follows:

Before the steady-state characteristics test, the fuel cell engine is in the warm-up state, and the test process should be carried out automatically without manual intervention. At least ten operating points are selected for uniformity in the fuel cell engine’s working range. After a warm-up, the engine is returned to idle speed for 10 s, loaded to the pre-determined operating points according to the specified loading method. The engine runs steadily for at least 3 min at none of the operational matters.

The measured data include fuel cell voltages and currents, hydrogen consumption, and voltage and current of the auxiliary system. The polarization characteristic curve (V-I), power curve, efficiency curve of the FCS, the power curve and efficiency curve of the fuel cell engine, the power curve of the auxiliary system, etc., can be obtained.

3.2.2. GB/T35178-2017

GB/T35178-2017, “Fuel Cell Electric Vehicle Hydrogen Consumption Measurement Method” specifies the measurement method of hydrogen consumption of FCEV, and this standard applies to FCEV, using compressed hydrogen. The units of relevant measurement parameters and the accuracy should be consistent with the provisions of

Table 5. Units and accuracy of pertinent measurement parameters.

Parameters	Unit	Accuracy
Temperature	K	±1
Gas pressure	Mpa	±1%
Mass	g	±0.5
Volumetric flow rate	L/s	±1%
Mass flow rate	g/s	±1%

.

(1)

The measurement method is as follows:

Vehicle condition: the test vehicle should be broken according to the manufacturer’s specifications. The break-in mileage should not be less than 1000 km. It is recommended to drive at least 300 km within seven days before the experiment. External hydrogen supply is used during the investigation, which is cut off in the fuel supply line.

Fuel is supplied from outside the vehicle through a bypass line installed in the fuel cell system’s fuel line at the manufacturer’s recommended pressure. A flow meter is installed in the supply line between the fuel cell and the outside supply source. The flow meter can be either a meter for mass flow, volumetric flow, or volumetric flow (vehicle A uses a mass flow meter, vehicle B cannot connect to a flow meter and uses the pressure-temperature method). The flow meter and bypass line should be installed reliably to prevent leakage, release, or air entry due to vibration. The flow meter is used to measure the volume or mass of oxygen consumed by the off-vehicle supply source.

(2)

The test procedure for flow method measurements is as follows:

Calculate the hydrogen consumption (by mass) by using the measured mass flow rate as in Equation (1):

$$\omega ={\displaystyle \underset{0}{\overset{\mathrm{t}}{\int }}{Q}_m}\mathrm{dt}$$

(1)

In Equation (1):

$\omega$ is the hydrogen consumption in grams (g) during the measurement time;

$Q_m$ is the gas mass flow rate in grams per second (g/s) during the test.

(3)

The test procedure for the pressure-temperature measurement method is as follows:

Before the test starts, the test hydrogen storage tank’s gas pressure and gas temperature are tested. After the test, the test hydrogen storage tank’s gas pressure and gas temperature should be tested. The gas pressure and temperature measured before and after the experiment are substituted into Equation (2) to calculate the hydrogen consumption:

$$\omega =m\times \frac{V}{R}\times \left(\frac{P_1}{Z_1\times T_1}-\frac{P_2}{Z_2\times T_2}\right)$$

(2)

In Equation (2):

ω is the amount of hydrogen consumed in grams (g) during the measurement time;

m is the molar mass of hydrogen molecules (2.016) in grams without moles (g/mol);

V is the total volume of the high-pressure part of the fuel tank and the Fujian (pressure reducing valve, piping, etc.), in liters (L);

R is the common gas constant, =0.0083145 [Mpa·L/(mol·K)];

P₁ is the number of gas molecules in the tube at the beginning of the test, pressure in megapascals (MPa);

P₂ is the number of gas molecules in the tube at the end of the test, pressure in megapascals (MPa);

T₁ is the number of gas molecules in the tube at the beginning of the test and the temperature in Kelvin (K);

T₂ is the number of gas molecules in the tube at the end of the test and the temperature in Kelvin (K);

Z₁ is the hydrogen compression factor at P₁, T₁;

Z₂ is the hydrogen compression factor at P₂, T₂.

3.3. The CLTC-P Cycle Test

CLTC-P test condition is the passenger car part of CATC (China Automotive Test Cycle), based on 41 cities, 3832 vehicles, accumulated 32.78 million kilometers, and 2 billion GIS traffic big data to define the standard working condition. Compared with NEDC, CLTC-P removes the part of super high-speed driving which does not match the actual working condition and redivides the traffic into 3-speed intervals, corresponding to low speed, medium speed, and high speed, with a total working time of 1800 s, of which the proportion of low-speed interval time is 37.4%, medium speed interval time is 38.5%, and high-speed interval time is 24.1%. The average speed, the maximum speed, and the idling speed are 29.0 km/h, 114.0 km/h, and 22.1%, respectively. CLTC-P reflects more realistic working conditions with Chinese characteristics, including a more reasonable definition of average speed and maximum speed, broader driving conditions, a more reasonable proportion of stopping modes, and more great dynamic acceleration and deceleration conditions. CLTC-P includes three speed intervals: low speed (Part 1), medium speed (Part 2), and high speed (Part 3), with a total working time of 1800 s. The working curves are shown in Figure 7. Furthermore, the statistical characteristics of the operating curves are shown in

Table 6. Statistical characteristics of CLTC-P working condition curve.

Features	Overall	Part 1	Part 2	Part 3
Running time/s	1800	674	693	433
Mileage/km	14.48	2.45	5.91	6.12
Maximum speed/(km/h)	114.00	48.10	71.20	144.00
Maximum acceleration/(m/s2)	1.47	1.47	1.44	1.06
Maximum deceleration/(m/s2)	−1.47	−1.42	−1.47	−1.46
Average speed/(km/h)	28.96	13.09	30.68	50.90
Average running speed/(km/h)	37.18	20.20	38.24	53.89
Average acceleration/(m/s2)	0.45	0.42	0.46	0.46
Average deceleration/(m/s2)	−0.49	−0.45	−0.50	−0.54
Relative positive acceleration/(m/s2)	0.17	0.14	0.16	0.18
Acceleration ratio/%	28.78	22.55	30.45	35.80
Deceleration ratio/%	26.44	21.51	28.43	30.95
Uniformity ratio/%	22.67	20.77	21.36	27.71
Idle speed ratio/%	22.11	35.16	19.77	5.54

.

4. Results and Discussion

Referring to the test method of GB/T24554-2009 “Fuel Cell Engine Performance Test Method”, the output power was selected for both models and tested at 10 kW, 20 kW, 30 kW, 40 kW, 50 kW, 60 kW, 70 kW, and 80 kW, respectively. During the testing, the actual hydrogen flow consumption, the percentage of time the hydrogen discharge valve was open, and the change of hydrogen emission rate was observed at steady-state power. Since the steady-state power control of the fuel cell reactors fluctuates within a range, the actual output power of test vehicles A and B reactors is not strictly the set power at the operating point. So, to increase the comparability of the data, the hydrogen consumption per unit of energy is compared below.

Referring to the three hydrogen measurement methods given in GB/T35178-2017 “Fuel Cell Electric Vehicle Hydrogen Consumption Measurement Methods”, test vehicle A can use easier flow method to calculate the actual hydrogen consumption due to the bypass line installed in the fuel cell system. The value is measured by the mass flow meter installed in the supply line. The actual hydrogen consumption of the vehicle can only be calculated by using the temperature and pressure of the hydrogen storage tank before and after the test combined with the total volume of the high-pressure part of the onboard storage system due to the inconvenience of dismantling the bypass line of the vehicle. For test vehicle B, it should be noted that this method is different from GB/T35178-2017 “Fuel Cell Electric Vehicle Hydrogen Consumption Measurement Method”. It should be noted that this method is not strictly consistent with the external hydrogen storage tank required in the standard Pressure-Temperature Method.

4.1. Comparations of Hydrogen Consumption and Utilization under Steady-State Conditions

4.1.1. Comparative Analysis of Hydrogen Consumption

The actual hydrogen flow consumption of the fuel cell reactors of test vehicles A and B at different power steady-state conditions are shown in

Table 7. Actual hydrogen flow consumption.

Set Power (kW)	Test Vehicle A			Test Vehicle B
	Actual Power (kW)	Actual Hydrogen Mass Flow Rate (kg/h)	Actual Hydrogen Flow Rate per Unit Power [kg/(kW·h)]	Actual Power (kW)	Actual Hydrogen Flow Rate (kg/h)	Actual Hydrogen Flow Rate per Unit Power [kg/(kW·h)]
10	10.3472	0.4821	0.0466	10.1839	0.4579	0.0450
20	20.4949	1.0127	0.0494	20.0545	0.9444	0.0471
30	30.1725	1.5160	0.0502	30.2895	1.4900	0.0492
40	40.5500	2.0549	0.0507	40.8430	2.0707	0.0507
50	50.6707	2.5871	0.0511	51.2121	2.6524	0.0518
60	60.2494	3.2075	0.0532	60.4346	3.4451	0.0570
70	70.1754	4.0295	0.0574	70.7718	3.7589	0.0531
80	80.0496	4.4795	0.0560	80.7534	4.4532	0.0551

. The hydrogen flow rate trend per unit power at steady-state power for test vehicles A and B are indicated in Figure 8.

Figure 8 shows that the hydrogen consumption per unit power of test vehicle A is higher than that of test vehicle B when the reactor output is below 40 kW, higher than that of test vehicle A when the reactor output is between 40 kW and 60 kW, and higher than that of test vehicle B when the reactor output is between 70 kW and 80 kW. It is concluded that the fuel economy of test vehicle A is slightly lower than that of test vehicle B in steady-state at low power output (10 kW to 30 kW). The fuel economy of test vehicles A and B in steady-state operation at power output greater than 40 kW (50 kW to 80 kW) is better and worse.

4.1.2. Comparison of Hydrogen Emission Rates at Steady-State Power

The actual hydrogen emission rates for the fuel cell reactors of test vehicles A and B models at different power steady-state operating conditions are compared in

Table 8. Actual hydrogen emission rate.

Set Power (kW)	Actual Power (kW)		Actual Hydrogen Flow Rate (kg/h)		Theoretical Hydrogen Flow Rate (kg/h)		Hydrogen Emissions (kg/h)		Hydrogen Emission Rate [g/(kW·h)]
	Vehicle A	Vehicle B	Vehicle A	Vehicle B	Vehicle A	Vehicle B	Vehicle A	Vehicle B	Vehicle A	Vehicle B
10	10.3472	10.1839	0.4821	0.4579	0.4742	0.4575	0.0079	0.00036	0.7606	0.0003
20	20.4949	20.0545	1.0127	0.9444	0.9935	0.9425	0.0192	0.00192	0.9449	0.0016
30	30.1725	30.2895	1.5160	1.4900	1.5100	1.4828	0.0060	0.00712	0.1843	0.0057
40	40.5500	40.8430	2.0549	2.0707	2.0517	2.0536	0.0032	0.01709	0.1401	0.0182
50	50.6707	51.2121	2.5871	2.6524	2.5795	2.6307	0.0076	0.02167	0.1325	0.0226
60	60.2494	60.4346	3.2075	3.4451	3.2001	3.1572	0.0074	0.28790	0.1231	0.3944
70	70.1754	70.7718	4.0295	3.7589	4.0165	3.7448	0.0130	0.01409	0.1845	0.0235
80	80.0496	80.7534	4.4795	4.4532	4.4585	4.3253	0.0210	0.12784	0.5413	0.2465

. The trend of hydrogen emission rate per unit power for test vehicles A and B at steady-state power depending on different power is shown in Figure 9.

It can be seen from Figure 9 that the hydrogen emission rate of test vehicle B is better than that of test vehicle A at low power output (10 kW to 50 kW). At higher power output (60 kW to 80 kW), the hydrogen emission rate of test vehicle A is better than that of test vehicle B at 60 kW output. In contrast, at 70 kW and 80 kW output, the hydrogen emission rate of test vehicle B is better than that of test vehicle A. It can be concluded that the hydrogen emission rate of test vehicle B is better than that of test vehicle A at steady-state power conditions.

4.1.3. Subsubsection

The comparison of the actual hydrogen discharge valve opening time ratio between test vehicles A and B fuel cell reactors under different power steady-state operating conditions is shown in

Table 9. Actual hydrogen discharge valve opening time as a percentage.

Set Power (kW)	Test Vehicle A		Test Vehicle B
	Actual Power (kW)	The Ratio of the Opening Time of the Hydrogen Discharge Valve	Actual Power (kW)	The Ratio of the Opening Time of the Hydrogen Discharge Valve
10	10.3472	0.0215	10.1839	0.0054
20	20.4949	0.0116	20.0545	0.0067
30	30.1725	0.0583	30.2895	0.0073
40	40.5500	0.1381	40.8430	0.0080
50	50.6707	0.2515	51.2121	0.0084
60	60.2494	0.3632	60.4346	0.0088
70	70.1754	0.5773	70.7718	0.0097
80	80.0496	0.6218	80.7534	0.0150

.

The trend of the percentage of hydrogen discharge valve opening time per unit power for test vehicles A and B at steady-state power according to different power is shown in Figure 10:

In terms of the opening timeshare of the hydrogen drain valve for steady-state power, the data on the opening timeshare of the hydrogen drain valve differs significantly between test vehicles A and B due to the difference in drain valve construction (vehicle A has an integrated drain function, while vehicle B only has a hydrogen drain function). The percentage of time that the hydrogen drain valve is open increases gradually with the output power of the reactor for both models. For test vehicle B, the rate of time that the hydrogen drain valve is available increases slowly when the output power of the reactor is less than 70 kW and increases rapidly when the output power of the reactor is greater than 70 kW. For test vehicle A, the percentage of time the hydrogen drain valve is open changes slowly when the output power is 10 kW and 20 kW and changes slowly when the output power is greater than 20 kW. After the power output is more significant than 20 kW, the percentage of time the hydrogen drain valve is open may be affected by the amount of water generated and proliferates.

4.2. Comparisons of Operating Characteristics under CLTC-P Cycle Conditions

The range of test vehicles A and B is based on the NEDC and WLTP conditions, respectively, and the content of test vehicle B is based on the NEDC conditions. The range of test vehicles A and B are tested for the first time in China under CLTC-P conditions. 13 CLTC-P cycle tests were carried out on the two models to fully accumulate test samples and observe more comprehensive test results. Consider that test car B cannot connect to the hydrogen flow measurement system on the laboratory test stand. The actual hydrogen emissions were calculated from the actual hydrogen consumption of the vehicle during each CLTC-P cycle and the theoretical hydrogen consumption measured by the mass flowmeter on the bypass line of test vehicle A and the actual hydrogen consumption of test vehicle B. The actual hydrogen emission was calculated by the temperature and pressure changes before and after the vehicle hydrogen storage bottle test combined with the volume of the hydrogen storage bottle. Theoretical hydrogen consumption is calculated according to the number of single cells in the stack and the instantaneous current of the stack. The actual hydrogen emission is obtained by subtracting the theoretical hydrogen consumption from the actual hydrogen consumption. The actual hydrogen emission rate is the ratio of the actual hydrogen emission to time.

4.2.1. Analysis of the Operating Characteristics of Test Vehicle A under Cyclic Conditions

The operating parameters of test vehicle A in 13 CLTC-P cycles are listed in

Table 10. Operating parameters of test vehicle A in 13 CLTC-P cycles.

Cycle Number	Cycle Mileage (kM)	Actual Hydrogen Consumption (g)	Actual Hydrogen Consumption per Unit Mileage (g/kM)	Theoretical Hydrogen Consumption (g)	Theoretical Hydrogen Consumption per Unit Mileage (g/kM)	Hydrogen Emission Rate (g/(kW·h))	Percentage of Time Hydrogen Discharge Valve Is Open
Average value	14.4	141.1147	9.7723	136.7344	9.4700	1.556	2.79%
1	14.6	141.6526	9.7022	138.5345	9.4887	1.106	2.88%
2	14.4	140.7565	9.7748	135.1728	9.387	1.988	2.67%
3	14.5	142.3718	9.8187	138.5345	9.5541	1.381	2.81%
4	14.5	142.9743	9.8603	138.3585	9.542	1.628	2.76%
5	14.4	146.4122	10.1667	142.3807	9.8876	1.376	2.89%
6	14.6	144.7451	9.911	140.7070	9.6375	1.382	2.53%
7	14.3	143.6235	10.042	139.7469	9.7725	1.351	2.68%
8	14.4	134.8275	9.3611	130.6854	9.0754	1.525	2.63%
9	14.4	138.3347	9.6042	133.9542	9.3024	1.578	2.77%
10	14.4	137.3234	9.5347	132.5717	9.2064	1.735	2.77%
11	14.4	146.1787	10.1458	141.3862	9.8185	1.621	3.11%
12	14.4	137.8389	9.5694	132.6882	9.2145	1.868	2.82%
13	14.4	137.4522	9.5451	132.8261	9.2240	1.694	2.92%

. In the 13 CLTC-P cycles, the distribution of mileage, actual hydrogen consumption, actual hydrogen consumption per unit mile, theoretical hydrogen consumption, theoretical hydrogen consumption per unit mile, hydrogen emission rate, and the percentage of hydrogen discharge valve opening time for each cycle are demonstrated.

For the 13 CLTC-P tests conducted by test vehicle A, the average driving range was 14.44 kM, the highest driving range was 14.6 kM (cycle 6), and the lowest driving range was 14.4 kM (cycle 7). The average actual hydrogen consumption was 141.093 g, the highest was 146.4 g (cycle 5), and the lowest was 134.8 g (cycle 8). In terms of actual hydrogen consumption per unit mile, the average hydrogen consumption per unit mile was 9.7723 g/kM, the highest value was 10.1667 g/kM (cycle 5), and the lowest value was 9.3611 g/kM (cycle 8). In terms of hydrogen emission rate, the average hydrogen emission rate was 1.556 g/(kW·h), the highest value of hydrogen emission rate was 1.988 g/(kW·h) (cycle 3), and the lowest value of hydrogen emission rate was 1.106 g/(kW·h) (cycle 2). Regarding the hydrogen vent opening time percentage, the average value of the hydrogen vent opening time percentage was 2.79%, the highest value was 3.11% (cycle 11), and the lowest value was 2.53% (cycle 6).

4.2.2. Analysis of Test Vehicle B Operating Characteristics under Cyclic Conditions

The mileage distribution, actual hydrogen consumption, actual hydrogen consumption per unit mile, theoretical hydrogen consumption, theoretical hydrogen consumption per unit mile, hydrogen emission rate, and the percentage of hydrogen discharge valve opening time for each of the 13 CLTC-P cycles are shown in

Table 11. Operating parameters of test vehicle B in 13 CLTC-P cycles.

Cycle Number	Cycle Mileage (kM)	Actual Hydrogen Consumption (g)	Actual Hydrogen Consumption per Unit Mileage (g/kM)	Theoretical Hydrogen Consumption (g)	Theoretical Hydrogen Consumption per Unit Mileage (g/kM)	Hydrogen Emission Rate (g/(kW·h))	Percentage of Time Hydrogen Discharge Valve Is Open
Average value	14.4	119.0038	8.2772	106.7597	7.4270	5.3967	0.68%
1	14.5	118.7670	8.1908	118.0480	8.1412	0.2762	0.81%
2	14.4	135.7284	9.4256	112.2915	7.798	9.4737	0.66%
3	14.3	137.3309	9.6036	104.4185	7.302	14.2604	0.60%
4	14.3	91.6100	6.4063	107.5723	7.5225	−6.7434	0.64%
5	14.4	122.9383	8.5374	112.5401	7.8153	4.198	0.66%
6	14.3	115.3396	8.0657	109.5605	7.6616	2.4043	0.68%
7	14.5	121.1332	8.354	98.4284	6.7882	10.4742	1.20%
8	14.4	118.0258	8.1962	103.2147	7.1677	6.5477	0.59%
9	14.4	95.7630	6.6502	107.7721	7.4842	−5.0496	0.58%
10	14.2	124.8953	8.7954	109.5134	7.7122	6.378	0.58%
11	14.5	113.0234	7.79472	99.2022	6.8415	6.3453	0.56%
12	14.5	135.6756	9.3569	97.2594	6.7075	17.906	0.58%
13	14.2	116.8191	8.2267	108.0544	7.6095	3.6866	0.70%
14	186.9	1476.6712	7.9009	1387.8756	7.4257	2.9063	0.68%

(Note: Test vehicle B cannot be connected to a hydrogen flow measurement system, so the onboard hydrogen pressure and temperature can only calculate the actual hydrogen consumption. (To eliminate the influence of pressure sensor accuracy to the greatest extent, we calculate the total hydrogen consumption and hydrogen consumption per unit mile for 13 CLTC-P cycles by recording the cylinder pressure and temperature at the beginning of the first CLTC-P cycle and the cylinder temperature and pressure at the end of 13 CLTC-P cycles).

.

In the 13 CLTC-P tests conducted by test vehicle B, the authors used two methods to calculate hydrogen consumption per unit mile, theoretical hydrogen consumption, theoretical hydrogen consumption per unit mile, and hydrogen emission rate. The first method calculated the above parameters using the pressure and temperature at the beginning/ending of each cycle. The average value of actual hydrogen consumption was calculated as 119.0038 g. The average hydrogen consumption per unit mile was 8.2772 g/kM. The average hydrogen emission rate was 5.3967 g/kW·h, and the available hydrogen discharge valve percentage was 0.68%. In the second method, by recording the starting temperature and pressure of the first cycle of 13 CLTC-P cycles and the temperature and pressure at the end of the thirteenth cycle, the total mileage driven in 13 CLTC-P cycles was calculated as 186.9 kM, the total hydrogen consumption was 1476.6712 g, and the hydrogen consumption per unit mileage was 7.9009 g/kM. This is the total theoretical hydrogen consumption. The authors believe that the hydrogen consumption per unit mileage from the second calculation method is closer to the actual value, so the second algorithm is also used for the 100 km hydrogen consumption mentioned below.

4.2.3. Comparative Analysis of the Operating Characteristics of the Two Tested Vehicles

The test results show that vehicle A has a lower hydrogen emission rate while vehicle B has a lower hydrogen consumption rate. The actual hydrogen consumptions rates of vehicles A and B are 9.77 g/kM and 8.28 g/kM, respectively. The average hydrogen emission rates for vehicles A and B are 1.56 g/(kW·h) and 2.91 g/(kW·h) (since the error in the first method is too large, the conclusion of the second method is quoted), respectively, as shown in Figure 11. Besides, the opening frequency of test vehicle A hydrogen drain valve is higher than that of test vehicle B, as the opening frequencies of the hydrogen drain valve for vehicle A and vehicle B are 2.79% and 0.68%, respectively.

The operating characteristics comparison of different vehicles is listed in

Table 12. Comparison of the operating characteristics of different electric vehicles.

	Honda CLARITY	Toyota Mirai I	Toyota Mirai II	Hyundai NEXO	BYD HAN	GAC AION S
Fuel cell power (kW)	103	114	128	95	/	/
Total power (kW)	130	114	134	135	363	162
Torque (NM)	300	335	300	395	680	350
Acceleration (s) (0–100 km/h)	8.8	10	/	9.7	3.9	6.8
Range	589 (EPA)	502 (EPA)	850 (WLTC)	754 (NEDC)	505 (NEDC)	602 (NEDC)
Hydrogen storage	141 L	4.6 kg	5.6 kg	6.33 kg	/	/

. FCEVs have fewer performance advantages but a more extended driving range than the current mainstream FCEVs and pure electric vehicles.

5. Conclusions

Through the analysis of hydrogen flow consumption at steady-state power, hydrogen flow consumption per unit power, hydrogen emission rate, and the percentage of time the hydrogen drain valve is open, the overall performance parameters of test vehicle B are better than those of test vehicle A. Test vehicle B’s hydrogen emission rate is higher than that of test vehicle A. However, test vehicle B’s hydrogen consumption per 100 km is still lower than that of test vehicle A, mainly due to the influence of the reactor performance, the hydrogen discharge valve opening strategy, and the performance of the brake energy recovery function.

Both test vehicles A and B consume similar amounts of hydrogen at the same steady-state power, with test vehicle B outperforming test vehicle A in terms of fuel economy in steady-state conditions at low power output (10 kW to 30 kW) and in steady-state conditions at power outputs more significant than 40 kW (50 kW to 80 kW). The hydrogen emission rate of test vehicle B is better than that of test vehicle A at steady-state power. In contrast, the actual hydrogen consumption of test vehicle B is approximately 7.90 g/km in the CLTC-P cycle and 9.77.23 g/km in test vehicle A, which is a significant difference.

In the 13 CLTC-P tests, the hydrogen consumption of vehicle B was erratic and jumped between individual cycles. In contrast, the hydrogen consumption of vehicle A was stable over a given cycle, with more minor variations between cycles. As vehicle A uses a hydrogen flow meter as the measurement method for hydrogen consumption, while vehicle B uses the temperature-pressure method for calculations, it can be concluded that the measurement method for hydrogen consumption has a significant influence on the experimental results. At the same time, the temperature-pressure method of the hydrogen consumption algorithm still has some shortcomings.