Fuel 348 (2023) 128223

Contents lists available at ScienceDirect

Fuel
journal homepage: www.elsevier.com/locate/fuel

Full Length Article

A tripartite evolutionary game model for the hydrogen fuel cell vehicle
industry development under government regulation in China
Wei Wang *, Jinzhong Li
School of Economics and Management, North China Electric Power University, Beijing 102206, PR China

A R T I C L E I N F O A B S T R A C T

Keywords: The development of the hydrogen fuel cell vehicle (HFCV) industry constitutes a necessary advancement in the
Hydrogen fuel cell vehicles transformation and upgrading of the automotive industry as a whole and is also an important strategic route to
Evolutionary game model help China achieve the objectives of carbon neutrality. How to effectively promote the production of HFCV
Evolutionary stable strategy
enterprises and the purchase of HFCVs by consumers has become a critical problem that needs to be solved.
Government regulation
Taking into consideration the limited rationality of the government, enterprises and consumers, we applied a
tripartite evolutionary game theory (EGT) model to analyze the impact of government regulation on the
development of the HFCV industry. The simulation results show that: (1) medium-intensity supervision can
promote robust and regulated HFCV industry development; (2) the higher the penalty, the more likely enterprises
are to transition to HFCV production; (3) the larger the research and development (R&D) subsidy, the more
beneficial it is for enterprises to overcome technical obstacles; (4) higher investment in infrastructure stimulates
consumers to purchase HFCVs. Based on these results, some recommendations have provided to support the
sustainable development of HFCV industry in China.

1. Introduction increased ownership of HFCVs [23,24]. Nowadays, the United States

(US), Japan, the European Union (EU), and China are taking action to
The increasing population and demand of the world have led to an encourage HFCV industry development.
overall rise in anthropogenic greenhouse gas emissions and extreme One of the first countries to accept hydrogen and fuel cell technology
climate change, which have caused natural disasters and shortages in as part of its national policy was the US [25]. In 2019, the Fuel Cell and
food and water in recent years [1]. As concerns about the effects of Hydrogen Energy Association issued the ROAD MAP TO A US
climate change grow, so do efforts to transition to a low-carbon energy HYDROGEN ECONOMY, which set a goal of promoting 200,000 and 5.3
system [2]. The application of new energy is now a compelling way to million HFCVs by 2025 and 2030, respectively [26]. Japan has adopted
tackle climate change and support a low-carbon energy transition [3–9]. hydrogen energy as a key strategy for advancing its energy transition. In
As a clean, abundant, low-carbon energy source with wide applicability, 2014, the vision of building a “hydrogen society” was proposed
hydrogen energy plays an important role in the low-carbon energy [25,27,28], and the Japanese administration began to subsidize the
transition [10–13]. Hydrogen energy is used extensively in trans­ construction and operation of hydrogen refueling stations and HFCVs.
portation, construction, and electricity, and one major implementation For instance, the subsidy for hydrogen fuel cell buses is half that of the
of hydrogen energy in transport is hydrogen fuel cell vehicles (HFCVs) bare vehicle price, the subsidy for cabs is a third of the bare vehicle
[14–17]. With their zero emissions, high-energy efficiency, and recy­ price; and the maximum subsidy for household passenger cars is 2.25
clable energy resources [18–21], HFCVs can effectively reduce trans­ million yen per vehicle [27,29]. The EU views hydrogen energy as the
portation energy consumption [22]. If HFCVs are commercially primary path toward energy security and transition [25]. The European
produced and marketed, they will contribute considerably to achieving Hydrogen Energy Roadmap: A sustainable Pathway for Europe’s Energy
carbon neutrality. The environment for human life will improve through Transition released in 2019, set the goal of promoting 4.24 million

Abbreviations: HFCV, Hydrogen fuel cell vehicle; ESS, Evolutionary stable strategies; EGT, Evolutionary game theory; NEV, New energy vehicle; R&D, Research
and development; EV, Electric vehicle; US, United States; HFC, Hydrogen fuel cell; EU, European Union; FCV, Fuel cell vehicle; FV, Fuel vehicle.
* Corresponding author.
E-mail address: [email protected] (W. Wang).

https://doi.org/10.1016/j.fuel.2023.128223
Received 3 January 2023; Received in revised form 13 March 2023; Accepted 23 March 2023
Available online 13 May 2023
0016-2361/© 2023 Elsevier Ltd. All rights reserved.
W. Wang and J. Li Fuel 348 (2023) 128223

HFCVs and building 3,000 hydrogen refueling stations by 2030 [30]. HFCVs is increasing in China, but it is always in the primary stage
China is the largest consumer and producer of hydrogen worldwide compared with developed countries, and it is behind other major
and has the largest hydrogen production capacity [31]. At present, the countries in developing HFCVs in the world at the technical and com­
hydrogen production potential in China is approximately 40 million tons mercial levels. Although the government has released various policies
per year, and the production capacity is approximately 37 million tons from multiple perspectives to promote the growth of HFCVs, such as
per year (shown in Fig. 1) [32]. The cumulative domestic production industry regulations, promotions and applications, tax benefits, tech­
and distribution of fuel cell vehicles (FCVs) reached 13,155 and 12,722 nical innovation, and infrastructure development, the development of
units, respectively, in the 2016–2022 period (shown in Fig. 2) [33], the HFVC industry is still slow. This is because most enterprises are still
making China one of the major markets for HFCVs. For the first time, in a wait-and-see situation for the production of HFCVs, and a few en­
hydrogen energy was included in the Government Work Report [34] in terprises are likely to commit subsidy fraud [38].
March 2019 to accelerate the construction of recharging and hydrogen Moreover, there is a misalignment between consumer satisfaction
refueling facilities in the public sector. In April 2020, the Energy Law of with HFCVs and government expectations, which leads to a large gap
the People’s Republic of China (Draft for Comments) [35] proposed between the actual and planned promotion volumes. Thus, encouraging
including hydrogen energy in the energy category. In September 2020, more vehicle manufacturers (enterprises) to produce HFCVs, more
the Ministry of Finance, the Ministry of Industry and Information consumers to purchase HFCVs, and regulating the HFCV market are
Technology, and five other departments collaboratively launched the critical issues to be resolved by the Chinese government in promoting
demonstration and application of FCVs and rewarded qualified urban the growth of the HFCV industry. Therefore, it is urgent and necessary to
agglomerations for tackling key problems and applying key core tech­ analyze the development of the HFCV industry to achieve these goals. To
nologies of FCVs. In October 2021, the Central Committee of the more clearly analyze the role of government regulation on the HFCV
Communist Party of China and the State Council issued Opinions on the industry development, this paper introduces the government and con­
Complete and Accurate Implementation of the New Development sumers into the HFCV industry system.
Concept to Do a Good Job in Carbon Dumping and Carbon Neutral Work Game theory can analyze the key decision factors of game players
[36] and made overall plans to encourage the development of the whole and help them make decisions. It assumes all players are perfectly
chain of hydrogen energy, “production - storage - transmission - use”. rational and can directly reach Nash equilibrium once, and decisions
The Medium and Long-Term Plan for the Development of the Hydrogen made based on this model are often far from reality [39]. However, the
Energy Industry (2021–2035) [37] was released in March 2022 jointly government, enterprises, and consumers are limited rational game
by the National Development, the Reform Commission and the National subjects, which cannot guarantee that they can make strategic choices
Energy Administration. Hydrogen energy is an essential component of beneficial to their own interests from the game beginning. Therefore,
the national low-carbon energy system of the future and represents a evolutionary game theory (EGT) can more intuitively discuss and
new strategic industry. research the strategy choices and evolution process of tripartite game
Even though China has been paying further attention to the subjects. It is a multidisciplinary theory, which is derived from biolog­
hydrogen energy industry, it is still at a preliminary stage compared with ical evolution and game theory, and is also their evolution and devel­
developed countries. Currently, the importance of hydrogen energy and opment [39]. EGT assumes that all players are finitely rational and
require a dynamic adjustment process to reach a Nash equilibrium [40].
Therefore, this paper aims to better comprehend the conflict of in­
terests in the Chinese HFCV industry and to address the lack of tradi­
tional game research that leads to the dynamic process being
overlooked. The government’s regulatory, enterprise production, and
consumer purchasing options are all systematically investigated after it
constructs a tripartite EGT model of the government, enterprises, and
consumers. Through the above analysis, the innovations of the paper are
as follows: (1) A tripartite EGT model is constructed for government,
enterprises, and consumers. (2) This paper obtained the evolutionary
stability strategy of every player and the equilibrium points of the HFCV
industry game system, and then discusses and analyzes their stability
based on the replicated dynamic equation. (3) By simulation analysis,
this paper studied and obtained the impact of different initial strategy
selection ratios and key factors on government, enterprise, and con­
sumer behavior strategies. Analyze the evolutionary mechanism of
HFCV industry to stimulate further its medium- and long-term high-
Fig. 1. China Hydrogen Production and Growth Rate (2016–2022). quality development.
The remainder of this paper is organized as follows: Section 2 re­
views the relevant literatures. Section 3 develops a tripartite EGT model
and develops replicator dynamics equations. Section 4 presents a
tripartite EGT simulation analysis. Finally, Section 5 presents the con­
clusions and directions for further research.

2. Literature review

2.1. Research on HFCV from different perspectives

As a necessary application area for hydrogen energy development,

HFCV is a topic of debate amongst scholars with different views. The
academic literature on HFCV has revealed the emergence of several
contrasting themes, such as HFCV development prospects and obstacles,
Fig. 2. China Fuel Cell Vehicle Production and Sales (2016–2022). public attitude and purchase intention, and HFCV market diffusion and

2
W. Wang and J. Li Fuel 348 (2023) 128223

penetration on. Jones et al. [41] advanced the conversation about governments, and enterprises to explore the impact of replacing sub­
HFCVs as a possible strategic option to solve sustainability issues in sidies with award policies in China on the hydrogen vehicles industry.
urban logistics. Zhao et al. [42] comprehensively analyzed China’s Simulation results revealed that supervision and competitive benefits
HFCV industry from three aspects: policy approval, market imple­ are critical to the technological upgrade t of the hydrogen fuel cell
mentation, and technology preparation level. According to Li and (HFC), and technology subsidies should be appropriately set to maintain
Kimura [43], HFCVs, such as passenger cars and buses will be able to the motivation for innovation.
compete with conventional powertrains by 2030. In summary, the studies presented thus far provide evidence that
Based on the comprehensive available literature, it is clear that the HFCVs have a bright future, and that their commercialization is just a
main obstacles to HFCVs are insufficient infrastructure [43–46], few steps away. Many researchers consider EGT models appropriate for
hydrogen fuel storage, and security [43,45]; an insufficient number of exploring the mechanisms of influence among the government, enter­
manufacturers[46];barriers to technology adoption(such as safety prises, and consumers. In addition, EGT models are useful for the
standards, user costs, vehicle mileage, and maintenance)[44,46]; eco­ quantitative analysis of some policy setting criteria. This literature re­
nomic barriers(such as support systems necessary for HFCVs, a lack of view reveals two important themes: HFCV development prospects and
resources, a high purchase price, and the profitability of refueling sta­ obstacles, and public attitude and purchase intention towards HFCV.
tions[45,46]; and politics [44]. However, these studies remain narrow in focus, and the literature
In an analysis of public attitude and purchase intention toward available on this topic has two limitations. First, most of the research
HFCVs, research on public attitudes is weaker than that on purchase using EGT to study HFCVs considered only the relationships between the
intentions. A combination of existing literature shows that public atti­ government and enterprises or between enterprises and consumers,
tudes toward HFCVs have shifted from an initial favorable bias toward disregarding the interplay of the strategies employed by all three
hydrogen and fuel cell technologies to slightly more rational support players. Second, most studies focus on the impact of government subsidy
[47], and that drivers of HFCVs tested in real-world situations have policies on the HFCV industry, thus overlooking the development of the
generally been receptive to vehicle performance and refueling [48]. The industry itself. Faced with the urgent need for a tripartite EGT model
public’s attitude towards HFCV depends on their level of knowledge between the government, enterprises, and consumers to examine HFCV
[49], sense of social value, socio-economic status [50–53], knowledge of industry development, this paper examines the following:
hydrogen technology [54], industrial structure of the city they live in
[55], and environmental awareness [52]. Consumers were generally (1) China’s HFCV industry is continually changing, rendering earlier
sensitive to factors such as HFCV costs [56,57], available infrastructure research on the industry no longer applicable.
[52,57,58], market potential, improvements in drivetrain technology (2) Few studies have examined the development of the HFCV in­
[59], fuel pricing[60], vehicle performance, industry development, dustry from the perspective of government regulation. We aimed
vehicle type and space [61], driving range, refueling time, accessibility to fill this gap by conducting quantitative analyzes,
[16,49,50], emission reductions [49,50], perceived benefits and risks, (3) This paper also focuses on the influence of consumer awareness
and policy interventions [62,63].They are less sensitive to factors such on the entire HFCV industry, which should not be neglected.
as the government’s participation in the technical objectives and pro­
motional activities of the “Low Carbon, Green Growth” agenda man­ 3. Methods
agement[64], fuel availability, and limited mileage[58].
3.1. The game relationships among the three stakeholders
2.2. Application of EGT to HFCVs
The EGT model of the HFCV industry includes three players: the
EGT has been widely applied in many fields, including those related government, enterprises, and consumers. The government can choose
to HFCV. A few studies have developed an EGT model with two par­ between two strategies: positive regulation and negative regulation.
ticipants involved. Wang et al. [65] employed the EGT model for gov­ Enterprises can choose between two strategies: producing HFCVs and
ernments and companies. The results suggested that the strategic not producing HFCVs. Consumers can choose from two strategies: pur­
options of enterprises would be affected by government subsidies, chasing HFCVs and not purchasing HFCVs. The game relationships
penalties, and, to a certain extent, consumer environmental concerns. among the three stakeholders are shown in Fig. 3.
Guan et al. [66] established an EGT model between the government and (1) Relationship between government and enterprises.
consumers in the car market. Studies show that increased government The government is an initiator, supporter, and supervisor of the
subsidy policy, reduced life cycle costs, and improved effectiveness of HFCV industry development. There are two strategies used by govern­
new energy vehicles (NEVs), will cause the model to evolve toward ments: positive and negative regulation. Under the negative regulation,
consumer purchase of NEVs. Liu et al. [67] developed an EGT model for the government will provide R&D subsidies to encourage enterprises to
vehicle manufacturing companies and governments. The simulation produce HFCVs and build infrastructure such as hydrogen refueling
results indicated that the influence of efficient political forces on con­ stations to provide technical support for HFCV production. Under pos­
sumer requests accelerated the evolution of the electric vehicle (EV) itive regulation, in addition to the above incentive measures, the gov­
industry. The policies of dynamic taxation and static subsidies were ernment will also supervise the behavior of enterprises. This is because,
more efficient than other policies in the development of the EV industry. in the production process of HFCVs, some enterprises will take advan­
Ji et al. [68] used the EGT model to explore the interactive mechanism of tage of government financial support to engage in subsidy fraud and
complicated behaviors among local governments and vehicle produce vehicles with unstable technical parameters, lower safety fac­
manufacturing companies. This proved that the gradual elimination of tors or even substandard emissions. Meanwhile, tougher tax policies will
funding policies can help develop the NEV industry. be implemented on FVs, and polluting manufacturing behavior will be
A tripartite EGT model can found in other literature. Sun et al. [69] punished to encourage enterprises to produce HFCVs. Government
developed a government, business, and consumer tripartite EGT model. policies will encourage enterprises to consider the trade-offs and make
The results of the study suggest that the Chinese government should not appropriate strategic decisions.
hurry to eliminate price subsidies for consumers, but instead should (2) Relationship between the government and consumers.
adjust the decline of subsidies dynamically and increase the additional The government will guide consumers’ purchase behavior and pro­
cost of fuel vehicles (FVs) for consumers. NEV companies should in­ vide purchase subsidies and tax relief policies (reduction of vehicle
crease their investments in the R&D of NEVs [69]. Tuofu et al. [40] purchase tax, exemption from vehicle tax, etc.) to stimulate consumers’
developed a tripartite EGT model for the central government, local consumption of HFCVs. The government also increased investment in

3
W. Wang and J. Li Fuel 348 (2023) 128223

Fig. 3. Relationships between government, enterprises, and consumers.

hydrogen refueling stations and improved infrastructure construction to system composed of enterprises, government, and consumers is
eliminate consumer concerns about purchasing HFCVs. considered a complete system. Each player in the system has limited
(3) The Relationship between enterprises and consumers. rationality and incomplete information symmetry, and they are all in the
Enterprises and consumers, as supply and demand parties in the initial stages of the game.
HFCV market, have strategic choices that influence each other’s bene­ (2) All players in the game should adjust their strategies dynamically
fits. If companies choose to produce HFCVs, consumers’ purchases will at different stages of the development in the HFCV industry.
bring additional benefits to companies. Consumers make purchase de­ (3) In the EGT model, while the government selects a negative
cisions based on the products and services offered by companies. regulation strategy, it will support enterprises producing HFCVs by
providing R&D subsidies S1 . If the government subsidy proportion is α,
3.2. Model assumptions and parameter settings the amount of R&D support from the government to enterprises will be
αS1 . To stimulate the popularization of HFCVs, the government will
The parameters used to develop the tripartite EGT model are pro­ invest in infrastructure construction S2 . If the investment proportion is β,
vided in Table 1. the investment amount is βS2 . While the government selects the positive
Several assumptions were made to analyze the divergence of benefits regulation strategy, it will regulate and punish the fraud compensation
and optimal strategies of the government, enterprises, and consumers: P1 of enterprises (the government will deduct the subsidy funds for
(1) Without considering other constraints in a natural environment, a enterprises, severely cancel the HFCV subsidy qualification, and give up
the government’s purchase of other models.). If the proportion of gov­
ernment supervision is γ, the loss of enterprises will be γP1. If an en­
Table 1
terprise produces only FVs but does not HFCVs, the government will
Description of Parameter Symbols.
punish it by levying an emission fee and a carbon tax P2 . If the penalty
Parameters Descriptions
proportion is μ, the penalty costs is μP2 . In particular, the government
Government parameters will provide tax benefits B1 to consumers who purchase HFCVs.
S1 Government R&D subsidies (4) The benefit of enterprises that produce HFCV is B2 , and the
α Government R&D subsidy proportion
benefit of enterprises that do not produce HFCV is B3 . The extra profits
αS1 Government R&D subsidies for enterprises
S2 Government investment in infrastructure construction of consumers’ contributions to the enterprises by purchasing HFCVs is
β Government investment proportion B4 , and the R&D cost paid by the enterprise for producing HFCV is C.
βS2 Government investment for HFCV infrastructure construction (5) The benefit to consumers who purchase HFCVs is B5 . The benefit
P1 Government fines for fraudulent subsidies
for consumers who do not buy HFCVs is B6 . The additional benefit
γ Government supervision proportion
γP1 Government supervision of fines for enterprises fraudulent subsidies
brought by the government’s construction of the hydrogen refueling
P2 Government levy emission fee and a carbon tax station infrastructure to consumers who buy HFCVs is γP1 , and the
μ Government penalties proportion additional benefit brought by the government’s positive regulation to
μP2 Government penalties enterprise for levy emission fee and a carbon tax consumers who buy HFCVs is ΔB.
B1 Tax benefit of consumers purchasing HFCVs
(6) Let x (0 ≤ x ≤ 1) represents the probability of governments that
x Probability of the government implement positive regulation
Enterprise parameters choosing to implement positive regulation and let 1 − x denotes the
B2 The benefit of enterprises producing HFCVs probability of governments choosing to implement negative regulation.
B3 The benefit of enterprises not producing HFCVs Suppose that y (0 ≤ y ≤ 1) denotes the probability of enterprises
B4 Additional benefits of consumer’s contributions to the enterprise by
choosing to produce HFCVs, and 1 − y denotes the probability of enter­
purchasing HFCVs
C R&D cost of each HFCV
prises choosing not to produce HFCVs. Finally, let z (0 ≤ z ≤ 1) denote
μP2 Cost of the enterprises not producing HFCVs the probability of consumers choosing to buy HFCVs and 1 − z denote the
y Probability of the enterprise producing HFCVs probability of consumers choosing not to buy HFCVs.
Consumer parameters
B5 The benefit of consumers purchasing HFCVs
B6 The benefit of consumers not purchasing HFCVs
βS2 Additional benefits of the government’s investment in HFCV 3.3. Payoffs of stakeholders
infrastructure construction to the consumers by purchasing HFCVs
ΔB Additional benefits of the government’s positive supervision to the With these assumptions above, the player’s payment matrix is shown
consumers’ by purchasing HFCVs in Table 2. The specific payout is shown in Table 3.
z Probability of consumers purchasing HFCVs
Thus, the replicator dynamic equations F(x), F(y), F(z) of the three

4
W. Wang and J. Li Fuel 348 (2023) 128223

Table 2
The payment matrix of the tripartite game.
Enterprises

producing HFCVs (y) non-producing HFCVs (1 − y)

Strategies
Consumers

purchasing HFCVs (z) non-purchase HFCVs (1 − z) purchasing HFCVs (z) non-purchase HFCVs (1 − z)

positive regulation (x) (a1 , b1 , c1 ) (a2 , b2 , c2 ) (a3 , b3 , c3 ) (a4 , b4 , c4 )

Government
negative regulation (1 − x) (a5 , b5 , c5 ) (a6 , b6 , c6 ) (a7 , b7 , c7 ) (a8 , b8 , c8 )

3.4.2. Progressive stability analysis of enterprises

Table 3
Whether the enterprises choose to produce HFCVs or not to produce
The specific payoff of government, enterprises, and consumers.
HFCVs strategy can be determined by replicator dynamics equation F(y)
Payoff Government Enterprises Consumers and seeking partial derivative [70–72].
(a1 , b1 , c1 ) γP1 − αS1 − βS2 − B1 B2 + B4 + αS1 − γP1 − C B1 + B5 + ΔB + βS2
dF(y)
(a2 , b2 , c2 ) γP1 − αS1 − βS2 B2 + αS1 − γP1 − C B6 (3)
′
U (y) = = (1 − 2y)[(μP2 − γP1 )x + B4 z + αS1 + B2 − B3 − C ]
(a3 , b3 , c3 ) μP2 − βS2 B3 − μP2 0 dy
(a4 , b4 , c4 ) μP2 − βS2 B3 − μP2 B6
(a5 , b5 , c5 ) − B1 − αS1 − βS2 B2 + B4 + αS1 − C B1 + B5 + βS2
According to the stability of the evolutionary game characteristic,
(a6 , b6 , c6 ) − αS1 − βS2 B2 + αS1 − C B6 the condition for the enterprises evolutionary game to reach stability is
(a7 , b7 , c7 ) B3 0 U (y)〈0. The analysis of Equation (3) is performed.
′
− βS2
(a8 , b8 , c8 ) B3 B6
(1) when x = B3 +C−μPB22−− γP
B4 z− αS1
, then F(y) = 0 , it is steady state for all
− βS2
1

y-axes, at this time the enterprises select any proportion of the produce
stakeholders can be obtained [70–72], and the three-dimensional dy­ HFCV strategy is a stable strategy. When the probability of government
namic system is: regulation reaches B3 +C−μPB22−− γP
B4 z− αS1
or the probability of consumer pur­
1
⎧ B3 +C− B2 − (μP2 − γP1 )x− αS1
⎪ chase of HFCV reaches B4 , the probability of enter­
⎨ F(x) = x(EX1− EX) = x(1− x)[(B1 +γP1 − μP2 )y+ μP2 ]
F(y) = y(EY1− EY) = y(1− y)[(μP2 − γP1 )x+B4 z+ αS1 +B2 − B3 − C] (1) prises choose to produce HFCV is stable.
⎪
⎩ F(z) = z(EZ1− EZ) = z(1− z)[(ΔBx+B1 +B5 +βS2 )y− B6 ] = B3 +C−μPB22−− γP
(2) When x ∕ B4 z− αS1
1
, let F(y) = 0, it is found that y = 0 and
y = 1 are the two stable states of y, respectively. It can be divided into
3.4. Progressive stability and results analysis two situations.
B3 +C− B2 − B4 z− αS1 dF(y) dF(y)
① When μP2 − γP1 < x < 1, dy |y=0 > 0, dy |y=1 < 0, at this
3.4.1. Progressive stability analysis of government instance y = 1 is the globally unique evolutionary stabilization strategy,
Whether the government select a negative or positive regulation that is when the probability of government positive regulation or the
strategy can be determined by replicator dynamics equation F(x) and probability of consumers purchase HFCVs rises to a certain level, en­
seeking partial derivative [70–72]. terprises will choose to produce HFCV as the optimal strategy.
dF(y) dF(y)
dF(x) ② When 0 < x < B3 +C−μPB22−− γP
B4 z−
1
αS1
, dy |y=0 < 0, dy |y=1 > 0, t this
(2)
′
U (x) = = (1 − 2x)[(B1 + γP1 − μP2 )y + μP2 ]
dx instance y = 0 is the globally unique evolutionary stabilization strategy,
that is when the probability of government positive regulation or the
According to the stability of the evolutionary game characteristic,
probability of consumers purchase HFCVs decrease to a certain extent,
the condition for the government evolutionary game to reach stability is
enterprises will choose to not produce HFCV as the optimal strategy.
U (x)〈0. The analysis of Equation (2) is performed.
′

A diagram of the dynamic evolution of enterprises management

(1) When y = B1 +μμPP22− γP1 , then F(x) = 0, it is steady state for all x-axes, decisions is shown in Fig. 5.
and any strategy chosen by the government in this state is a stable
strategy. It also means that the probability of government chooses to 3.4.3. Progressive stability analysis of consumers
positive regulation the HFCV industry is stable when the probability that Whether the consumers choose to purchase HFCVs or not to purchase
enterprises choose to produce HFCV reaches y = B1 +μμPP22− γP1 .
HFCVs strategy can be determined by replicator dynamics equation F(z)
(2) When y ∕ μP2 and seeking partial derivative [70–72].
= B1 +μP2 − γP1 , let F(x) = 0, then we can find x = 0 and x =
1 are the two stable states of x, respectively. It can be divided into two dF(z)
(4)
′
U (z) = = (1 − 2z)(ΔBx + B1 + B5 + βS2 )y − B6
situations. dz
① When B1 +μμPP22− γP1 <y< 1, dF(x)
dx |x=0 < 0, dF(x)
dx |x=1 > 0, at this instance According to the stability of the evolutionary game characteristic,
x = 0 is the globally unique evolutionary stabilization strategy, which the condition for the consumers evolutionary game to reach stability is
means when the probability of enterprises producing HFCV reach a U (z)〈0. The analysis of Equation (4) is performed.
′

certain level, it is optimum for the government to choose negative (1) When x = B6 − (B1 +B5 +βS2 )y
, then F(z) = 0 , it is steady state for all z-
regulation of the HFCV industry.
ΔBy

dF(x)
axes. When the probability of government regulation reaches
② When 0 < y < B1 +μμPP22− γP1 , dx |x=0 > 0, dF(x)
dx |x=1 < 0, at this instance B6 − (B1 +B5 +βS2 )y
or the probability of enterprises produce HFCVs reaches
x = 1 is the globally unique evolutionary stabilization strategy, which
ΔBy
B6
means when the probability of enterprises producing HFCVs decreases ΔBx+B1 +B5 +βS2 , the probability of consumers chose to purchase HFCVs is
to a certain degree, the government chooses to support the HFCV in­ stable.
dustry with subsidies is the optimal strategy. For example, the subsidies = B6 −
(2) When x ∕ (B1 +B5 +βS2 )y
ΔBy , let F(z) = 0, it is found that z = 0 and
for enterprises to develop HFCVs and subsidies for consumers to pur­ z = 1are the two stable states of y, respectively. It can be divided into
chase HFCVs. two situations.
A diagram of the dynamic evolution of government management ① When B6 − (B1 +B5 +βS2 )y
< x < 1,dF(z) dF(z)
dz |z=0 > 0, dz |z=1 < 0, at this
decisions is shown in Fig. 4.
ΔBy

5
W. Wang and J. Li Fuel 348 (2023) 128223

Fig. 4. Dynamic evolution of government management decisions.

Fig. 5. Dynamic evolution of enterprises management decisions.

Fig. 6. Dynamic evolution of consumer s management decisions.

point z = 1 is the global evolutionary stable strategy, which means when

Table 4
the probability of government positive regulation or the probability of
The stability points of the tripartite EGT model.
enterprises select to produce HFCVs increase to a certain degree, con­
sumers will choose purchase HFCVs as the optimal strategy. Number Stability point

② When 0 < x < B6 − (B1 +B5 +βS2 )y dF(z)

ΔBy , dz |z=0 < 0, dF(z)
dz |z=1 > 0, at this 1 E1 (0, 0, 0)
2 E2 (1, 0, 0)
point z = 0 is the global evolutionary stabilization strategy, which 3 E3 (0, 1, 0)
means when the probability of government positive regulation or the 4 E4 (0, 0, 1)
probability of enterprises choose to produce HFCV decreases to a certain 5 E5 (1, 1, 0)
extent, consumers choose not to purchase HFCV as the optimal strategy. 6 E6 (1, 0, 1)
7 E7 (0, 1, 1)
A diagram of the dynamic evolution of enterprises management
8 E8 (1, 1, 1)
decisions is shown in Fig. 6. 9 E9 [1, B6 /(B1 + B5 + μP2 ), − (B2 + αS1 − B3 − C)/B4 ]
10 E10 [0, B6 /(B1 + B5 + βS2 ), − (B2 − B3 − C + αS1 )/B4 ]
3.4.4. Results analysis 11 E11 [ − (B1 + B5 − B6 + μP2 )/ΔB, 1, − γP1 /B1 ]
12 E12 [(B2 − B3 − C + αS1 )/(γP1 − μP2 ), − μP2 /(γP1 − μP2 ), 0]
To find the equilibrium point of the three-party EGT model, we set
13 E13 [1, B6 /(ΔB + B1 + B5 + βS2 ), − (B2 − B3 − C + αS1 − γP1 + μP2 /B4 ]
replicator dynamics in Equations (1) to zero. The stability points of the 14 E14 [− (B4 μ2 P22 + B1 B3 B6 + B1 B6 C + B4 B6 γP1 + B1 B4 μP2 + B1 B5 μP2 B1 B2 ⋅
tripartite EGT model are shown in Table 4. B6 − B1 B2 αS1 − B4 B6 μP2 )/ (B1 B6 γP1 + B4 μP2 ), − (B1 B6 γP1 + ΔBB4 μP2 − B1 ⋅
The equilibrium point must be a pure strategic Nash equilibrium B6 μP2 )/(B21 μ2 P22 + ΔBB1 B3 + ΔBB1 C + B21 μP2 + ΔBB4 γP1 + B1 B5 μP2 − ΔB⋅
when it is asymptotically stable in a tripartite EGT [70].E9 − E14 can be B1 B2 − B1 2 γP1 − ΔBB4 μP2 − ΔBB1 S1 − B1 γP1 μP2 ), − (B6 γ2 P1 2 + B6 μ2 P2 2 +
determined as non-sink due to it being a mix-strategy Nash equilibrium. γ2 P1 2 μ2 P2 2 + ΔBB2 μP2 + ΔBαS1 μP2 + B1 γP1 μP2 + ΔBαS1 μP2 + B1 γP1 μP2 −
It does not satisfy the stability criterion of the evolutionary game of the μ3 P32 − B5 μP2 − ΔBB3 μP2 − ΔBCμP2 − 2B6 γP1 μP2 )/(B1 B6 γP1 + ΔBB4 μP2 −
B1 B6 μP2 )]
multi-participant subject model. This equilibrium solution makes the

6
W. Wang and J. Li Fuel 348 (2023) 128223

tripartite EGT non-optimal, therefore only the E1 -E8 equilibrium point funds or even disqualifies enterprises from subsidies after heir fraudu­
needs to be considered. Consequently, through the use of the partial lent behavior. This may also suggest that the government’s policy in­
derivatives of F(x), F(y), and F(z) regarding x, y, and z, the Jacobian centives are not sufficiently strong. With high R&D costs and limited
matrix (J) can be obtained [73], and the components of the Jacobian application prospects, companies more likely to choose to commit
matrix (J) can be obtained, and the components in the Jacobian matrix fraudulent subsidies. Government behavior will eventually tend toward
(J) correspond to Equations (5). positive regulation. Enterprises will eventually tend not to produce

⎡ ⎤
⎢ ∂F(x) ∂F(x) ∂F(x) ⎥
⎢ ∂x ∂y ∂z ⎥
⎢ ⎥
⎢ ⎥
⎢ ∂F(y) ∂F(y) ∂F(y) ⎥
J=⎢
⎢ ∂x
⎥
⎥
⎢ ∂y ∂z ⎥
⎢ ⎥
⎢ ∂F(z) ∂F(z) ∂F(z) ⎥
⎣ ⎦
∂x ∂y ∂z
⎡ ⎤
⎡ ⎤
⎢ (γP1 − μP2 ⎥ (5)
⎢ (1 − 2x)⎣ ⎦ x(1 − x)(γP1 + B1 z − βS2 ) B1 xy(1 − x) ⎥
⎢ +B1 z)y + μP2 ⎥
⎢ ⎥
⎢ ⎥
⎢ ⎡ ⎤ ⎥
⎢ ⎥
⎢ (μP2 − γP1 )x + B2 + B4 z ⎥
⎢
= ⎢ y(1 − y)(μP2 − γP1 ) (1 − 2y) ⎣ ⎦ B4 y(1 − y) ⎥
⎥
⎢ +αS1 − B3 − C ⎥
⎢ ⎥
⎢ ⎡ ⎤ ⎥
⎢ ⎥
⎢ (ΔBx + B1 + B5 ⎥
⎢ (1 − z)ΔByz z(1 − z)(ΔBx + B1 + B5 + βS2 ) (1 − 2z)⎣ ⎦ ⎥
⎣ ⎦
+βS2 )y− B6

HFCVs, and consumers will eventually tend not to choose to purchase

The E1 -E8 equilibrium points are substituted into the Jacobian ma­ HFCVs.
trix, and the eigenvalues for the equilibrium points [73], corresponding (2) Scenarios 2.
to the Jacobian matrix, can be obtained separately, which is summarized If E5 (1, 1, 0) is an ESS, Condition 2 must be satisfied. It is evident
as shown in Table 5. from condition 2 that when B1 +B5 +ΔB +μP2 − B6 < 0 and B5 + γP1 +
According to EGT, that the equilibrium point satisfying all eigen­ C − B2 − αS1 − μP2 < 0. This indicates that when the cost of government
values of the Jacobian matrix is non-negative is the ESS of the replicated regulation is less than the sum of the penalties levied on enterprises by
dynamic system of equations [74]. The results shown in Table 4 indicate government regulation, the government chooses to regulate positively.
that there are three points that can reach asymptotic stable equilibrium Higher government support, and lower R&D costs incentivize enter­
when certain conditions are satisfied. Taking E2 (1, 0, 0) as an example, prises to produce HFCVs. However, the gain from not purchasing HFCVs
the conditions for satisfying the evolutionary stabilization strategy are is higher than the sum of purchasing HFCVs and government subsidies,
discussed. If λ3 satisfy the scenarios 1, that is B2 +
μP2 − B3 − αS1 − γP1 − C < 0. The rest of the stability conditions at the
three equilibrium points are shown in Table 6. Table 6
Stability conditions of the equilibrium point for tripartite evolutionary game.
Serial Equilibrium Stability Scenarios
3.5. Game Scenarios for HFCVs industry development
Number Point

1 E2 (1, 0, 0) B2 + μP2 − B3 − αS1 − γP1 − C < 0

(1) Scenarios 1.
2 E5 (1, 1, 0) B1 + B5 + ΔB + μP2 − B6 < 0, B5 + γP1 +
If E2 (1, 0, 0) is an ESS, then the condition 1 must be satisfied. It is C − B2 − αS1 − μP2 < 0
evident from Condition 1 that when B2 + μP2 − B3 − αS1 − γP1 − C < 0. 3 E8 (1, 1, 1) B6 − B1 − B5 − ΔB − μP2 < 0, B3 + C +
This implies that the government has established a high standard of γP1 − B2 − B4 < 0
regulatory measures. At the same time, the government deducts subsidy

Table 5
The characteristic values of equilibrium points.
Equilibrium Point Characteristic Values Incremental stability

λ1 λ2 λ3

E1 (0, 0, 0) μP2 − B6 B2 + αS1 − B3 − C instability

E2 (1, 0, 0) − B6 − μP2 B2 + μP2 − B3 − αS1 − γP1 − C condition 1
E3 (0, 1, 0) γP1 B1 + B5 + βS2 − B6 B3 + C − B2 − αS1 instability
E4 (0, 0, 1) B6 μP2 B2 + B4 + αS1 − B3 − C instability
E5 (1, 1, 0) − γP1 B1 + B5 + ΔB + μP2 − B6 B5 + γP1 + C − B2 − αS1 − μP2 condition 2
E6 (1, 0, 1) B6 − μP2 B2 + B4 + αS1 + μP2 − B3 − C − γP1 instability
E7 (0, 1, 1) B1 + γP1 B6 − B1 − B5 − μP2 B3 + C − B2 − B4 − αS1 instability
E8 (1, 1, 1) − B1 − γP1 B6 − B1 − B5 − ΔB − μP2 B3 + C + γP1 − B2 − B4 condition 3

7
 W. Wang and J. Li Fuel 348 (2023) 128223

and consumers eventually choose not to purchase HFCVs. industry to develop is through active regulation. Governments tend to
(3) Scenarios 3. vary in terms of initial proportions of positive regulation, but all
If E8 (1, 1, 1) is an ESS, then Condition 3 must be satisfied. It is evident generally tend to adopt a positive regulation strategy after a period of
from Condition 3 that B6 − B1 − B5 − ΔB − μP2 < 0 and time, suggesting that a gradual choice of negative regulation by the
B3 +C +γP1 − B2 − B4 < 0. This illustrates that when the cost of govern­ government might not be an efficient long-term strategy. In all three
ment regulation is greater than the sum of the penalties levied on the situations, the higher the initial value, the less time it takes to reach the
enterprise by government regulation, the government chooses to regu­ ESS. Under current market conditions, if positive regulation continues, it
late actively. Government subsidies to companies are greater than R&D can effectively promote enterprises to produce HFCVs and stimulate
costs and penalties due to positive regulation, so enterprises will choose consumers to purchase HFCVs.
to produce HFCVs. The sum of government subsidies and investment in (2) Producing high-quality HFCVs will be the preferred choice for
supporting facilities for consumers purchasing HFCVs are greater than enterprise in the future. The evolutionary trend of enterprises choosing
the gain of consumers not purchasing HFCVs. Consumers choose to to produce HFCVs is the same when they have different primary pro­
purchase HFCVs because of the lure of high subsidies and the portions, all of them tend to produce HFCVs after a period of time. In all
improvement of supporting facilities. three situations, the higher the initial value, the less time it takes to
Considering the above three scenarios, only Condition 3 is the ideal reach the ESS. However, it is worth mentioning that at the initial pro­
expectation for HFCV industry development in the medium to long term. portion of 0.3, there is a trend of the number of enterprises producing
Then,E8 (1, 1, 1) is an ESS in accordance with the future expectations of HFCVs increasing, decreasing, and increasing respectively, which in­
the market. Section 4.1 will confirm this view. dicates that the government subsidy policy gradually plays an important
role and choice of HFCVs production is the correct strategy to adapt to
4. Simulation analysis of the tripartite EGT model the medium and long-term growth of the market.
(3) For consumers, choosing to purchase HFCVs is the best strategy,
According to the theoretical analysis, the constraints and the repli­ which is the government’s final goal of implementing positive regula­
cated dynamic equations, using Matlab2022a to imitate the evolu­ tions. While the primary proportion of consumers choosing to buy
tionary process of enterprises, government, and consumers behaviors, HFCVs varies, they evolve in essentially the same way, with all con­
and analyze the impact of each parameter change on the evolutionary sumers tending to purchase HFCVs after a certain period. In all three
results. cases, the larger the initial value, the shorter the time to reach the ESS.
However, at an initial proportion of 0.3, the number of consumers
4.1. Effect of the initial market proportions of the three parties on the choosing to purchase HFCVs decreases gradually at first, reaching the
Game’s equilibrium lowest point in 1.2 years, and then the number of consumers choosing to
purchase HFCVs increases gradually. This phenomenon indicates that
In this section, we discuss the evolutionary trends when there are government subsidies and the progressive improvement of infrastruc­
different initial proportions of governments, enterprises, and consumers. ture shift consumers’ purchase intentions.
In order to examine the impact of different primary proportions of
government, enterprises, and consumers on their choice of strategies. 4.2. Analysis of the influence of other factors on the strategy selections of
Let x , y and z denoting the primary proportions of the government, game groups
′ ′ ′

enterprises, and consumers. We conducted simulations where the orig­

inal proportions were 0.3, 0.5, and 0.9 [69,75,76]. The results of the In this section, we select four variables and investigate their impact
simulation are in Fig. 7. on the evolution strategy selection for the three players. The four vari­
(1) For the government, one of the most effective ways for the ables are the strength of regulation γ, punishment μ, subsidy α, and in

Fig. 7. The Tripartite evolutionary tracks.

8
­
W. Wang and J. Li Fuel 348 (2023) 128223

Fig. 8. The Tripartite evolutionary tracks with different supervision strengths.

vestment β. From the previous section, we realize that the three players’ evolve towards purchasing HFCVs. Under high supervision, it will
original proportions only change the time and speed of evolution, and do initially promote consumer purchasing behavior, so the percentage of
not influence the ultimate result of evolution. We performed the consumers purchasing HFCVs will gradually increase, reaching the
following simulations using the initial proportions (0.5, 0.5, 0.5). highest point in 0.67 years. However, with time, consumers will even­
tually evolve towards not purchasing HFCVs. This is because even with
4.2.1. The effect of supervision strengths γ on game equilibrium high supervision, some enterprises with core competitiveness are still
Assume that γ is set to 0.2, 0.5, and 0.9 [75–79], which correspond to able to produce high-quality HFCVs to meet consumer demand. Con­
the low, medium, and high strengths of government supervision, sumers enjoy a good driving experience, and their expected revenue
respectively. The simulation results are in Fig. 8. increases. However, with the change in time, some enterprises were
With proper supervision, enterprises will evolve toward producing exposed to fraudulent subsidies under the government’s supervision,
HFCVs, consumers will evolve toward purchasing HFCVs, and eventu­ which led to the continuous decline and loss of market share and
ally remain stable. Too much supervision will cause enterprises not to customer groups, and finally the abandonment of HFCV production. The
produce HFCVs, consumers not to purchase HFCVs, and government fraudulent behavior of these enterprises makes consumers question the
regulation will not have a predetermined effect. quality and safety of HFCVs, combined with concerns about the con­
(1) When supervision is of low or medium intensity, the government struction of supporting facilities, which affects consumer purchasing
will always choose the strategy of positive regulation. When regulation behavior and makes the HFCV industry market more unpredictable,
is high-intensity, the initial government regulation can stimulate the with the core enterprises eventually evolving toward not producing
development of the HFCV industry, and the government will move to­ HFCVs and consumers eventually evolving towards not buying HFCVs.
wards positive regulation, reaching the highest point after 1 year. In summary, medium-intensity government supervision can facilitate
However, with time changes, the high intensity of regulation has the scientific and organized growth of the HFCV industry; low-intensity
conversely inhibited HFCV industry development. Thus, the rate of supervision will disturb market discipline, and high-intensity supervi­
government evolution to positive supervision gradually decreases. sion will restrain the development of the HFCV industry.
Eventually, the government will abandon heavy-handed supervision and
focus on the construction of infrastructure and support for R&D. 4.2.2. The effect of penalty strengths μ on game equilibrium
(2) When supervision is low, enterprises evolve toward producing Assume that μ is set to 0.2, 0.5, and 0.9 [75–79], which correspond to
HFCVs. When the supervision is at medium strength, enterprises will the low, medium, and high strengths of government penalties, respec­
first evolve towards not producing HFCVs, reaching the lowest point in tively. The simulation results are shown in Fig. 9.
1.9 years. However, as time changes, the market environment for HFCVs When the penalty strength is low, enterprises will evolve toward not
under government supervision steadily improves, leading to an increase producing HFCVs, consumers will evolve toward not purchasing HFCVs,
in the production of HFCVs with certified technology and high-security and eventually remain stable. Increasing the penalty will make enter­
factors, which are promptly introduced to the market with the support of prises evolve toward producing HFCV, consumers choose to purchase
the government. Meanwhile, the number of consumers who buy HFCVs HFCV, and government regulation will have a considerable effect.
gradually increases and indirectly influences the purchasing behavior of (1) The speed of government evolution toward positive regulation
other consumers, causing an increase in market demand for HFCVs. accelerates with an increase in penalties. Low penalties are insufficient
Eventually, all enterprises choose the strategy of producing HFCVs. to restrict the production of enterprises. High R&D costs, inadequate
When the government supervises at a high intensity, enterprises will infrastructure, consumer purchase intentions, and other factors restrict
eventually abandon the production of HFCVs because they cannot afford enterprise production. When the penalty is high enough, it will restrict
the government’s fines for fraudulent behavior. enterprises’ polluting behavior and indirectly promote HFCVs, causing
(3) When supervision is at low or medium strength, consumers will more enterprises to choose to produce HFCVs. Therefore, the higher the

9
W. Wang and J. Li Fuel 348 (2023) 128223

Fig. 9. Tripartite evolutionary tracks with different penalty strengths.

government penalty, the more favorable the development of the HFCV a high point in 1.1 years. However, as time changes, the high R&D costs
industry. and the penalty will cause enterprises to have eventually evolved toward
(2) When the penalty strength is low, all companies choose not to not producing HFCVs. When the penalty is high, the proportion of en­
produce HFCVs after 1.2 years. Due to the high market share and profits terprises choosing to produce HFCV will initially increase, reaching the
of existing traditional vehicles, the HFCV industry is still at an initial highest point in 1.7 years. Eventually, enterprises will evolve towards
stage of development. Although there are penalties for producing HFCVs production, and after 0.7 years, all enterprises will choose to
traditional vehicles (such as incurring emission fees and carbon taxes), produce HFCVs. Therefore, the penalty for pollutants produced by
enterprises will eventually choose not to produce HFCVs because the HFCVs enterprises should be sufficiently large to promote their prod­
penalties’ strengths are small. When penalty strengths increase moder­ ucts. Therefore, companies shoule choose to produce energy-saving and
ately, the proportion of HFCV production will increase at first, reaching environment friendly HFCs that meet the laws of automotive

Fig. 10. Tripartite evolutionary tracks with different subsidy strengths.

10
W. Wang and J. Li Fuel 348 (2023) 128223

development and social needs. R&D cost of enterprises, stimulate enthusiasm for technological inno­
(3) When the penalty is low, the proportion of consumers purchasing vation and encourage enterprises to choose to produce HFCVs. There­
HFCVs increases at first, reaching the highest point after 0.6 years, but fore, besides providing direct financial support and purchase subsidies,
shortly after this proportion decreases, finally evolving into HFCVs no the government should pay attention to HFCV industry technology
longer being purchased. It is closely related to the strategic choice of innovation and provide enterprise subsidies for the R&D of core tech­
enterprises not to choose to produce HFCVs under such government nologies and key components to boost the growth of China’s HFCV
regulation. When the penalty is high, consumers eventually evolve to­ industry.
ward purchasing HFCVs; the higher the penalty, the faster the rate of (3) Government subsidies to enterprises for R&D affect their pro­
evolution of consumer purchase behavior. Therefore, the government duction behavior, subsequently impacting consumers’ purchasing
should penalize enterprises that generate pollution through the pro­ behavior. Thus, low subsidy is not conducive for consumers to pur­
duction of traditional energy vehicles and encourage them to produce chasing HFCVs, and a high subsidy will encourage consumers to pur­
HFCVs and constantly improve their technology to bring more added chase HFCVs.
value to consumers.
4.2.4. The effect of investment strengths β on game equilibrium
4.2.3. The effect of subsidy strengths α on game equilibrium Assume that β is set to 0.2, 0.5, and 0.9 [75–79], which correspond to
Assume that α is set to 0.2, 0.5, and 0.9 [75–79], which correspond to the low, medium, and high strengths of government investment,
the low, medium, and high strengths of government penalties, respec­ respectively. The simulation results are shown in Fig. 11.
tively. The simulation results are shown in Fig. 10. When government infrastructure investment strength is low, enter­
When government subsidy intensity is low, enterprises evolve to­ prises evolve toward not producing HFCVs, consumers evolve toward
ward not producing HFCVs, and consumers choose not to buy HFCVs, not purchasing HFCVs, and finally remain stable. When government
and eventually remain stable. When the subsidy strength is high, en­ infrastructure investment strength is high, it will make enterprises shift
terprises evolve toward producing HFCVs, consumers buy HFCVs, and to produce HFCVs and consumers buy them, and suggesting that gov­
the government regulation will have a considerable effect. ernment regulation has been effective.
(1) As the intensity of subsidies increase, the rate of evolution of the (1) The intensity of investment does not affect the results of the
government to active regulation slowly increases. While the government government’s evolution toward positive regulation. The government’s
is subsidizing companies’ technology and R&D, it must regulate com­ evolution to active regulation occurs at a lower speed as its investments
pany behavior and standardize the market order that continuously im­ become progressively more intense. This is because while the govern­
proves the companies’ R&D achievements and the efficiency of ment is investing in infrastructure, it also supervises the behavior of
government financial expenditures. When government subsidies are enterprises and regulates the market order. When government invest­
high, the HFCV market can evolve healthily, as enterprises mature in ment strength is high, infrastructure gradually improves, rendering the
their R&D technologies. At this point, the government abandons high- HFCV market healthy. At this time, the government abandons high-
intensity regulation and penalties, thus slowing down the rate of posi­ intensity regulation and punishments, thus slowing down the pace of
tive regulation evolution. positive regulatory action.
(2) When the government provides a low subsidy to enterprises on (2) With insufficient investment in infrastructure and inadequate
R&D, enterprises will choose the strategy of not producing HFCV. This is support facilities, it is difficult to change consumers’ mindset, resulting
because the technology of HFCVs is immature, the R&D cost of core in a downturn in the HFCV market, and enterprises will evolve in the
technologies is high, and enterprises can hardly pay the high R&D cost direction of not producing HFCVs. When the investment in infrastruc­
and therefore give up the production of HFCs. When the government ture is moderate or high, enterprises will initially choose not to produce
provides high funding for enterprises to develop HFCV, it can reduce the HFCVs owing to the time delay of the infrastructure investment effect.

Fig. 11. Tripartite evolutionary tracks with different investment strengths.

11
W. Wang and J. Li Fuel 348 (2023) 128223

However, as the investment effect continues to show, the infrastructure hydrogen supply chains, and fully absorb the advanced policy experi­
will improve (such as the establishment of hydrogen refueling stations), ence of countries leading in HFCV development. Finally, government
and consumers become more motivated to purchase HFCVs. Enterprises subsidies for HFCV R&D ought to increase. For instance, strengthening
will change their product strategy and gradually move towards the the intensity of subsidies for core HFC technologies can improve enter­
production of HFCVs. prises’ ability to self-innovation.
(3) The government is constantly building and improving infra­ ④ Attention should be paid to the construction and improvement of
structure, the most direct beneficiaries of which are consumers. When infrastructure. The central budget investment and financial loans should
the investment strength in infrastructure is low, consumers will not be provided to support the construction of hydrogen refueling station
evolve towards purchasing HFCVs evidently, the greater the investment, infrastructure across the country, and establish reward and subsidy
the faster the rate of evolution of consumers’ purchasing of HFCVs. In policies. The construction of a sensible investment return mechanism
China, the construction of infrastructure such as hydrogen refueling and the creation of a reasonable business model could help encourage
stations is a prerequisite for promoting HFCV. social capital to enter the construction and hydrogen refueling station
operation and improve the economic benefits of HFCV.
5. Conclusions and discussions for further research (2) The government should clarify how HFCV policies affect con­
sumer behavior and decision-making. At present, HFCVs are still in the
5.1. Conclusions and implications early stage of market promotion, and consumers have not developed a
complete product awareness. Policies are still required to change con­
In this paper, we establish a game model for the evolution of HFCV sumers’ purchase intentions. The government can focus on promoting
industry participants according to the assumption of limited rationality the HFCV industry, such as increasing the incentive funds for the pro­
of game parties and analyze the evolutionary behavior of enterprises and motion and application of HFCV, which has a positive guiding effect on
consumers under government regulation using Matlab2022a numerical consumers. As the concept of environmental protection has becomes
simulation. The results show that: increasingly popular, the greatest advantage of HFCVs over traditional
(1) Appropriate supervision can boost the production and con­ vehicles is that they are pollution-free. If consumers can receive certain
sumption of HFCVs. Medium-intensity supervision can facilitate the policy subsidies and related tax incentives when purchasing, they can
growth of HFCV research and industry. Low-intensity supervision dis­ also promote their purchase intention.
turbs market discipline, while high-intensity supervision restrains the (3) It is critical to clarify the relationships between enterprises and
development of the HFCV industry. consumers to facilitate the development and growth of the HFCV mar­
(2) The more severe the government penalties for pollutant emis­ ket. The marketing of HFCVs must focus on key customers and on pro­
sions, the more likely enterprises will transition to HFCV production and moting product information and services through these customers.
consumers will buy them. Owing to the operating characteristics of the electric motor, HFCVs are
(3) The higher the government’s subsidy for technology and R&D for smoother and run with relatively little noise. Structurally, the power­
HFCV production, the more beneficial it is for enterprises to invest in train of an HFCV requires less room, increasing the space inside the
research to overcome technological barriers, make original break­ vehicle. These features and performance will enhance the customer’s
throughs, and subsequently active consumer purchase of HFCV. goodwill towards HFCVs. In developing word-of-mouth marketing, en­
(4) When government infrastructure investment is low, consumers terprises must actively communicate with consumers after selling
choose not to buy HFCVs, whereas when it. is moderate and high, HFCVs, and pay attention to consumers’ experiences and feedback on
consumers will actively purchase HFCVs, and enterprises will eventually relevant information platforms, including promptly responding to con­
move toward producing HFCVs under the influence of government sumers’ feedback to alleviate their dissatisfaction when product-related
regulation and consumer purchasing behavior. problems arise. The more frequently companies interact with con­
The government, enterprises, and consumers are the three main ac­ sumers, the more engaged feedback in linked to their products or ser­
tors in the HFCV industry and their strategic choices influence each vices, and the more influence in spreading positive word-of-mouth. This
other. In other words, the evolutionary strategy of any one party is is the key to achieving long-lasting, high-quality development in the
influenced by the those of the other two parties. From the above con­ HFCV market.
clusions, the following insights were obtained:
(1) The government should clarify how it will influence its rela­ 5.2. Research limitations and future directions
tionship with enterprises in HFCV development based on the following
four points. In this study, we have made simulations of HFCV industry, and have
① The government’s regulation mechanism should improve and drawn some important conclusions, but some limitations remain. First,
enhance the supervision of enterprises fraudulent behavior, including the assumptions on the HFCV industry are more desirable. Second,
increasing the fines for enterprises that do not meet environmental because of the limitation of the parameters of the evolutionary game
standards. For enterprises in severe circumstances, the government may payment matrix, all the factors affecting the development of the HFCV
order them to stop production and rectification. industry are not considered. Third, infrastructure construction only
② An effective penalty mechanism should be established with high considers hydrogen refueling stations and ignores hydrogen production
rewards and penalties for meeting and failing to meet emission stan­ and transportation. Fourth, because of the limitation of the investigation
dards, thus promoting emission technology improvement. In particular, objects and conditions, the parameter assignment of the payment matrix
the government should introduce more environmental regulatory laws only indicates the general state and lacks actual data support. Finally,
to increase the violation cost, decrease the compliance cost, and evolutionary behavior changes are mainly discussed in the context of
strengthen implementation, so that more enterprises will choose to changes in government regulation, not in the context of changes in
produce HFCVs. business or consumer behavior.
③ The government should continue to play a leading role in pro­ There are several possible directions for further research. Although
moting HFCV R&D and developing science and technology support HFCV development is encouraged by the government, the HFCV in­
policies conducive to the growth of the HFCV industry. Simultaneously, dustry in China is still complex and dynamic. There is still much room
we should accelerate the revolutionary breakthrough of HFCV, promote for progress, and it will be interesting to analyze the behavior of
HFCV core technology, and form an industry-university-research-and- stakeholders in the HFCV market, given the predictions of the enormous
application system. Furthermore, we should strengthen international expected growth of the industry. After the rapid development of the
technical cooperation, enthusiastically join the co-construction of HFCV industry, actual data support is available to increase the

12
W. Wang and J. Li Fuel 348 (2023) 128223

credibility of the EGT model. Finally, the future EGT analysis of the [17] Hao H, Mu Z, Liu Z, Zhao FQ. Abating transport GHG emissions by hydrogen fuel
cell vehicles: chances for the developing world. Fron Energy 2018;12(3):466–80.
HFCV industry, can be made either from the perspective of companies or
https://doi.org/10.1007/s11708-018-0561-3.
from the perspective of consumers. [18] Han W, Zhang G, Xiao J, Bénard P, Chahine R. Demonstrations and marketing
strategies of hydrogen fuel cell vehicles in China. Int J Hydrogen Energ 2014;39
(25):13859–72. https://doi.org/10.1016/j.ijhydene.2014.04.138.
Funding [19] Lan H, Hao D, Hao W, He Y. Development and comparison of the test methods
proposed in the Chinese test specifications for fuel cell electric vehicles. Energy Rep
This research was funded by the Beijing Natural Science Foundation 2022;8:565–79. https://doi.org/10.1016/j.egyr.2022.02.006.
[20] Kovač A, Paranos M, Marciuš D. Hydrogen in energy transition: A review. Int
(9224037); funded by the Fundamental Research Funds for the Central
JHydrogen Energ 2021;46(16):10016–35. https://doi.org/10.1016/j.
Universities (2022FR004). ijhydene.2020.11.256.
[21] Salvi BL, Subramanian KA. Sustainable development of road transportation sector
using hydrogen energy system. Renew Sust Energ Rev 2015;51:1132–55. https://
CRediT authorship contribution statement doi.org/10.1016/j.rser.2015.07.030.
[22] Chang X, Ma T, Wu R. Impact of urban development on residents’ public
Wei Wang: Conceptualization, Methodology, Writing – original transportation travel energy consumption in China: An analysis of hydrogen fuel
cell vehicles alternatives. Int J Hydrogen Energ 2019;44(30):16015–27. https://
draft. Jinzhong Li: Conceptualization, Writing – review & editing, doi.org/10.1016/j.ijhydne.2018.09.099.
Supervision. [23] Hosseini SE. Hydrogen has found its way to become the fuel of the future. Future.
Energy 2022;1(2):11–2. https://doi.org/10.55670/fpll.fuen.1.3.2.
[24] Habib AKMRR, Butler K. Environmental and economic comparison of hydrogen
Declaration of Competing Interest fuel cell and battery electric vehicles. Future Technol 2022;1(3): 25-33. https://do
i.org/10.55670/fpll.futech.1.2.3.
[25] China Hydrogen Alliance. China’s hydrogen energy and fuel cell industry white
The authors declare that they have no known competing financial paper, http://www.h2cn.org.cn/Uploads/File/2019/07/25/u5d396adeac15e.pdf;
interests or personal relationships that could have appeared to influence 2019 [accessed 29 June 2022].
[26] Fuel Cell & Hydrogen Energy Association. ROAD MAP TO A US HYDROGEN
the work reported in this paper.
ECONOMY, https://static1.squarespace.com/static/53ab1feee4b0bef0179a1563/
t/5e7ca9d6c8fb36 29d399fe0c/1585228263363/Road+Map+to+a+US+
References Hydrogen+Economy+Full+Report.pdf;2019 [assessed 21 June 2022].
[27] Automotive business review. China, the United States, Japan and South Korea
hydrogen energy and fuel cell policy comparison, https://d1ev.com/kol/92316;
[1] Okolie JA, Patra BR, Mukherjee A, Nanda S, Dalai AK, Kozinski JA. Futuristic
2019 [accessed 20 June 2022].
applications of hydrogen in energy, biorefining, aerospace, pharmaceuticals and
[28] Guotai Junan Securities Research. Japan’s hydrogen strategy and the advent of the
metallurgy. Int J Hydrogen Energ 2021;46(13):8885–905. https://doi.org/
fuel cell era.https://finance.sina.com.cn/stock/stockzmt/2019-03-26/doc-
10.1016/j.ijhydene.2021.01.014.
ihsxncvh5596144.shtml;2019 [accessed 26 June 2022].
[2] Esily RR, Chi Y, Ibrahiem DM, Chen Y. Hydrogen strategy in decarbonization era:
[29] Japan’s Ministry of Economy, Trade and Industry. Basic hydrogen strategy,
Egypt as a case study. Int J Hydrogen Energ 2022;47(43):18629–47. https://doi.
https://www.meti.go,jp/press/2017/12/20171226002/20171226002.html ;2019
org/10.1016/j.ijhydene.2022.04.012.
[accessed 20 June 2022].
[3] Chapman A, Itaoka K, Farabi-Asl H, Fujii Y, Nakahara M. Societal penetration of
[30] Fuel Cells and Hydrogen Joint Undertaking (FCHJU). Hydrogen Roadmap Europe:
hydrogen into the future energysystem: impacts of policy, technology and carbon
A sustainable pathway for the European Energy Transition, https://www.fch.
targets. Int J Hydrogen Energ 2020;45(7). https://doi.org/10.1016/j.
europa.eu/news/hydrog en-roadmap-europe-sustainable-pathway-european-
ijhydene.2019.12.112.
energy-transition; 2019[accessed 26 June 2022].
[4] Hou C, Wang B, Murugadoss V, Vupputuri S, Chao Y, Guo Z, Wang C, Du W. Recent
[31] Deloitte. Powering the Future of Mobility: Hydrogen Energy and Fuel Cell Mobility
advances in Co3O4 as anode materials for high-performance lithium-ion batteries.
Solutions, https://www2.deloitte.com/cn/zh/pages/about-deloitte/articles/pr-
Eng Sci 2020;11(5):19–30. https://doi.org/10.30919/es8d1128.
fcevs-to-be-cheaper-to-run-than-bevs-and-ice-vehicles-within-10-years.html; 2020
[5] Hu X, Wu H, Liu S, Gong S, Du Y, Li X, Lu X, Qu J. Fabrication of organic shape-
[accessed 20 June 2022].
stabilized phase change material and its energy storage applications. Eng Sci 2021;
[32] Chen WG. China Hydrogen Energy Market Annual Data Interpretation 2022-2023,
17:1–27. https://doi.org/10.30919/es8d474.
http://goodsfu.10jqka.com.cn/20230106/c644100915.shtml;2023 [accessed 27
[6] Huang J, Luo Y, Weng M, Yu J, Sun L, Zeng H, Liu Y, Zeng W, Min Y, Guo Z.
February 2023].
Advances and applications of phase change materials (PCMs) and PCMs-based
[33] 2022 sales up 138.90% year-on-year! Learn about China’s fuel cell vehicle
technologies. ES Mater Manuf 2021;13(23–39). https://doi.org/10.30919/
production and sales in 2022 in one article, https://h2.in-en.com/html/h2-
esmm5f458.
2421418.shtml;2023 [accessed 27 February 2023].
[7] Xie Y, Yang Y, Liu Y, Wang S, Guo X, Wang H, et al. Paraffin/polyethylene/graphite
[34] Sohu. Hydrogen energy is included in the Government’s Work Report for the first
composite phase change materials with enhanced thermal conductivity and
time, https://www.sohu.com/a/302100209_120046996; 2019 [accessed 27
leakage-proof. Adv Compos Hybrid Ma 2021;4(3):543–51. https://doi.org/
February 2023].
10.1007/s42114-021-00249-6.
[35] National Energy Administration. Announcement of the National Energy
[8] Zhao J, Wei D, Zhang C, Shao Q, Murugadoss V, Guo Z, Jiang Q, Yang X. An
Administration on the Public Consultation on the Energy Law of the People’s
overview of oxygen reduction electrocatalysts for rechargeable zinc-air batteries
Republic of China (Draft for Comments), http://www.nea.gov.cn/2020-04/10/c_
enabled by carbon and carbon composites. Eng Sci 2021;15:1–19. https://doi.org/
138963212.htm; 2020 [accessed 27 February 2023].
10.30919/es8d420.
[36] Xinhua News Agency. Opinions of the Central Committee of the Communist Party
[9] Zhou Y, Wu S, Ma Y, Zhang H, Zeng X, Wu F, Liu F, Ryu JE, Guo Z. Recent advances
of China (CPC) and the State Council on the complete and accurate implementation
in organic/composite phase change materials for energy storage. ES Energy
of the new development concept and the proper implementation of carbon peaking
Environ 2020;9:28–40. https://doi.org/10.30919/esee8c150.
and carbon neutral work, http://www.gov.cn/zhengce/2021-10/24/content_
[10] Hosseini SE, Butler B. An overview of development and challenges in hydrogen
5644613.htm; 2021[accessed 27 February 2023].
powered vehicles. Int J Green Energy 2020;17(1):13–37. https://doi.org/10.1080/
[37] National Energy Administration. The Medium and Long-Term Plan for the
15435075.2019.1685999.
Development of the Hydrogen Energy Industry (2021-2035), http://zfxxgk.nea.
[11] Apak S, Atay E, Tuncer G. Renewable hydrogen energy and energy efficiency in
gov.cn/2022-03/23/c_1310525630.htm; 2020 [accessed 23 June 2022].
Turkey in the 21st century. Int J Hydrogen Energ 2017;42(4):2446–52. https://doi.
[38] Ministry of Finance of the People’s Republic of China. Circular of the Ministry of
org/10.1016/jijhydene.2016.05.043.
Finance on Local Budget and Account Disclosure and Special Inspection of Subsidy
[12] Capurso T, Stefanizzi M, Torresi M, Camporeale SM. Perspective of the role of
Funds for the Promotion and Application of New Energy Vehicles, http://www.
hydrogen in the 21st century energy transition. Energ Convers Manage 2022;251:
gov.cn/xinwen/2016-09/08/content_5106603.htm ; 2016 [accessed 21 June
114898. https://doi.org/10.1016/j.enconman.2021.114898.
2022].
[13] Damman S, Sandberg E, Rosenberg E, Pisciella P, Graabak I. A hybrid perspective
[39] Wang G, Chao Y, Cao Y, Jiang T, Han W, Chen Z. A comprehensive review of
on energy transition pathways: Is hydrogen the key for Norway? Energy Res Soc Sci
research works based on evolutionary game theory for sustainable energy
2021;78:102116. https://doi.org/10.1016/j.erss.2021.102116.
development. Energy Rep 2022;8:114–36. https://doi.org/10.1016/j.
[14] Li W, Long R, Chen H, Chen F, Zheng X, He Z, et al. Willingness to pay for hydrogen
egyr.2021.11.231.
fuel cell electric vehicles in China: A choice experiment analysis. Int J Hydrogen
[40] Tuofu H, Changhao H, Qingyun H, Dongxiao Y, Tian H, Yi F. A coordination
Energ 2020;45(59):34346–53. https://doi.org/10.1016/j.ijhydene.2020.01.046.
analysis of stakeholder interests on the new subsidy policy of hydrogen fuel cell
[15] Luo Y, Wu Y, Li B, Mo T, Li Y, Feng SP, et al. Development and application of fuel
vehicles in China: From the perspective of the evolutionary game theory. Int J
cells in the automobile industry. J Energy Storage 2021;42:103124. https://doi.
Hydrogen Energ 2022;47(58):24493–510. https://doi.org/10.1016/j.
org/10.1016/j.est. 2021.103124.
ijhydene.2022.02.218.
[16] Shao L, Li X, Wu S. Medium and Long-term Sales Forecast of Hydrogen Fuel Cell
[41] Jones J, Genovese A, Tob-Ogu A. Hydrogen vehicles in urban logistics: a total cost
Vehicles (HFCV) Based on System Dynamics and Discrete Selection Model.E3S Web
of ownership analysis and some policy imply cations. Renew Sust Energ Rev 2020;
of Conferences EDP Sciences 2020; 218: 02037. https://doi.org/10.1051/e3s
119:109595. https://doi.org/10.1016/j.rser.2019.109595.
conf/202021802037.

13
W. Wang and J. Li Fuel 348 (2023) 128223

[42] Zhao F, Mu Z, Hao H, Liu Z, He X. Hydrogen fuel cell vehicle development in China: [60] Greene DL, Lin Z, Dong J. Analyzing the sensitivity of hydrogen vehicle sales to
An industry chain perspective. Energy Technol 2020;8:2000179. https://doi.org/ consumers’ preferences. Int J Hydrogen Energ 2013;38(36):15857–67. https://doi.
10.1002//ente.202000179. org/10.1016/j.ijhydene.2013.08.099.
[43] Li Y, Kimura S. Economic competitiveness and environmental implications of [61] Southall GD, Khare A. The feasibility of distributed hydrogen production from
hydrogen energy and fuel cell electric vehicles in ASEAN countries: The current renewable energy sources and the financial contribution from UK motorists on
and future scenarios. Energ Policy 2021;148:111980. https://doi.org/10.1016/j. environmental grounds. Sustain Cities Soc 2016;26:134–49. https://doi.org/
enpol.2020.111980. 10.1016/j.scs.2016.05.009.
[44] Hu M. The current status of hydrogen and fuel cell development in China. [62] Zhang L, Yu J, Ren J, Ma L, Zhang W, Liang H. How can fuel cell vehicles bring a
J Electrochem Energy Convers Storage 2020;17(3). https://doi.org/10.1115/ bright future for this dragon? Answer by multi-criteria decision making analysis.
1.4045702. Int J Hydrogen Energ 2016;41(39):17183–92. https://doi.org/10.1016/j.
[45] Wu Y, Liu F, He J, et al. Obstacle identification, analysis and solutions of hydrogen ijhydene.2016.08.044.
fuel cell vehicles for application in China under the carbon neutrality target. Energ [63] Kar SK, Bansal R, Harichandan S. An empirical study on intention to use hydrogen
Policy 2021;159:112643. https://doi.org/10.1016/j.enpol.2021.112643. fuel cell vehicles in India. Int J Hydrogen Energ 2022;47(46):19999–20015.
[46] Ambrose AF, Rasiah R, Al-Amin AQ, Chen Z. Introducing hydrogen fuel cell https://doi.org/10.1016/j.ijhydene.2022.04.137.
vehicles in Malaysia. J Asia Pac Econ 2018;23(2):263–78. https://doi.org/ [64] Rosales-Tristancho A, Brey R, Carazo AF, Brey JJ. Analysis of the barriers to the
10.1080/13547 860.2018.1442151. adoption of zero-emission vehicles in Spain. Transport Res Part A-Pol 2022;158:
[47] Trencher G, Edianto A. Drivers and barriers to the adoption of fuel cell passenger 19–43. https://doi.org/10.1016/j.tra.2022.01.016.
vehicles and buses in Germany. Energies 2021;14(4):833. https://doi.org/ [65] Wang S, Fan J, Zhao D, Wu Y. The impact of government subsidies or penalties for
10.3390/en14040833. new-energy vehicles a static and evolutionary game model analysis. J Transp Econ
[48] Itaoka K, Saito A, Sasaki K. Public perception on hydrogen infrastructure in Japan: Policy 2015;49(1):98–114. https://www.jstor.org/stable/jtranseconpoli.49.1.
influence of rollout of commercial fuel cell vehicles. Int J Hydrogen Energ 2017;42 0098.
(11):7290–6. https://doi.org/10.1016/j.ijhydene.2016.10.123. [66] Guan X, Zhang G, Liu D, Tan X, Wu D. The behavior of consumer buying new
[49] Lipman TE, Elke M, Lidicker J. Hydrogen fuel cell electric vehicle performance and energy vehicles based on stochastic evolutionary game. Filomat 2016;30(15):
user-response assessment: Results of an extended driver study. Int J Hydrogen 3987–97. https://doi.org/10.2298/FI L1615987G.
Energ 2018;43(27):12442–54. https://doi.org/10.1016/j.ijhydene.2018.04.172. [67] Liu C, Huang W, Yang C. The evolutionary dynamics of China’s electric vehicle
[50] Li H, Welsh R, Morris A. Emergency responders’ perceptions of Hydrogen Fuel Cell industry–Taxes vs. subsidies. Comp Ind Eng 2017;113:103–22. https://doi.org/
Vehicle: A qualitative study on the UK fire and rescue services. Int J Hydrogen 10.1016/j.cie.2017.08.026.
Energ 2021;46(65):32750–61. https://doi.org/10.1016/j.ijhydene.2021.07.146. [68] Ji S, Zhao D, Luo R. Evolutionary game analysis on local governments and
[51] Khan U, Yamamoto T, Sato H. Understanding attitudes of hydrogen fuel-cell manufacturers’ behavioral strategies: Impact of phasing out subsidies for new
vehicle adopters in Japan. Int J Hydrogen Energ 2021;46(60):30698–717. https:// energy vehicles. Energy 2019;189:116064. https://doi.org/10.1016/j.
doi.org/10.1016/j.ijhydene.2021.06.195. energy.2019.116064.
[52] Khan U, Yamamoto T, Sato H. An insight into potential early adopters of hydrogen [69] Sun H, Wan Y, Lv H. System dynamics model for the evolutionary behaviour of
fuel-cell vehicles in Japan. Int J Hydrogen Energ 2021;46(18):10589–607. https:// government enterprises and consumers in China’s new energy vehicle market.
doi.org/10.1016/j.ijhydene.2020.12.173. Sustainability 2020;12(4):1578. https://doi.org/10.3390/su12041578.
[53] Apostolou D, Welcher SN. Prospects of the hydrogen-based mobility in the private [70] Selten R. A note on evolutionarily stable strategies in asymmetric animal conflicts.
vehicle market. A social perspective in Denmark. Int J Hydrogen Energ 2021;46(9): J Theor Biol 1988;84(1):93–101.
6885–900. https://doi.org/10.1016/j.ijhydene.2020.11.167. [71] Ritzberger K, Weibull JW. Evolutionary selection in normal-form games.
[54] Chen M, Zhang WH. Purchase intention for hydrogen automobile among Chinese Econometrica: J Econom Soc 1995;63(6):1371–99.
citizens: The influence of environmental concern and perceived social value. Int J [72] Wainwright J. A dynamical systems approach to Bianchi cosmologies: orthogonal
Hydrogen Energ 2021;46(34):18000–10. https://doi.org/10.1016/j. models of class A. Class Quant Grav 1989;6(10):1409. https://doi.org/10.1088/
ijhydene.2020.11.099. 0264-9381/6/10/011.
[55] O’Garra T, Mourato S, Pearson P. Analysing awareness and acceptability of [73] Friedman D. Evolutionary games in economics. Econometrica: J Econom Soc 1991:
hydrogen vehicles: A London case study. Int J Hydrogen Energ 2005;30(6):649–59. 637–66. https://doi.org/10.2307/2938222.
https://doi.org/10.1016/j.ijhydene.2004.10.008. [74] Lyapunov AM. The general problem of the stability of motion. Int J Control 1992;
[56] Nie X, Wei X, Xia Q, Zhou M. Customers’ purchase intention for hydrogen vehicle 55(3):531–4. https://doi.org/10.1080/00207179208934253.
and industrial agglomeration: Evidence from Jiangsu Province. China I Int J [75] Yang Y, Yang W, Chen H, Li Y. China’s energy whistleblowing and energy
Hydrogen Energ 2021;46:18011–9. https://doi.org/10.1016/j. supervision policy: An evolutionary game perspective. Energy 2020;213:118774.
ijhydene.2020.10.055. https://doi.org/10.1016/j.energy.2020. 118774.
[57] Stotts R, Lopez-Jaramillo OG, Kelley S, Krafft A, Kuby M. How drivers decide [76] Zhao T, Liu Z, Jamasb T. Developing hydrogen refueling stations: An evolutionary
whether to get a fuel cell vehicle: An ethnographic decision model. Int J Hydrogen game approach and the case of China. Energ Econ 2022;115:106390. https://doi.
Energ 2021;46(12):8736–48. https://doi.org/10.1016/j.ijhydene.2020.12.042. org/10.1016/j.eneco.2022.106390.
[58] Hardman S, Chandan A, Shiu E, Steinberger-Wilckens R. Consumer attitudes to fuel [77] Yang W, Yang Y, Chen H. How to stimulate Chinese energy companies to comply
cell vehicles post trial in the United Kingdom. Int J Hydrogen Energ 2016;41(15): with emission regulations? Evidence from four-party evolutionary game analysis.
6171–9. https://doi.org/10.1016/j.ijhydene.2016.02.067. Energy 2022;258:124867. https://doi.org/10.1016/j.energy.2022.124867.
[59] Park SY, Kim JW, Lee DH. Development of a market penetration forecasting model [78] Xu JZ, Sun Y. Evolutionary game research on cooperative innovation of new energy
for Hydrogen Fuel Cell Vehicles considering infrastructure and cost reduction vehicle industry under market mechanism and government regulation. Oper Res
effects. Energ Policy 2011;39(6):3307–15. https://doi.org/10.1016/j. Manag Sci 2020;29(5):143–51. https://doi.org/10.12005/orms.2020.0128.
enpol.2011.03.021. [79] Peng P, He XT. Government subsidies and development of new energy vehicle
industry—evolutionary game analysis based on system dynamics. Oper Res Manag
Sci 2021;30(10):31–8. https://doi.org/10.12005/orms.2021.0310.

14