Energy 295 (2024) 131069

Contents lists available at ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

Analysis of the influence of electric flywheel and electromechanical

flywheel on electric vehicle economy
Binbin Sun a, Bo Li a, *, Jilei Xing b, Xiao Yu c, Mengxue Xie a, Zihao Hu a
a
School of Transportation and Vehicle Engineering, Shandong University of Technology, Zibo, 255000,China
b
School of Vehicle and Mobility, Tsinghua University, Beijing, 100084, China
c
National Engineering Research Center of Electric Vehicles, Beijing Institute of Technology, Beijing, 100081, China

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

Handling editor: X Ou Improving energy utilization efficiency to extend the range of vehicle is the common issue concerned by various
forms of electric vehicles. In order to reveal the influence of electric flywheel and electromechanical flywheel on
Keywords: vehicle economy, two kinds of hybrid energy systems are studied. Firstly, based on the operating characteristics
Electric flywheel of the two types of flywheels, the topology schemes of the two hybrid energy systems are designed. On this basis,
Electromechanical flywheel
to make full use of the advantages of electric flywheel and electromechanical flywheel, energy management
Composite energy system
strategies based on wavelet algorithm and logic threshold are designed. Finally, economy tests of the two hybrid
Energy management
Vehicle economy energy systems are conducted. Results show that compared with the single energy scheme with lithium battery,
under CLTC, as the control motor of the electric flywheel operates under high speed and low torque range
frequently, the energy consumption improvement of lithium battery is not enough to compensate for the flywheel
energy loss. The net loss of the lithium battery-electric flywheel energy system increases by 2.61%. Profit from
efficiency improvement of lithium battery system, increase of regenerative energy recovery and better efficiency
of main drive motor, the net loss of the lithium battery-electromechanical flywheel energy system decreases by
6.44%.

source plays an important role in the optimization of the main one [13,
1. Introduction 14]. By parameters design of the composite energy system and coordi­
nated control of auxiliary energy source, the negative influence of
1.1. Introduction of research background vehicle conditions on main energy source is able to be reduced. Owing to
this, stable and efficient operation of main energy source is achieved
Improving utilization efficiency of on-board energy remains the [15–17]. The second method is the apply of distributed drive [18,19]. To
focus of the automotive industry [1,2]. Optimization design of on-board improve the efficiency of power system, two or more motors are used to
energy system plays an important role in the improvement of vehicle improve the matching degree between power system and vehicle oper­
energy efficiency [3]. Different topology schemes of energy and power ations [20,21]. Parameters design and coordinated control of the
systems require different optimization design of on-board energy. As for distributed motors are essential [22,23].
the fuel vehicle, development of efficient engine is essential. Technol­
ogies, such as advanced intake and exhaust system design, efficient 1.2. Introduction of flywheel system
thermodynamic cycle design and so on, have been developed [4–6]. To
improve the economy of hybrid vehicle, different parameters design and High power density and long life are the advantages of flywheel
power distribution methods of engine and motor have been proposed energy storage device [24,25]. Based on the configuration characteris­
[7–10]. Consequently, mature and efficient hybrid systems have been tics of different flywheels, auxiliary power source or energy source has
developed and used. been developed and used [26]. Owing to the high power and fast
For pure electric vehicle, technical ways to improve the efficiency of response intervention, flywheel devices are able to effectively reduce the
on-board energy system can be summarized into two categories. The impact of vehicle load change on main power source or main energy
first one is the use of composite energy system [11,12]. Auxiliary energy source [27,28]. Based on the perspective of structure and energy flow,

* Corresponding author.
E-mail address: [email protected] (B. Li).

https://doi.org/10.1016/j.energy.2024.131069
Received 16 October 2023; Received in revised form 11 February 2024; Accepted 20 March 2024
Available online 25 March 2024
0360-5442/© 2024 Elsevier Ltd. All rights reserved.
B. Sun et al. Energy 295 (2024) 131069

and patented magnetically loaded composite are used to ensure efficient

Nomenclature and safe operation of the high speed flywheel [24]. Consequently, the
cost is very high, influencing its application on vehicle.
CVT Continuously variable transmission Mechanical flywheel achieves energy transfer directly by CVT. As the
SOE State of energy scheme has no energy conversion, it shows the advantages of high ef­
SOC State of charge ficiency and compact structure. Mechanical flywheel device shows
CLTC China light-duty vehicle test cycle perfect energy-saving effect on vehicle, which has been proven [33].
PGS Planetary gear set Taking the famous Flybrid flywheel system for example, the flywheel is
VCU Vehicle control unit connected to the transmission of the vehicle via a CVT. By manipulation
HIL Hardware-in-the-loop testing of the CVT ratio, the energy storage and recovery of the flywheel are
HWFET Highway fuel economy test achieved. Specially, the flywheel energy is stored when the ratio is
UUDS Urban dynamometer driving schedule changed so as to speed up the flywheel. Similar to this, energy recovery
FTS Flywheel and transmission shaft is achieved when the ratio is changed so as to slow down the flywheel. In
RTS Real-time simulator Jaguar XF and Volvo Car, Flybrid flywheel system running at over 60000
WHP Williams Hybrid Power Limited r.min− 1 to provides an additional high power to dramatically improve
acceleration. More importantly, a 20% fuel economy improvement is
achieved for both the Jaguar and Volvo applications [24]. However,
apart from high cost caused by the very high rotational speed of
flywheel devices can be divided into three types, that is the electric flywheel, due to the input and output control of mechanical energy, this
flywheel, the mechanical flywheel and the electromechanical flywheel, scheme can only be used as an auxiliary power source to provide
as shown Fig. 1 [29–31]. short-term but high-power regulation for main power engine. In addi­
As for electric flywheel, a built-in motor with high design speed is tion, due to the range limitation of speed ratio adjustment of CVT, me­
used to achieve energy conversion between kinetic energy and electrical chanical flywheel is unable to realize deep release and feedback of
one. As the scheme is able to realize the input and output control of flywheel energy, which limits the energy-saving effect and application
electric energy, it is usually used as an auxiliary energy to adjust and of the device [34]. At present, the mechanical flywheel device is only
optimize the efficiency of main energy source [32]. Taking the famous applied to fuel vehicle.
WHP flywheel system for example, the system uses the electrical pow­ High speed flywheel design can indeed ensure energy storage for
ertrain with motors and generators, rather than a mechanical power­ electric and mechanical flywheel systems, while accompanied by high
train. With the use of WHP flywheel, Porsche 911 GT3 R shows cost challenge. The special characteristic of vehicle urban condition,
significant economy performance by regenerative braking that achieves frequent brake, provides feasibility for low-speed flywheel design.
25% improvement of fuel economy. Furthermore, at full charge, the Through frequent braking energy recovery, the flywheel is able to
electric flywheel is able to deliver a total of 120 kW for about 6–8 s to maintain energy supply for vehicle. With the help of PGS, the electro­
accelerate the car. However, in order to maximize the energy storage as mechanical flywheel device is able to achieve input and output control
much as possible, the maximum speed of the flywheel rotor is approx­ of both electric and kinetic energy. The electric energy control is ach­
imately 40000 r.min− 1. Hybrid ceramic ball bearings, vacuum chamber ieved by speed control motor. The kinetic energy control is realized by

Fig. 1. Structure and energy flow of different types of flywheel systems.

2
B. Sun et al. Energy 295 (2024) 131069

flywheel. Consequently, as shown in Table 1, compared with the other What are the efficiency influences of it on main energy source, main
two types of flywheels, electromechanical flywheel device shows greater power source and composite energy system?
freedom for design, control and application. By taking the advantage of Secondly, as the flywheel device has itself energy loss issue, economy
motor vector control, the device can accurately control the bidirectional analysis from the perspective of the composite energy system is neces­
power and energy depth of flywheel. Therefore, the electromechanical sary. Specifically, what is the characteristic of the energy loss of flywheel
flywheel device can be designed as an energy source to adjust and device? How the energy loss effects the system economy? Does the en­
optimize the working efficiency of main energy source. Moreover, it can ergy loss is greater or less than the energy optimization of the main
also be used as a power source to optimize the working state and effi­ energy source?
ciency of main power source. Finally, as the energy management strategy based on rule-based,
global optimization and local optimization algorithms are unable to
1.3. Introduction of energy management strategy balance energy optimization and real-time performance, based on
existing hardware conditions, how to design an appropriate energy
The core of energy management is to develop efficient energy allo­ management strategy for the composite energy system that can achieve
cation control strategy that matches the topology of the vehicle energy engineering application with best energy optimization？
and power system under different working conditions [35,36]. It is the
key to ensuring the efficient performance of vehicle. According to the 1.5. Introduction of novelty and creativity in this paper
different algorithms used, the current energy management strategies for
vehicle hybrid systems can be summarized into three types: strategy Given the above analysis, this paper provides a deep analysis on the
based on rule-based algorithm, global optimization algorithm, and local influence of electric flywheel and electromechanical flywheel on electric
optimization algorithm [37–39]. vehicle economy. This paper attempts to make three notable contribu­
The energy management strategy based on rule-based algorithm uses tions and improvements to the current study, as shown in the followings:
given mode switching rule by experience to control energy. It has the Specially structured electromechanical flywheel device was
characteristics of simple algorithm structure, strong real-time perfor­ designed: To solve the challenges of limited energy control range and
mance, but relatively poor energy efficiency [40]. The energy manage­ extremely high cost of electric and mechanical flywheel systems, the
ment strategy based on global optimization algorithms, such as genetic electromechanical flywheel based on planetary gear was designed. The
algorithm, population algorithm and so on, are able to deeply explore device is able to act as both auxiliary energy source and power source,
the potential of efficient energy utilization, and shows perfect energy thereby, can realize the efficiency improvement of lithium battery and
efficiency. However, as the algorithm is complex, poor real-time per­ main drive motor. The new flywheel system is able to achieve flywheel
formance restricts its practical engineering applications [41]. The last energy control from 0 to 100% by a speed regulating motor and plan­
type of energy management focus on local optimization, by using dy­ etary gear. Furthermore, it can achieve frequent braking energy recov­
namic programming, can solve the efficient application of system energy ery. Thereby, the maximum design speed of flywheel is reduced. Due to
within limited time interval working conditions [42]. The algorithm still the reduction in the maximum design speed, it is unnecessary to use
has very high hardware requirement for the controller. expensive designs such as vacuum chamber and magnetic levitation
bearing, which is conducive to the promotion and application of it.By
1.4. Introduction of proposing research question using the new flywheel system, under CLTC cycle, compared with the
electric flywheel scheme, the net consumption of the lithium battery and
In general, compared with electric and mechanical flywheel systems, the whole hybrid system decreases by 8.05% and 113.88%.
the electromechanical flywheel shows significant advantages on low Energy management strategy based on two-stage power split
cost and easy application. However, as the electromechanical flywheel was proposed: To fully leverage the fast-response and high-power ad­
has special structure and energy flow, there are still unresolved issues. vantages of electric flywheel, an energy management based on two-stage
First of all, since the new scheme uses steel bearing and atmospheric power split was proposed. The first stage was designed by using rule-
environment, does it really have advantage in efficiency optimization? based control strategy to achieve the basic power allocation. The sec­
ond one was proposed with the use of wavelet algorithm to decompose
Table 1 the high and low frequency of the power required by vehicle. Compared
Comparison of different types of flywheels. to the rule-based strategy, the new energy management strategy shows
Representative Electric Mechanical Electromechanical greater advantage on improvement of lithium battery performance
schemes flywheel of WHS flywheel of flywheel proposed by while balances real-time performance. Under CLTC cycle, the proportion
Flybrid this paper of working conditions of electric flywheel increases significantly.
The input and Electric drive Mechanical Both electric and Consequently, the peak power and current of lithium battery decrease
output paths of system transmission mechanical systems by 82.67% and 90.43%. The average drive power and current decrease
flywheel energy system by 18.18% and 23.56%. The average efficiency improves by 1.42%.
Operating Vacuum Vacuum Atmospheric
Specific impacts of flywheel on electric vehicle economy were
conditions of environment environment environment
flywheel revealed: For the first time, from the perspective of vehicle economy,
Energy regulation Limited by gear 75%–100% 0–100% the impact of different types of flywheel systems on the efficiency of the
amplitude ratio entire energy system was analyzed. The shortcomings and advantages of
Maximum 40000–45000 64500 25000 them were identified. Research results provide important reference
flywheel speed
(r.min− 1)
value for the design and vehicle application of different flywheel sys­
Flywheel material Patented Carbon fiber Metal tems. The electric flywheel is able to regulate the charge and discharge
magnetically composite current of lithium battery by instantaneous and high-power interven­
loaded material tion, thus improving lithium battery efficiency. However, due to the
composite
frequent operation of electric flywheel in the high speed and low torque
Requirements for Ceramic bearing Hybrid Steel bearings
bearing ceramic ball range, the efficiency of control motor is very low. The long-time and
bearings low-efficiency operation of the control motor will not only affect the
Difficulty in Very high Very high Relatively low efficiency of electric flywheel device, but also have a negative impact on
promoting the whole hybrid energy system. On the perspective of vehicle economy,
application
there are two key issues need to be solved urgently for the application of

3
B. Sun et al. Energy 295 (2024) 131069

the electric flywheel to electric vehicle. Deign to improve the efficiency {

of control motor is essential, or new method is required to avoid the Tc : Ts : Tr = (1 + k) : − 1 : − k
(2)
Pc : Ps : Pr = (nr rr + ns rs ) : − ns rs : − nr rr
long-time operation of control motor in low efficiency range.
Where Ts, Tc and Tr are the torque applied to sun gear, planetary frame
2. Topology scheme design of composite energy system
and gear ring by external forces; Ps, Pc and Pr are the power applied to
sun gear, planetary frame and gear ring; rs is the radius of sun gear; rr is
2.1. Lithium battery-electric flywheel composite energy system
the radius of gear ring.
As shown in Table 2, based on the planetary gear device, there are six
As the electric flywheel has the characteristic of high-power density
topological connection schemes between flywheel, speed control motor
but low-energy storage, it is designed as an auxiliary energy for the
and vehicle transmission system. To confirm the optimal one, the power
lithium battery. The topology of the lithium battery-electric flywheel
and torque relations between the three devices needs to be derived. The
energy system can be designed into four forms. Specially, the passive
power of sun gear and gear ring are assumed to be Ps and Pr. The torque
scheme, the semi-active scheme on flywheel side, the semi-active
of speed control motor is set as T. According to the kinematic and dy­
scheme on lithium battery side and the active scheme. Based on the
namic equations of planetary gear, the power and torque relationships
following reasons, the second topology scheme is adopted in this study,
for different topological design schemes is derived as shown Table 3 and
as shown in Fig. 2.
Table 4.
The first reason is that the voltage change range of the lithium bat­
Based on the derived result, it is clear that the power coupling output
tery is very small during working process, thus the bus voltage is rela­
of flywheel and speed control motor can be realized by using scheme 1
tively stable. As shown in Fig. 3, the voltage variation of lithium battery
and scheme 5. Thus, the output power is amplified to achieve greater
is presented under six CLTC cycles. The initial SOC of the test is 0.8.
power performance. Conversely, the other four schemes present the
During each test cycle, the difference between the maximum voltage and
opposite pattern. Under the premise of satisfying the same dynamic
the minimum one is about 4V. For the sixth cycle, compared with the
performance of vehicle, scheme 1 and scheme 5 are able to realize the
first one, the SOC changes from the initial 0.8 to 0.56. The maximum
miniaturization design of speed control motor and flywheel. Based on
voltage variation of is lithium battery about 8V.
the analysis, the other four schemes are excluded.
Secondly, by the design of input and output power flow without
Compared with flywheel, the speed control motor is able to realize
passing through DC/DC converter, the lithium battery energy loss can be
active regulation control. According to the torque relationship in Ta­
improved. The last one is that the voltage of electric flywheel device
bles 4 and it can be further deduced that, when the output torque of
changes significantly. By controlling working state of the bidirectional
speed control motor is T, the output shaft of scheme 1 achieves the
DC/DC, the bus voltages of the two energy sources are able to collabo­
maximum multiplication of T. Therefore, on the premise of meeting the
rate. Consequently, the power distribution of the hybrid system can be
same low speed climbing and acceleration performance of vehicle,
conducted based on energy management strategy, so that the purpose of
compared with scheme 5, scheme 1 can use a motor with smaller peak
fully utilizing the energy of the flywheel is achieved.
torque. The advantage is conducive to the miniaturization design of
electromechanical flywheel. In summary, scheme 1 is adopted is this
2.2. Lithium battery-electromechanical flywheel composite energy system
study. Fig. 4 shows the topology of the designed electromechanical
flywheel.
The electromechanical flywheel device mainly consists of flywheel,
As shown in Fig. 5, the topology scheme of the lithium battery-
speed control motor, planetary gear, flywheel brake, reverse locking
electromechanical flywheel composite energy system is given.
device and flywheel reducer. The planetary gear plays an important role
Different from electric flywheel, who can only take part in vehicle drive
as the connection of flywheel, speed control motor and vehicle trans­
or brake in the way of energy flow without mechanical coupling rela­
mission system. The kinematic equation between sun gear, planet carrier
tionship, the planet frame of electromechanical flywheel is mechanically
and gear ring of the planetary gear is as following:
coupled directly to vehicle transmission system. Consequently, the
ns + knr = (1 + k)nc (1) electromechanical flywheel is able to output power to vehicle under
drive mode and accept power from vehicle under regenerative brake
Where, ns, nc and nr mean sun gear speed, planetary frame speed and mode.
gear ring speed respectively; k is the characteristic parameter of plane­ Constrained by the overall dimension of electromechanical flywheel
tary gear. device, the peak torque of the speed control motor is limited. Therefore,
In this paper, it is assumed that the counterclockwise direction is the the electromechanical flywheel is unable to meet the requirements of
positive direction of torque. When the planetary gear device operates vehicle dynamic performance. To solve this problem, a main drive motor
under balanced state, the dynamic equation of the external forces is required. As the original vehicle is drived by rear motor, on the
applied to the planetary gear is derived as following: premise of meeting the installation space requirements, in this design,
the electromechanical flywheel device is arranged in the front drive

Fig. 2. Topology scheme of lithium battery-electric flywheel composite energy system.

4
B. Sun et al. Energy 295 (2024) 131069

Fig. 3. The voltage variation of lithium battery under CLTC.

3.1.1. Primary power split based on logical threshold

Table 2
Essential parameters are designed for the first stage. The critical
Connection methods for different topology solutions.
power, P1, is used as the threshold for mode switching. The SOE of
Scheme Speed control motor Flywheel Output shaft flywheel and SOC of lithium battery play important role in energy
position position position
management. The lower and upper limits of SOE are defined as SOEfwmin
1 Sun gear Gear ring Planet carrier and SOEfwmax. The referable state of eSOEfw1 and the lower limit SOCbat1
2 Sun gear Planet carrier Gear ring that can be charged by lithium battery to flywheel, are used as the
3 Planet carrier Sun gear Gear ring
4 Planet carrier Gear ring Sun gear
threshold for energy management. Specific control rules are as
5 Gear ring Sun gear Planet carrier followings.
6 Gear ring Planet carrier Sun gear
(1) When the required power of the main drive motor, Preq, is greater
than P1, if SOEfw > SOEfwmin, the lithium battery provides con­
Table 3 stant power P1 and the electric flywheel provides the part beyond
Power relations for different topology solutions. P1. If SOEfw ≤ SOEfwmin, the lithium battery provides all the
Scheme Speed control motor position Flywheel power Output shaft power
required power.
(2) When 0<Preq ≤ P1, if SOEfw < SOEfw1 and SOCbat > SOCbat1, it
1 Ps -Pr Pr + Ps
means that the electric flywheel needs to be charged while the
2 Ps Pr + Ps -Pr
3 Pr + Ps Ps Pr lithium battery is able to charge it. Thus, the lithium battery
4 Pr + Ps Pr Ps provides constant power P1. Part of it is used to meet the demand
5 Pr -Ps Pr + Ps of main drive motor. The rest is designed to charge the electric
6 Pr Pr + Ps -Ps
flywheel. If SOEfw < SOEfw1 and SOCbat ≤ SOCbat1, it indicates
that the lithium battery is unable to charge electric flywheel.
Consequently, the lithium battery provides all the required power
Table 4 and the flywheel system does not work.
Torque relations for different topology solutions. (3) When Preq = 0, both the lithium battery and the flywheel system
Scheme Speed control motor torque Flywheel torque Output shaft torque do not work.
1 T -kT (1+k)T
(4) When Preq <0, the main drive motor works in generator state. The
2 T (1+k)T -kT maximum charging power of flywheel system is limited by the
3 T T/(1+k) kT/(1+k) peak power of the control motor. If the absolute generating power
4 T kT/(1+k) T/(1+k) of the main drive motor is less than or equal to the peak power of
5 T -T/k (1+k)T/k
flywheel control motor Pmotf_max, it means that the generating
6 T (1+k)T/k -T/k
power is relatively small. Under this condition, if SOEfw <
SOEfw1, in order to quickly supplement the energy of flywheel to
axle. The rear drive axle still uses the original drive scheme. prepare for later driving condition, the output power from
lithium battery is set as P1 to charge flywheel together with the
3. Design of energy management strategy power generation from main drive motor. If SOEfw1≤SOEfw <
SOEfwmax, only the flywheel system bears the power generation
3.1. Lithium battery-electric flywheel energy system and recovers the brake energy. If SOEfw ≥ SOEfwmax, only the
lithium battery takes up the power generated by the main drive
In order to give full play to the advantages of instantaneous and high- motor and recovers the brake energy.
power response of electric flywheel, as shown in Fig. 6, an energy (5) If Preq is less than zero and its absolute value is greater than
management strategy based on two-stage power split by using logic Pmotf_max, it indicates that the power of the main drive motor is
threshold control and wavelet algorithm is proposed. With the use of relatively large. Under this condition, if SOEfw < SOEfwmax, the
rule-based control strategy, the first stage is able to achieve basic power flywheel system recycles brake energy with the maximum power.
allocation. In the second one, high and low-frequency of the required The rest part is recovered by the lithium battery. If SOEfw ≥
power from vehicle are decomposed by wavelet transform. Then, the SOEfwmax, the generated power is fully recovered by the lithium
high-frequency power is distributed to electric flywheel and the low one battery.
is assigned to lithium battery. Owing to the design, the negative impact
of transient and high power on lithium battery can be reduced.

5
B. Sun et al. Energy 295 (2024) 131069

Fig. 4. Topology scheme of electromechanical flywheel.

Fig. 5. Topology of lithium battery-electromechanical flywheel energy system.

3.1.2. The second-stage power split based on wavelet transform frequency approximation signals by discrete wavelet using high-pass
The rapidly changing and transient power affects the efficiency and filter H1(z) and low-pass filter H0(z). Then, the decomposed signal is
life of lithium battery. Method to identify the severe working condition reconstructed by reconstructing filter G1(z) and G0(z). Since the data
and decompose the high-frequency power is essential. In this paper, the amount after signal decomposition is about twofold of the original one,
discrete wavelet transform is used to decompose the lithium battery the data scale needs to be adjusted. The down and up samplers are used
power distributed from the upper stage. The high frequency and tran­ during signal decomposition and reconstruction. In order to ensure the
sient power after decomposition is assign to the electric flywheel. The orthogonality of filter, two-channel orthogonal filter is adopted.
low frequency power is distributed to the lithium battery. The discrete Considering frequency limitation in actual system and the goal of
wavelet transform of signal f(t) is as following: maintaining computational simplicity, a third-order Haar wavelet
∫ +∞ decomposition for power demand signal is selected. As shown in Fig. 7,
1 (t − τ )
W(τ, s) = f (t) √̅̅ ψ dt (3) the third-order Haar wavelet for decomposition and reconstruction of
s s
− ∞
input signal f(t) is present.
In order to avoid the negative impact of transient power on the
Where, W(τ,S) is the wavelet coefficient; f(t) is the original signal; s is the
lithium battery, it is necessary to perform the third-order Haar wavelet
scale factor, s = 2j j ∈ Z; τ is the translation factor, τ = k2j, k ∈ Z; Ψ is
decomposition on the power distributed to the lithium battery Pbat1 from
wavelet function.
the first stage. The high-frequency transient power is supported by the
In order to simplify program composition and improve code execu­
electric flywheel, while the low-frequency power is distributed to the
tion rate, Haar wavelet algorithm is used in this paper. The Haar wavelet
lithium battery. The power distribution result after decomposition by
function is as following:
wavelet transform is as following:
⎧
⎨ 1, x ∈ [0， 0.5) ⎧
ψ (x) = − 1, x ∈ [0.5， 1) (4) ⎨ Pbat1 = Pbat2 + Pfw2
⎩ P = f0 (t) (5)
0, otherwise ⎩ bat2
Pfw2 = f1 (t) + f2 (t) + f3 (t)
The original signal is decomposed into low-frequency and high-

6
B. Sun et al. Energy 295 (2024) 131069

Fig. 6. Two-stage power split energy management strategy.

Where, Pfw2 is the power allocated by wavelet transform to the flywheel

system.
The power distribution between the lithium battery and the electric
flywheel after the two-stage strategy is as following:
{
Pbat = Pbat2
(6)
Pfw = Pfw1 + Pfw2

Where, Pfw is the power distributed to the flywheel system after the two-
stage power distribution; Pfw1 is the power assigned to the flywheel by
the logic threshold strategy.
In order to avoid the SOE of flywheel is lower or higher than its limit
value after the third-order Haar wavelet, a design margin of 0.01 is used
on the basis of SOEfwmin and SOEfwmax. When SOEfw > SOEfwmin+0.01 or
SOEfw < SOEfwmax-0.01, the Haar wavelet is conducted. While in other
cases, the decomposition of Pbat1 ends. The power of the lithium battery
and the electric flywheel are obtained by the logic threshold control
strategy.

Fig. 7. Schematic diagram of the third-order Haar wavelet.

7
B. Sun et al. Energy 295 (2024) 131069

3.2. Lithium battery-electromechanical flywheel energy system 18000r•min− 1. The capacity and energy storage of lithium battery are
220A•h and 73 kW•h. The hardware-in-the-loop (HIL) test platform is
As electromechanical flywheel is able to act as a power source, in designed for compiling the energy management strategy. The compiled
order to better distribute the torque between the main drive motor and code is conducted on the test bench. The electric dynamometer is used to
the electromechanical flywheel to improve the efficiency of the hybrid load the operation resistance, so as to simulate the driving resistance of
system, a drive torque distribution control strategy based on logic the vehicle under different test cycles. Due to the limitation of labora­
threshold control is designed, as shown Fig. 8. The lower limit SOC value tory space, a safe and controllable performance test chamber for lithium
of the lithium battery is set as SOCbat_min, the maximum torque of the battery is unable to be installed. To solve this problem, a lithium battery
speed control motor is Tm2_max, the maximum torque of the main drive simulator is adopted. By presetting parameters of a given lithium battery
motor is Tm1_max. The three parameters are designed as the threshold and on-line control, the charging and discharging performance of it can
values of the logic strategy. be simulated.
When SOC is less than or equal to the lower limit value, the output For the test platform of lithium battery-electromechanical flywheel
torques from the two motors are zero. When SOC is higher than the composite system, the power coupler is used to couple power of main
lower limit value, the control strategy is as followings. drive motor and electromechanical flywheel device. Thus, four-wheel
drive scheme of the hybrid system is achieved. As the lithium battery-
(1) When the torque Treq required by driver is less than Tmot2_max, Treq electric flywheel energy system has no mechanical power coupling
is completely provided by the speed control motor. The main with vehicle, the power coupler is not used during the test of it. For the
drive motor does not participate in the work. test platform of the lithium battery-electromechanical flywheel com­
(2) When Tmot2_max ≤ Treq ≤ Tmot1_max, Treq is fully provided by the posite system, the power coupler is used to couple power of the main
main drive motor, the speed regulating motor does not partici­ drive motor and the electromechanical flywheel. Thus, four-wheel drive
pate in the work. scheme of the hybrid system is achieved. As the lithium battery-electric
(3) When Tmot1_max < Treq ≤ Tmot1_max + Tmot2_max, the main drive flywheel energy system has no mechanical power coupling with vehicle,
motor provides constant torque Tmot1_max, the insufficient part is the power coupler is not used during the test of it.
provided by the speed regulating motor.
(4) When Treq > Tmot1_max + Tmot2_max, Treq needs to be corrected. The
main drive motor provides constant torque Tmot1_max and the 4.2. Influence of energy strategy on lithium battery-electric flywheel
speed control motor provides constant torque Tmot2_max. system

4. Performance test and data analysis As shown in Fig. 10, under CLTC cycle, the influences of energy
strategy on lithium battery-electric flywheel system are given. Results
4.1. Introduction of the test platform indicate that, compared to the rule-based strategy, the proposed energy
management strategy based on two-stage power split shows greater
As shown in Fig. 9, test platform is developed for performance test. advantage on improvement of the lithium battery performance.
The maximum energy of electric flywheel is 0.67 kW•h with the First of all, the new strategy is able to increase the proportion of
maximum speed of 42000r•min− 1. The maximum energy of electro­ working conditions where electric flywheel participates in work. Owing
mechanical flywheel is 0.11 kW•h with the maximum speed of to this, the lithium battery is able to operate more stably. The peak
power and current of it decrease by 82.67% and 90.43%. The average

Fig. 8. Torque distribution control strategy flow chart.

8
B. Sun et al. Energy 295 (2024) 131069

Fig. 9. Composite energy system performance test platform.

drive power and current of the lithium battery decrease by 18.18% and scheme 1, the maximum discharge current of the lithium battery is
23.56%, respectively. Finally, benefiting by the improvement of the new reduced by 30.98% to 303A. However, the maximum charge current is
strategy on lithium battery operations, its average efficiency improves increased by 15.28% to 166A. The average discharge current of the
by 1.42%. lithium battery is reduced by 1.73%–39.77A. Whlie, the average charge
current is increased by 31.07%–37.08A.
Compared with the electric flywheel, the electromechanical flywheel
4.3. Influence of flywheel device on lithium battery current shows less effect on the regulation of the lithium battery current. The
main reasons are as followings: First of all, compared with the electric
As shown in Fig. 11, the discharge current variation of the lithium flywheel, the maximum design energy storage of the electromechanical
battery under CLTC cycle is given. For the original scheme with single flywheel is reduced by 83.58%, which inevitably affects the energy
energy (scheme 1), the maximum discharge and charge currents of the regulation range of the lithium battery. Thus, the larger energy flow of
lithium battery are 439A and 144A, respectively. The average discharge the lithium battery means the larger lithium battery current under
and charge currents are 40.47A and 28.29A. As for the lithium battery- scheme 3. Secondly, due to the unique topology of the electromechan­
electric flywheel composite energy scheme (scheme 2), due to the ical flywheel, constrained by the planetary gear mechanism, when the
regulation of the electric flywheel, the maximum discharge and charge flywheel attached to ring gear releases energy, the speed control motor
currents of the lithium battery are reduced by 90.21% and 100%, to 43A connected to sun gear has to work at the same time. The operating motor
and 0 A. The average discharge and charge currents are reduced by requires energy from the lithium battery, which in turn increases the
37.39% and 100%, respectively, to 25.34A and 0 A. It shows that the discharge current of it.
electric flywheel has a perfect regulation effect on the charge and Compared with scheme 1, both the maximum and the average charge
discharge currents of lithium battery. currents of the lithium battery increase under scheme 3. The main
For the lithium battery-electromechanical flywheel composite en­ reason is that scheme 1 uses the single-motor and two-wheel drive
ergy scheme (scheme 3), different from scheme 2, it shows different power system. This scheme can only realize two-wheel regenerative
regulation rules for the currents of the lithium battery. Compared with

9
B. Sun et al. Energy 295 (2024) 131069

Fig. 10. Influence of energy strategy on lithium battery-electric flywheel system.

Fig. 11. Current variation of lithium battery under different schemes.

10
B. Sun et al. Energy 295 (2024) 131069

brake recovery. scheme 3 adopts the dual-motor and four-wheel drive control motor under CLTC is given. Results show that, in order to ensure
power system. Therefor, four-wheel regenerative brake energy recovery high SOE, the electric flywheel has to run frequently in the high-speed
is achieved. The results show that the total brake energy recovery is range. As there is no variable speed device between the flywheel and
increased by 27.14%. Furthermore, as shown in Fig. 12, as the maximum the control motor, the motor needs to operate in the high-speed and low-
design energy of the electromechanical flywheel is small, the conversion torque range for a long time, where the motor efficiency is very low. The
between vehicle kinetic energy to flywheel kinetic energy is limited. average efficiency of the flywheel control motor is only 75.92%,
Consequently, most of the recovered brake energy is converted to resulting in the low efficiency of the electric flywheel device.
electric energy, which actually increases the maximum and average The losses of the electromechanical flywheel mainly include the
charge currents of the lithium battery. wind resistance loss from flywheel rotor, the bearing loss, the mechan­
ical friction loss and the speed control motor loss. By measuring the
4.4. Effect of flywheel device on lithium battery efficiency input and output torques of the electromechanical flywheel under
different speeds, the operation efficiency of the electromechanical
As shown in Fig. 13, under CLTC test cycle, the initial SOC of lithium flywheel is obtained. Fig. 15 shows the efficiency characteristic of the
battery is set to 0.8, the charge and discharge efficiencis of the lithium electromechanical flywheel under CLTC condition. Results show that,
battery are achieved. For scheme 1, the average efficiency of the lithium compared with the electric flywheel, the frequency that the electrome­
battery is 91.36%. Owing to the current reduction, the average effi­ chanical flywheel participates in CLTC operation increases by 136.25%.
ciency of the lithium battery is increased by 4.49%–95.46% under The main reason is that the maximum energy capacity of theelec­
scheme 2. Unlike scheme 2, the electromechanical flywheel presents tromechanical flywheel is low. In order to ensure the adjustment effect
different influence on the efficiency regulation of the lithium battery. of the device, the flywheel is controlled to participate in CLTC operation
Compared with scheme 1, the discharge efficiency of the lithium battery more frequently.
is increased by 2.12% while the charge efficiency is reduced by 2.79%. Furthermore, the average efficiency of the electromechanical
The average efficiency is 91.03% with an overall reduction of 0.36%. flywheel is higher, that is 86.06%. Owing to the influence of the plan­
The main reason is that, as discussed above, the intervention of the etary gear, there is a speed coupling relationship between the flywheel
electric flywheel reduces the charge and discharge currents of the and the speed regulating motor. The probability of the speed control
lithium battery. It is helpful for improving the lithium battery efficiency. motor running in low efficiency area, high speed and low torque, is
Although, the electromechanical flywheel is able to reduce the discharge significantly reduced. It is helpful for improving the condition efficiency
current of the lithium battery and improve the discharge efficiency, it, at of the electromechanical flywheel device.
the same time, increases the charge current of it and causes larger
negative effect on the charge efficiency. Consequently, the overall effi­ 4.6. Analysis of system net energy consumption
ciency of the lithium battery reduces under scheme 3.
As shown in Fig. 16, the system net energy consumptions of different
4.5. Analysis of flywheel device efficiency schemes under CLTC condition are given. As the lithium battery, electric
flywheel and electromechanical flywheel have the characteristic of
As magnetic suspension bearing and vacuum chamber are used in the bidirectional energy flow during operation, the system net energy con­
electric flywheel, the efficiency of it is mainly affected by the flywheel sumption is defined in this paper. That is, taking the electric flywheel for
control motor. By measuring the input and output torques of the electric example, the system net energy consumption of it means the difference
flywheel at different speeds, the efficiency of the device can be derived. between the actual energy consumption and the recovered brake energy.
As shown in Fig. 14, the efficiency characteristic diagram of the flywheel If the difference is negative, the recovered brake energy is greater than

Fig. 12. Regenerative brake of flywheel under two composite energy systems.

11
B. Sun et al. Energy 295 (2024) 131069

Fig. 13. Efficiency of lithium battery under different schemes.

Fig. 14. Efficiency diagram of electric flywheel.

Fig. 15. Efficiency diagram of electromechanical flywheel.
its energy consumption, which is helpful for the improvement of system
economy. measuring the energy consumption and recovery of the lithium battery
As for scheme 1, the system net energy consumption is obtained by simulator. For scheme and 3, the system net energy consumptions

12
B. Sun et al. Energy 295 (2024) 131069

importantly, its cycle efficiency is higher with an average effi­

ciency of 86.76%.
(2) scheme 3, with four-wheel drive advantage, is able to achieve
more brake energy recovery. The recovered energy increases by
27.14% and 24.39% compared with scheme 1 and scheme 2.
(3) As shown in Fig. 17, under the premise of meeting the same
power performance of vehicle, scheme 3 with four-wheel drive is
able to use a smaller main drive motor. Consequently, the effi­
ciency of the main drive motor improves owing to the increase of
the load rate. Compared with scheme 2, the average efficiency of
the main drive motor improves by 4.72%.

Fig. 18 gives the system net energy consumption of the three schemes
under UUDS and HWFET cycles. They represent the urban and highway
operations, respectively. Different from the test results under CLTC and
Fig. 16. System net energy consumption under different schemes. HWFET cycles, the system net energy consumption of the electric
flywheel is negative under UUDS. Frequent brake is the unique char­
acteristic of the urban cycle, which definitely increases the recovered
consist of the lithium battery net consumption and the flywheel one.
energy of the electric flywheel. Consequently, the energy recovery of
Results show that, the net energy consumption of the lithium battery
flywheel is greater than its energy consumption. However, under
under scheme 2 is the highest. Compared with scheme 1 and 3, it in­
HWFET cycle, the system net energy consumption of the electric
creases by 2.61% and 8.76%. Moreover, the system net energy con­
flywheel shows a significant raise. Compared with CLTC and UUDS cy­
sumption of scheme 2 is 6.55% and 12.93%, higher than scheme 1 and 3.
cles, it increases by 17.08% and 42.45%. Stable operating condition is
The main reasons are as followings.
the special characteristic of HWFET cycle, which means less regenera­
tive brake and energy recovery of the electric flywheel. Furthermore, the
(1) The energy conversion efficiency of the electric flywheel is low.
low efficiency of the flywheel device has further deterioration on the
The use of the electric flywheel is able to improve the lithium
system net energy consumption, since the lithium battery is controlled to
battery charge and discharge efficiency, compared with scheme
charge the flywheel frequently to ensure that it operates above the
1, the average cycle efficiency is increased from 91.36% to
minimum speed limit.
95.46%. However, the average efficiency of the electric flywheel
Moreover, compared with CLTC and UUDS cycles, under HWFET
is very low, only 75.92%. It means that, when the electric
condition, for the three schemes, the net energy consumptions of the
flywheel takes part in operation, the energy consumption
lithium battery increase in different degrees. The main reason is that the
improvement of the lithium battery is not enough to compensate
less brake means the smaller energy recovery for the lithium battery and
for the additional energy consumption of the flywheel.
flywheel, thus the lithium battery operates with high power and current
(2) With the use of the designed energy management strategy, the
frequently. Low efficiency of the lithium battery is inevitable.
lithium battery is controlled to charge the flywheel according to
its SOE. Thus, the flywheel speed can be maintained above the
5. Conclusion
minimum limit speed.
(1) The electric flywheel is able to regulate the charge and discharge
The net energy consumption of the lithium battery under scheme 3 is
currents of the lithium battery by instantaneous and high-power
the lowest. It is 5.65% and 8.05% lower than scheme 1 and scheme 2. In
intervention, thus improving the lithium battery efficiency.
addition, scheme 3 shows the lowest system net energy consumption,
However, due to the frequent operation of the electric flywheel in
that is 6.44% and 12.19% lower than the other two schemes respec­
high-speed and low-torque range, the efficiency of the control
tively. The main reasons are as followings.
motor is very low. The long-time and low-efficiency operation of
the control motor not only affects the efficiency of the electric
(1) Compared with the electric flywheel, the participation proportion
flywheel, but also has a negative impact on the whole hybrid
of the electromechanical flywheel in CLTC cycle is higher. More

Fig. 17. Efficiency of the main drive motor of the composite energy system schemes.

13
B. Sun et al. Energy 295 (2024) 131069

Fig. 18. System net energy consumption under different cycles.

energy system. On the perspective of vehicle economy optimi­ Data availability

zation, there are two key issues need to be solved urgently for the
application of the electric flywheel. Deign to improve the effi­ Relevant data has been provided in the article.
ciency of the control motor is essential, or new method is required
to avoid the long-time operation of the control motor in low ef­ Acknowledgements
ficiency range.
(2) Vehicle operating condition has great influence on the system This work was supported in part by the Shandong Provincial Talents
efficiency. Under urban operation, as the auxiliary electric Team Cultivation Plan of University Preponderant Discipline
flywheel and electromechanical flywheel are able to participate (2021KJ083), the National Natural Science Foundation Project of China
in brake energy recovery frequently, the recovered energies are (51805301, 52375105), the Shandong Province Excellent Youth Talent
greater than their energy consumptions, resulting in the positive Fund Project (ZR2022YQ51), the Key Technology Research and Devel­
effect on the net energy consumptions of the lithium battery and opment Program of Shandong Province (2020CXGC010405,
the hybrid system. While under other vehicle conditions, espe­ 2020CXGC010406 and 2021CXGC010703).
cially the highway operation, the recovered energy of the
flywheel is unable to cover the energy loss, leading to the increase References
of the system net energy consumption.
(3) The electromechanical flywheel shows special effect on system [1] Xiong Y, Wang Z, Cheng Q. Effects of new energy vehicle adoption on provincial
energy efficiency in China: from the perspective of regional imbalances. Energy
efficiency. With the use of the flywheel energy control, both the 2023;281:128324.
current and efficiency of the lithium battery are improved. By [2] Jayaprabakar J, Aravind Kumar J, Parthipan J, et al. Review on hybrid electro
controlling the torque of the speed control motor, the load rate of chemical energy storage techniques for electrical vehicles: technical insights on
design, performance, energy management, operating issues & challenges. J Energy
the main drive motor is increased with higher efficiency. The Storage 2023;72(Part D):108689.
electromechanical flywheel shows more advantages in improving [3] Yuan X, Zang C, Hong G, et al. Method for evaluating the real-world driving energy
the economy of electric vehicle. consumptions of electric vehicles. Energy 2017;141:1955–68.
[4] Li Y, Wang S, Duan X, Liu S, Hu S. Multi-objective energy management for atkinson
cycle engine and series hybrid electric vehicle based on evolutionary nsga-ii
In conclusion, compared with the single energy scheme with lithium algorithm using digital twins. Energy Convers Manag 2021;230(38):113788.
battery, under CLTC operation, by using the electric flywheel, the [5] Kazmouz S, Haworth D, Lillo P, Sick V. Extension of a thickened flame model to
highly stratified combustion - application to a spark-ignition engine. Combust
average operation efficiency of the lithium battery improves by 4.49%,
Flame 2022;236:111798.
but the net loss of the composite energy system increases by 2.61%. With [6] Ortiz-Soto E, Yang X, Ess J, Owlia S, Joshi A, Younkins M. Controls and hardware
the use of the electromechanical flywheel, the average operation effi­ development of multi-level miller cycle dynamic skip fire (mdsf) technology. SAE
ciency of the lithium battery increases only by 0.39%, while the net loss International Journal of Advances and Current Practices in Mobility 2021;(4):3.
[7] Itani A K, Bernardinis D, Khatir Z, et al. Comparative analysis of two hybrid energy
of the composite energy system decreases by 6.44%. Future work will storage systems used in a two front wheel driven electric vehicle during extreme
focus on the testing of the different energy systems in real vehicles under start-up and regenerative braking operations. Energy Convers Manag 2017;144:
the actual traffic scenario. 69–87.
[8] Enang W, Bannister C. Modelling and control of hybrid electric vehicles (A
comprehensive review). Renew Sustain Energy Rev 2017;74:1210–39.
CRediT authorship contribution statement [9] Du B, Qin D, Yang Y. Coordinated control strategies for a full hybrid electric vehicle
with single motor during mode changes. Int J Electr Hybrid Veh (IJEHV) 2016;8
(1):19–40.
Binbin Sun: Writing – review & editing, Writing – original draft, [10] Huang K, Xiang C, Langari R. Model reference adaptive control of a series-parallel
Validation, Methodology, Investigation, Formal analysis. Bo Li: Vali­ hybrid electric vehicle during mode shift. Proc IME J Syst Control Eng 2017;231
dation, Investigation, Formal analysis. Jilei Xing: Validation, Method­ (7):541–53.
[11] Yang Y, Xu Y, Zhang H, et al. Research on the energy management strategy of
ology. Xiao Yu: Writing – review & editing, Writing – original draft, extended range electric vehicles based on a hybrid energy storage system. Energy
Investigation. Mengxue Xie: Writing – original draft, Methodology. Rep 2022;8:6602–23.
Zihao Hu: Writing – original draft, Investigation. [12] Itani K, Bernardinis A, Khatir Z, Jammal A. Comparative analysis of two hybrid
energy storage systems used in a two front wheel driven electric vehicle during
extreme start-up and regenerative braking operations. Energy Convers Manag
2017;144:69–87.
Declaration of competing interest [13] Khan F, Mohammad G, Sayem A, Nirmal K. Design and optimization of lithium-ion
battery as an efficient energy storage device for electric vehicles: a comprehensive
The authors declare that they have no known competing financial review. J Energy Storage 2023;71:108033.
[14] Hasan M, Mahmud M, Habib Motakabber S, Islam S. Review of electric vehicle
interests or personal relationships that could have appeared to influence energy storage and management system: Standards, issues, and challenges.
the work reported in this paper. J Energy Storage 2021;41:102940.

14
B. Sun et al. Energy 295 (2024) 131069

[15] Papageorgiou P, Oureilidis K, Christoforidis G. A systematic review of hybrid [29] Bottiglione F, Carbone G, Novellis LD, et al. Mechanical hybrid KERS based on
superconducting magnetic/battery energy storage systems: applications, control toroidal traction drives: an example of smart tribological design to improve
strategies, benefits, limitations and future prospects. Renew Sustain Energy Rev terrestrial vehicle performance. Advances in Tribology 2013;2013:918387.
2023;183:113436. [30] Zhang Z, Yang B, Zhang Y, Li L, Zhao B, Zhang T. Powertrain modeling and
[16] Wilberforce T, Answer A, Swamy J, Opoku R. An investigation into hybrid energy performance simulation of a novel flywheel hybrid electric vehicle. Energy Rep
storage system control and power distribution for hybrid electric vehicles. Energy 2023;9:4401–12.
2023;279:127804. [31] Erhan O, Özdemir E. Prototype production and comparative analysis of high-speed
[17] Li F, Gao Y, Wu Y, et al. Incentive learning-based energy management for hybrid flywheel energy storage systems during regenerative braking in hybrid and electric
energy storage system in electric vehicles. Energy Convers Manag 2023;293: vehicles. J Energy Storage 2021;43:103237.
117480. [32] Lyu X, Di L, Yoon SY, Lin Z, Hu Y. A platform for analysis and control design:
[18] Sun B, Gao S, Ma C. System power loss optimization of electric vehicle driven by Emulation of energy storage flywheels on a rotor-AMB test rig. Mechatronics 2016;
front and rear Induction motors. Int J Automot Technol 2018;19(1):121–34. 33:146–60.
[19] Wei H, Zhang N, Liang J, et al. Deep reinforcement learning based direct torque [33] Read MG, Smith RA, Pullen KR. Optimisation of flywheel energy storage systems
control strategy for distributed drive electric vehicles considering active safety and with geared transmission for hybrid vehicles. Mech Mach Theor 2015;87:191–209.
energy saving performance. Energy 2022;238(Part B):121725. [34] Dhand A, Pullen K. Optimal energy management for a flywheel assisted battery
[20] Xu T, Zhao Y, Deng H, Guo S, Li D, Lin F. Integrated optimal control of distributed electric vehicle. Proc Inst Mech Eng - Part D J Automob Eng 2015;229(12):
in-wheel motor drive electric vehicle in consideration of the stability and economy. 1672–82.
Energy 2023;282:128990. [35] Duan B, Wang Q, Zeng X, et al. Calibration methodology for energy management
[21] Cui H, Ruan J, Wu C, Zhang K, Li T. Advanced deep deterministic policy gradient system of a plug-in hybrid electric vehicle. Energy Convers Manag 2017;136:
based energy management strategy design for dual-motor four-wheel-drive electric 240–8.
vehicle. Mech Mach Theor 2023;179:105119. [36] Tran Dai-Duong, Vafaeipour Majid, El Baghdadi Mohamed, et al. Thorough state-
[22] Sun B, Gao S, Wu Z, et al. Parameters design and economy study of an electric of-the-art analysis of electric and hybrid vehicle powertrains: Topologies and
vehicle with powertrain systems in front and rear axle. International Journal of integrated energy management strategies. Renew Sustain Energy Rev 2020;119:
Engineering Transactions A: Basics 2016;29(4):454–63. 109596.
[23] Zhang J, Yang Y, Hu M, Yang Z, Fu C. Longitudinal–vertical comprehensive control [37] Qi Li, Chen Weirong, Yankun Li. Energy management strategy for fuel cell/battery/
for four-wheel drive pure electric vehicle considering energy recovery and ride ultra capacitor hybrid vehicle based on fuzzy logic. Int J Electr Power Energy Syst
comfort. Energy 2021;236:121417. 2012;43(1):514–25.
[24] Hansen J, O’Kane D. An Assessment of flywheel high power energy storage [38] Nassar Mohamed Y, Shaltout Mohamed L, et al. Multi-objective optimum energy
Technology for hybrid vehicles [R]. Oak Ridge National Laboratory; 2012. management strategies for parallel hybrid electric vehicles: a comparative study.
[25] Dhand A, Pullen K. Review of battery electric vehicle Propulsion systems Energy Convers Manag 2023;277:116683.
Incorporating flywheel energy Storge. Int J Automot Technol 2015;16(3):487–500. [39] Lorenzo S, Simona O, Giorgio R. A comparative analysis of energy management
[26] Hansen J, O’Kane D. An Assessment of flywheel high power energy storage strategies for hybrid electric vehicles. J Dyn Syst Meas Control 2011;133(3):1–9.
Technology for hybrid vehicles [R]. Oak Ridge National Laboratory; 2012. [40] Enang W, Bannister C. Modelling and control of hybrid electric vehicles (A
[27] Arani A, Karami H, Gharehpetian G, et al. Review of flywheel energy storage comprehensive review). Renew Sustain Energy Rev 2017;74:1210–39.
systems structures and applications in power systems and microgrids. Renew [41] Liu T, Tan W, Tang X, et al. Driving conditions-driven energy management
Sustain Energy Rev 2017;69:9–18. strategies for hybrid electric vehicles: a review. Renew Sustain Energy Rev 2021;
[28] Dhand A, Pullen K. Review of flywheel based internal combustion engine hybrid 151:111521.
vehicles. Int J Automot Technol 2013;14(5):797–804. [42] Ganesh AH, Xu B. A review of reinforcement learning based energy management
systems for electrified powertrains: Progress, challenge, and potential solution.
Renew Sustain Energy Rev 2022;154:111833.

15