Energy

Multi-speed transmission optimization of electric vehicles based on shifting pattern

considering equivalent inertia variation
--Manuscript Draft--

Manuscript Number: EGY-D-22-09652

Article Type: Full length article

Keywords: Electric vehicle; Two-speed transmission; Shifting pattern; Gear ratio; Equivalent
inertia variation; multi-objective optimization

Abstract: Multi-speed transmissions for electric vehicles (EVs) can achieve superior economic
and dynamic performance than single-speed transmissions. Since gear shifting causes
an equivalent inertia variation in multi-speed transmissions, the optimal shifting pattern
should be determined by considering the inertia variation effect as well as the motor
and transmission efficiencies to maximize the economic and dynamic performance of
EVs. Since the analysis results of the optimal shifting patterns when considering and
not considering the inertia variation differ significantly in terms of EV performances, it is
essential to apply the optimal shifting pattern considering the inertia variation effect for
the superior transmission design of EVs. A multi-objective optimization problem is
formulated that includes the design variables as gear ratios and shifting patterns and
the objective functions as energy efficiency and acceleration ability. To verify the
importance of the equivalent inertia effect on the economic and dynamic performances,
the gear ratios and shifting patterns are optimized by considering the inertia effect and
none. The different optimum solutions and objective function values demonstrate the
necessity of considering the inertia variation effect due to gear shifting; the economic
and dynamic performances are improved from 2.7% to 7.8% and 2.8% to 3.0%,
respectively, in various driving cycles.

Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation
Highlights

Title: Multi-speed transmission optimization of electric vehicles based on shifting patter

n considering equivalent inertia variation
Highlights:

 Analysis of equivalent inertia variation from gear-shifting in transmissions

 Equivalent inertia effect on economic and dynamic performances of electric vehicle

 New formulation to optimize gear-shifting patterns including inertia effect

 Optimization results of gear ratios and shifting patterns under various cycles

 Effectiveness of optimal shifting patterns considering equivalent inertia variation

Manuscript Click here to view linked References

1
2
3
4
5
6
7
8
9 Multi-speed transmission optimization of electric
10
11
vehicles based on shifting pattern considering
12 equivalent inertia variation
13
14
15 Kihan Kwona , Kijong Parkb,∗
16 a Department of Automotive Engineering, Honam University, Gwangju, Korea
17 b Research and Development Division, Hyundai Motor Company, Hwaseong, Korea
18
19
20
21
22 Abstract
23
24 Multi-speed transmissions for electric vehicles (EVs) can achieve superior eco-
25
26 nomic and dynamic performance than single-speed transmissions. Since gear
27
28 shifting causes an equivalent inertia variation in multi-speed transmissions, the
29 optimal shifting pattern should be determined by considering the inertia vari-
30
31 ation effect as well as the motor and transmission efficiencies to maximize the
32 economic and dynamic performance of EVs. Since the analysis results of the
33
34 optimal shifting patterns when considering and not considering the inertia vari-
35
ation differ significantly in terms of EV performances, it is essential to apply the
36
37 optimal shifting pattern considering the inertia variation effect for the superior
38
39 transmission design of EVs. A multi-objective optimization problem is formu-
40 lated that includes the design variables as gear ratios and shifting patterns and
41
42 the objective functions as energy efficiency and acceleration ability. To verify
43 the importance of the equivalent inertia effect on the economic and dynamic
44
45 performances, the gear ratios and shifting patterns are optimized by consider-
46
47 ing the inertia effect and none. The different optimum solutions and objective
48 function values demonstrate the necessity of considering the inertia variation ef-
49
50 fect due to gear shifting; the economic and dynamic performances are improved
51 from 2.7% to 7.8% and 2.8% to 3.0%, respectively, in various driving cycles.
52
53
54 ∗ Correspondingauthor
55 Email address: [email protected] (Kijong Park)
56
57
58
59 Preprint submitted to Energy October 11, 2022
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9 Keywords: Electric vehicle, Two-speed transmission, Shifting pattern, Gear
10
ratio, Equivalent inertia variation, Multi-objective optimization
11
12
13
14 1. Introduction
15
16 The electric vehicle (EV) market has been rapidly growing in the automo-
17
18 tive industry as an alternative to vehicles that consume fossil fuels [1]. In an
19 EV powertrain, a transmission that converts the torque and speed of an elec-
20
21 tric motor into wheels generally employs a single-speed gear because it has a
22
simple structure with no gear shifting, enabling excellent high transmission ef-
23
24 ficiency and drivability [2]. However, compared to multi-speed transmissions,
25
26 single-speed transmissions have several drawbacks in terms of economic and dy-
27 namic performance [3–5]. For economic performance, even if a motor drives
28
29 in the low powertrain efficiency region under a specific vehicle driving condi-
30
tion, a single-speed transmission cannot control the motor operation because
31
32 the motor torque and speed are determined by a fixed transmission gear ratio
33
34 and the requested vehicle speed and force [6, 7]. However, a multi-speed trans-
35 mission allows the motor to operate in a high-efficiency region with appropriate
36
37 gear shifting depending on the vehicle’s driving conditions. For dynamic per-
38 formance, the vehicle traction force is represented by the product of the motor
39
40 torque and gear ratio, and the vehicle speed is expressed as the motor speed
41
divided by the gear ratio. Therefore, it is appropriate to use high and low gear
42
43 ratios to satisfy high traction force and vehicle speed, respectively.
44
45 Multi-speed transmissions are more advantageous than single-speed trans-
46 missions for EV performance because selecting the desired speed gear with an
47
48 appropriate gear ratio among various speed gears is possible. The effect of gear
49
shifting on powertrain efficiency and traction force is illustrated in Figure 1.
50
51 Since multi-speed transmissions are more complex than single-speed transmis-
52
53 sions, design optimization of the transmission is essential to find an optimal
54 solution that maximizes the economic and dynamic performances of EVs [8].
55
56 Gear ratio optimization is essential because the critical design parameter of
57
58
59 2
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9 a multi-speed transmission is the gear ratio of each speed [9]. However, the
10
gear-shifting pattern controls the driving gear according to the vehicle driving
11
12 conditions; therefore, optimizing the shifting pattern is also significant. Sev-
13
14 eral previous studies have proposed shifting patterns for multi-speed EVs. A
15 detailed review of these studies is presented in Table 1.
16
17 This literature survey revealed the following: First, the shifting pattern
18
generally focuses on economic performance, and the optimal pattern is deter-
19
20 mined primarily to maximize only motor efficiency. Second, some studies have
21
22 conducted gear ratio optimization simultaneously because the optimal shifting
23 pattern depends on the combination of gear ratios [9–12]. Thus, to effectively
24
25 optimize the shifting pattern, economic and dynamic performances should be
26 considered in the shifting pattern optimization since both are equally impor-
27
28 tant. Some studies have proposed a shifting pattern that focuses on both
29
performances[9–11, 13, 14]. However, these studies have practical limitations
30
31 because the shifting patterns differ for each performance rather than the inte-
32
33 grated pattern. Moreover, increased speed gears should reduce the transmission
34 efficiency because of additional shifting components [15]. Therefore, it is neces-
35
36 sary to consider the transmission efficiency that varies with the gear shift to fur-
37
ther accurately optimize the shifting pattern [12] as well as the motor efficiency.
38
39 In addition, optimizing both gear ratios and shifting patterns can potentially
40
41 improve the economic and dynamic performances of EVs. In summary, the
42 gear-shifting pattern for EVs should be elaborately designed to improve their
43
44 economic and dynamic performances, considering variations in the motor and
45 transmission efficiencies with gear ratio optimization.
46
47
48
49
50
51
52
53
54
55
56
57
58
59 3
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30 (a) Powertrain efficiency
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
(b) Traction force
54
55
56 Figure 1: Example of gear shifting effect.
57
58
59 4
60
61
62
63
64
65
65
64
63
62
61
60
59
58
57
56
55
54
53
52
51
50
49
48
47
46
45
44
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1

Table 1: Summary of previous research on shifting patterns for multi-speed EV.

Target performance Variable efficiency component
Reference Year Gear speed Gear ratio optimization
Economic Dynamic Motor Transmission
Walker et al.[16] 2013 2 ⃝ ⃝
Zhu et al.[13] 2015 2 ⃝ ⃝ ⃝
Liu et al.[17] 2016 4 ⃝ ⃝
Tan et al.[18] 2018 2 ⃝ ⃝

5
Ruan et al.[19] 2018 4 ⃝ ⃝
Lin et al.[20] 2019 4 ⃝ ⃝
Li et al.[10] 2020 2 ⃝ ⃝ ⃝ ⃝
Sun et al.[11] 2021 2 ⃝ ⃝ ⃝ ⃝
Kwon et al.[12] 2021 2 ⃝ ⃝ ⃝ ⃝
Lin et al.[14] 2021 2 ⃝ ⃝ ⃝ ⃝
Wu et al.[9] 2022 3 ⃝ ⃝ ⃝ ⃝
Zhao et al.[21] 2022 2 ⃝ ⃝
 1
2
3
4
5
6
7
8
9 When a vehicle is driven, translational and rotational motions exist in various
10
vehicle components. Based on the law of conservation of kinetic energy, these
11
12 motions can be represented by the motion of equivalent inertia, which is virtual
13
14 inertia, and its kinetic energy is equal to the sum of the kinetic energies of each
15 vehicle component [22], as shown in Figure 2. The equivalent inertia of the input
16
17 side (motor and part of the transmission) differs according to the driving gear
18
ratio. Assuming that the vehicle travels at a constant speed, the operating speed
19
20 of the motor using a high or low gear ratio of the transmission is high or low,
21
22 respectively. As the kinetic energy is proportional to the square of the rotational
23 speed, using a high gear ratio increases the equivalent inertia of the vehicle.
24
25 This effect is negligible as vehicle acceleration approaches zero. However, it
26 significantly affects the requested motor power as vehicle acceleration increases.
27
28 For economic performance, up- or down-shifting can improve the motor and
29
transmission efficiencies by changing the motor torque and speed. However, the
30
31 gear shift entails an additional inertia effect on the input side under acceleration
32
33 conditions. Although the motor and transmission efficiencies after the gear shift
34 are higher or lower than the previous gear, the total energy consumption rate
35
36 may be reversed by considering the variation in inertial energy due to vehicle
37
acceleration. For dynamic performance, using a high gear ratio can multiply the
38
39 input torque more than that when using a low gear ratio. Although a significant
40
41 increase in the input torque ensures a high acceleration of the vehicle, a high
42 acceleration also involves a large inertia resistance. Therefore, using a high
43
44 gear ratio increases the equivalent inertia of the input side, which may result
45
46
47
48
49
50
51
52
53
54
55
56 Figure 2: Concept of equivalent vehicle inertia.
57
58
59 6
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9 in negative effects on vehicle acceleration. In summary, the equivalent inertia
10
variation in the speed gear selection process should be considered to maximize
11
12 the economic and dynamic performances of EVs.
13
14 Previous studies have determined the gear-shifting pattern based on the in-
15 put torque request and vehicle speed only. However, to derive a further optimal
16
17 shifting pattern, it should be expressed by including the effect of the equivalent
18
inertia variation. Therefore, the shifting pattern should be optimized according
19
20 to the degree of vehicle acceleration to reflect the inertia variation effect for each
21
22 speed gear. In addition, the optimal shifting pattern and gear ratios can ensure
23 excellent economic and dynamic performances. However, their enhancement is
24
25 challenging because of the trade-off relationship meaning that optimizing the
26 shifting pattern and gear ratios for economic performance can diminish the dy-
27
28 namic performance [9, 12]. Therefore, a multi-objective optimization method
29
can be an alternative to this problem by providing a Pareto front consisting of
30
31 diverse optimal design solutions [23].
32
33 To overcome the aforementioned research limitations on the transmission
34 optimization of EVs, this study proposes an effective optimization method of
35
36 multi-speed transmission for EVs considering the effect of equivalent inertia
37
variation on the gear-shifting pattern. A reference powertrain configuration of
38
39 a two-speed transmission EV was introduced to analyze the equivalent inertia
40
41 for each speed. Based on this, the equivalent inertia was mathematically de-
42 rived and an EV analysis model was developed to evaluate the economic and
43
44 dynamic performances of EVs. To verify the importance of the equivalent iner-
45 tia variation, the optimal shifting patterns with inertia effect and nothing were
46
47 derived, and the performance results were compared. To address the trade-off
48
relationship between economic and dynamic performance, a multi-objective op-
49
50 timization problem was formulated that included the design variables as gear
51
52 ratios and shifting patterns and the objective functions as energy efficiency and
53 acceleration ability. The optimization results presented different Pareto fronts
54
55 as the optimal solutions based on the shifting pattern from an equivalent iner-
56
tia effect. These optimization results demonstrate the necessity of an equivalent
57
58
59 7
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9 inertia effect in the design of gear-shifting pattern. The contributions of this
10
study can be summarized as follows:
11
12
13 1. By referring to the limitations of previous studies on the multi-speed trans-
14 mission optimization of EVs, this study conducted an optimization of gear
15
16 ratios and shifting patterns considering both economic and dynamic per-
17
formances and variable efficiencies of the motor and transmission.
18
19 2. An optimization process of the shifting pattern, including inertia variation
20
21 for each speed, was proposed to confirm the importance of the equivalent
22 inertia effect on energy efficiency and vehicle acceleration. The comparison
23
24 results of economic and dynamic performances in consideration of inertia
25
variation or none quantitatively exhibited the effect of inertia variation on
26
27 EV performances.
28
29 3. The different optimization results of the gear ratios and shifting patterns
30 according to the equivalent inertia variation showed better objective values
31
32 of the optimal design based on the inertia variation effect. Therefore, it
33
verifies the performance superiority of the transmission design considering
34
35 the equivalent inertia variation due to the gear ratios and shifting.
36
37
38 2. Analysis of Multi-speed EV
39
40
41 An EV model employing multi-speed transmission was developed to analyze
42 the economic and dynamic performances. Previous studies on multi-speed EVs
43
44 have proposed various powertrain configurations. In the gear-shifting process,
45 gear-shifting events generally occur more when vehicle acceleration is considered
46
47 than when not considered. Since frequent gear-shifting causes poor drivability,
48
the multi-speed transmission structure should address this shortcoming appro-
49
50 priately. Therefore, this study adopted a dual-clutch transmission (DCT) that
51
52 exhibits high transmission efficiency and stable drivability [24].
53 The equivalent inertia for each driving speed gear can be derived from the
54
55 powertrain configuration equipped with the DCT. To analyze the equivalent
56
inertia of EV, the reference inertia component is determined. Several studies
57
58
59 8
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26 (a) First-speed gear
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46 (b) Second-speed gear
47
48 Figure 3: Power flows of DCT.
49
50
51 have calculated the equivalent inertia by referring to the wheel, which converts
52 the rotational motion to the translational motion of a vehicle. Figure 3 shows
53
54 the power flow of each driving-speed gear. Although the transmission consists
55
of various rotational components, such as clutches, gears, bearings, and shafts,
56
57
58
59 9
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9
Table 2: Inertia values of each component.
10
11 Component Symbol Value (kg ·m2 )
12 Motor Jm 0.0650
13
14 First clutch Jc1 0.0098
15
16 Second clutch Jc2 0.0172
17 Gear Jg1 0.0019
18 First speed
19 Pinion Jp1 0.0001
20 Gear Jg2 0.0009
21 Second speed
22 Pinion Jp2 0.0002
23
Gear Jgf 0.0413
24 Final
25 Pinion Jpf 0.0002
26
27
28 this study considers clutches and gears to calculate the equivalent inertia for
29
30 the transmission because the inertia values of other components are negligible
31 compared to those of clutches and gears. The inertia values of each component
32
33 are summarized in Table 2 and determined from the component specifications
34
35 introduced in [25]. Considering motor, transmission, and vehicle inertias, the
36 equivalent inertia at the wheel driven by the first-speed gear (Jeq.1 ) can be
37
38 expressed as follows:
39
40
2
41 2

r1 rf
42 Jeq.1 = (Jm + Jc1 + Jp1 )·(r1 rf ) +(Jg1 + Jg2 + Jpf )·rf2 +Jp2 · +Jgf +M Rt2
r2
43 (1)
44
45 where M is the mass of the vehicle, and Rt is the effective tire radius. Likewise,
46
the equivalent inertia at wheel driven by the second-speed gear (Jeq.2 ) can be
47
48 expressed as follows:
49
2
50

2 r2 rf
51 Jeq.2 = (Jm + Jc2 + Jp2 )·(r2 rf ) +(Jg1 + Jg2 + Jpf )·rf2 +Jp1 · +Jgf +M Rt2
r1
52 (2)
53
54 Using the equivalent inertia (Jeq ), vehicle acceleration (a) and speed (v) are
55
56
57
58
59 10
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9 determined as follows:
10
Twhl Rt
11 a=
12 Jeq
Z (3)
13
v = a dt
14
15
16 where Twhl is the wheel torque, represented by the driving (Tdrv ) and resistance
17
(Tres ) torques (Twhl = Tdrv − Tres ). Tdrv can be expressed as follows:
18
19
20 Tdrv = Tm ri rf · ηt (4)
21
22 where Tm is the motor output torque, ri is the speed gear ratio (r1 or r2 ), and
23
24 ηt is the transmission efficiency. Here, ηt is determined by considering the losses
25
26 of the clutch, gear, bearing, and concentric shaft in a DCT [12]. Tres can be
27 expressed as follows:
28  
29 1 2
30 Tres = Rt ρcd Af v + M g (µr cos θ + sin θ) + Tbrk (5)
2
31
32 where ρ is the air density, cd is the drag coefficient, Af is the frontal area, g is
33
34 the acceleration owing to gravity, µr is the rolling resistance coefficient, θ is the
35 road slope, and Tbrk is the braking torque. Tbrk can be classified into mechanical
36
37 (Tmec ) and regenerative torques, as follows:
38
39 Tbrk = Tmec − Tm ri rf · ηt (6)
40
41
42 Here, Tmec is generated by the mechanical brake, and Tm is the negative value
43 determined by the regenerative braking torque distribution, as mentioned in
44
45 [26].
46
Energy efficiency and acceleration ability were adopted as quantitative mea-
47
48 sures to evaluate the economic and dynamic performances of the EVs. For
49
50 energy efficiency, many vehicle manufacturers have presented official efficiency
51 values based on various standard driving cycles such as, the urban dynamometer
52
53 driving schedule (UDDS), highway fuel economy test (HWFET), new European
54 driving cycle (NEDC), and world harmonized light-duty vehicles test procedure
55
56 (WLTP). The energy efficiency of an EV (EFF) means the distance that can be
57
58
59 11
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9 driven using 1 kWh of energy and it is calculated as follows:
10
11 dcyc
EFF = (7)
12 Cbat · dSOC
13
14 where dcyc is the driving cycle distance, Cbat is the battery capacity, and dSOC
15 is the state of charge (SOC) consumption. The dSOC is calculated from the
16
17 analysis of the battery equivalent circuit. The battery voltage (Vbat ) was derived
18
using the circuit voltage equation, as follows:
19
20
21 Vbat = VOCV − Ri Ibat (8)
22
23 where VOCV is the open-circuit voltage, Ri is the internal battery resistance,
24
25 and Ibat is the battery current. Here, Ibat is determined by the battery charging
26
and discharging conditions using the mechanical-electrical power relationship,
27
28 as follows:
29
Tm ωm

30 
 if Tm ≥ 0 (discharging)
31


 ηm ηi Vbat
32 Ibat = (9)
33 
 ηm ηi Tm ωm


34
 if Tm < 0 (charging)
Vbat
35
36 where ωm is the motor speed, and ηm and ηi are the motor and inverter effi-
37
38 ciencies, respectively. dSOC is calculated by integrating Ibat as follows:
39 R
VOCV · Ibat dt
40 dSOC = (10)
41 Cbat
42
For dynamic performance, the 0−100 km/h acceleration time, maximum
43
44 speed, and ascendable gradient have been generally employed as evaluation
45
46 measures in several studies. These measures can be determined under different
47 vehicle speed conditions. Under a wide-open throttle (WOT) condition, the
48
49 0−100 km/h acceleration time (ta ) satisfies the following equation:
50 Z ta
51 v= a dt = 100 km/h (11)
52 0
53
54 Next, the maximum speed (vmax ) can be theoretically determined as follows:
55
ωmax
56 vmax = Rt (12)
57 r2 rf
58
59 12
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9 where ωmax denotes the maximum motor speed. However, the driving torque
10
(Tdrv ) must be greater than the resistance torque (Tres ) at this speed. Oth-
11
12 erwise, vmax is determined at a speed that satisfies the condition Tdrv > Tres .
13
14 Lastly, because the ascendable gradient (θgrad ) considers the drivable condition
15 (v > 0), assuming that the vehicle speed is zero, it can be determined as a
16
17 gradient value equalizing Tdrv and Tres . Therefore, θgrad satisfies the following
18
equation:
19
20 Tmax (0) · r1 rf = Rt M g (µr cos θgrad + sin θgrad ) (13)
21
22 where Tmax is the maximum motor torque, which varies with the motor speed.
23
24 Here, if θgrad is large, µr cos θgrad is negligible. Therefore, θgrad can be expressed
25 as:
26 
Tmax (0) · r1 rf

−1
27 θgrad = sin (14)
Rt M g
28
29 The significant parameter values of a reference EV are summarized in Table 3.
30
31
32 Table 3: Specifications of reference EV.
33
34 Item Specification
35 Vehicle mass 1,605 kg
36
37 Drag coefficient 0.29
38
Frontal area 2.27 m2
39
40 Rolling resistance coefficient 0.01
41
42 Tire radius 0.323 m
43 Final gear ratio 4.058
44
45 Maximum torque 280 Nm
46 Motor Maximum power 80 kW
47
48 Maximum speed 10,000 RPM
49
Open circuit voltage 350 V
50
51 Battery Internal resistance 0.1 Ω
52
53 Capacity 60 kWh
54
55
56
57
58
59 13
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9 3. Effect of equivalent inertia variation on optimal shifting pattern
10
11 The economic and dynamic performances vary considerably depending on
12
13 the gear ratio used under certain driving conditions. Therefore, the design of
14
the gear-shifting pattern is as important as that of the gear ratios. The gear-
15
16 shifting pattern determines the speed gear according to the accelerator pedal
17
18 sensor (APS) and vehicle speed values, which represent the driver’s request and
19 vehicle status, respectively. In the two-speed transmission, the shifting pattern
20
21 represents an up- or down-shift, meaning the first to the second gear shift, or vice
22 versa. Since the up-shift occurs mainly under vehicle acceleration, the shifting
23
24 pattern is generally determined by considering the optimal up-shift conditions,
25
26 and the down-shift is determined by offsetting the vehicle speed on the up-shift
27 to avoid frequent gear shifts [13].
28
29 With respect to economic performance, gear shift should be performed to
30 minimize the required motor power under the same wheel power, represented
31
32 as the product of wheel torque and speed, which maximizes the motor and
33
transmission efficiencies. Although the motor efficiency (ηm ) is determined only
34
35 by the input torque and speed, the transmission efficiency (ηt ) additionally
36
37 varies with the driving gear. Therefore, the powertrain efficiency expressed as
38 the product of the motor and transmission efficiencies (ηm ηt ) is different for
39
40 each driving gear, as shown in Figure 4. These efficiency maps can be converted
41 equally into efficiency values according to the APS and vehicle speed, and an
42
43 optimal shifting line can be determined by selecting a gear in a more efficient
44
45 area when driven with each gear, as shown in Figure 5. Based on this shifting
46 line, the optimal shifting pattern can be determined to maximize the economic
47
48 performance of the EV.
49 With respect to dynamic performance, gear shift should be performed to
50
51 maximize the output torque according to the vehicle speed. Therefore, the
52
driving torques (Tdrv ) of each speed gear were calculated for the APS values
53
54 from 0% to 100%, and the gear-shifting speeds were determined from the cross-
55
56 torque points, as shown in Figure 6. These methods utilize only the static
57
58
59 14
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
(a) First gear (r1 = 3.321) (b) Second gear (r2 = 1.681)
22
23
Figure 4: Powertrain efficiency (ηm ηt ) maps for each driving gear.
24
25
26 characteristics of efficiency and output torque to determine the gear-shifting
27
28 pattern. Although many previous studies have used these methods, they are
29
not the best under dynamic conditions such as vehicle acceleration. To obtain a
30
31 more optimized shifting pattern for EVs, the static and dynamic characteristics
32
33 should be considered together in the optimal shifting pattern process as follows.
34 When the vehicle is accelerating, the acceleration resistance torque on the
35
36 wheel can be calculated as follows:
37
38 Tacc = Jeq · αwhl (15)
39
40 where αwhl denotes the angular acceleration of the wheel. From Eqs. (1) and
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56 Figure 5: Optimal shifting pattern for economic performance.
57
58
59 15
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22 Figure 6: Optimal shifting pattern for dynamic performance.
23
24
25 (2), the equivalent inertia variation (Jvar ) between the first and second gear
26 drives can be derived by subtracting Jeq.2 from Jeq.1 as follows:
27
28
29
Jp1 r12 Jp2 r22 2
 
Jvar = Jm r12 − r22 + Jc1 r12 − Jc2 r22 + Jp1 r12 − Jp2 r22 +


30 − rf
31 r22 r12
32 (16)
33 When the gear ratios (r1 , r2 ) are fixed, Jvar is determined as a constant value. In
34
35 addition, a large difference in gear ratios also increases the inertia variation, indi-
36
cating that the equivalent inertia depending on the gear shift significantly affects
37
38 vehicle performance. Therefore, the method proposed in this paper considers
39
40 the motor and transmission efficiencies for the optimal gear-shifting pattern, as
41 well as the inertia efficiency, which indicates the requested torque variation from
42
43 the inertia effect by the gear shift as follows.
44 For economic performance, the shifting pattern optimization problem con-
45
46 sidering only the static characteristic can be formulated to maximize the pow-
47
48 ertrain efficiency (ηm ηt ) according to the APS and vehicle speed (v) as follows:
49
50
51 maximize fE (P) = ηm ηt (APS, v)
P
52 (17)
 i N

53 where P = vopt , · · · , vopt , i = 1, · · · , N
54
55 where P is the shifting speed set, vopt is the optimal shifting speed, and su-
56
perscript i is the total number of APS cases (0−100%). The shifting pattern
57
58
59 16
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9 optimization that considers dynamic characteristics can be defined by including
10
the acceleration resistance torque variation effect to Eq. (17).
11
12 The motor torque request (Treq ) by APS is expressed as follows:
13
14
15 Treq = Tmax (ωm ) · APS (18)
16
17 Tmax is the variable value according to the motor speed (ωm ) and can be deter-
18
19 mined by the vehicle speed and driving gear as follows:
20
v
21 ωm = · ri rf (19)
22 Rt
23
24 Therefore, Treq can be determined if APS and v are provided. Because of r1 > r2
25 in the gear ratio design, from Eqs. (1) and (2), Jeq.1 is always greater than Jeq.2 .
26
27 Hence, driving with the first gear under acceleration is more disadvantageous in
28 terms of inertia resistance than driving with the second gear. To compare the
29
30 inertia effect with the motor and transmission efficiency, the inertia resistance
31
torque can be converted into an expression in a unit of efficiency expression
32
33 using the ratio of the resistance torque to the motor torque request (Treq ) as
34
35 follows:
36 Jvar αwhl
ηin = (20)
37 r2 rf Treq
38 where ηin is the inertia efficiency, which is the difference in efficiency when
39
40 shifting from the first-speed gear to the second. Since ηin is the relative value
41
of the first gear when driving in the second gear, as expressed in Eq. (17), it
42
43 is subtracted from fE (P) when only the first-speed gear is driven. Therefore,
44
45 the objective function fE (P) can be revised by considering the inertia effect, as
46 follows: 
47  η η (APS, v) − ηin (αwhl ) if ri = r1
 m t

48 
49 fE (P) = (21)
50



51
 ηm ηt (APS, v) if ri = r2
52 For dynamic performance, the shifting pattern optimization problem consid-
53
54 ering only the static characteristic can be formulated to maximize the driving
55
56
57
58
59 17
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9 torque (Tdrv ) according to the APS and vehicle speed (v) as follows:
10
11 maximize fP (P) = Tdrv (APS, v)
12 P
(22)
13  i
where P = vopt n
, · · · , vopt

, i = 1, · · · , N
14
15 From this optimization problem, the optimal P can be easily obtained by com-
16
17 bining the values of Tdrv for each speed gear. Using the first gear during low-
18
19 speed driving is generally advantageous in vehicle traction because a high motor
20 torque is multiplied by a large gear ratio. However, an evaluation of dynamic
21
22 performance should be conducted by considering the inertia variation effect be-
23 cause the acceleration that occurs under dynamic performance is greater than
24
25 that in the economic performance evaluation. Therefore, when the accelera-
26
tion is large, even at a low speed, the gear up-shift (first to second) may be
27
28 advantageous because it can significantly reduce the inertia resistance torque.
29
30 Unlike maximizing the efficiency for economic performance, the analysis of
31 dynamic performance, such as acceleration ability, should consider the time-
32
33 variant behavior of the vehicle. Therefore, it is advisable to directly evaluate
34 the acceleration ability through simulations using the developed EV model. To
35
36 consider Tdrv with the inertia effect for the optimal gear-shifting pattern, the
37
38 optimization problem can be expressed using Eq. (11) as follows:
39
40 minimize ta (P)
P
41 Z ta (23)
42 subject to a dt = vtar
43 0
44
45 where vtar is the speed of the target vehicle. In several previous studies, vtar
46 was adopted at 100 km/h, and the acceleration time (ta ) was evaluated under
47
48 WOT conditions. Therefore, Eq. (23) is acceptable for obtaining an optimal
49 gear-shifting speed at an APS of 100%.
50
51 Most APS values did not exceed 60% in the standard driving cycle, as shown
52
in Figure 7. Therefore, to obtain an optimal shifting pattern that considers eco-
53
54 nomic and dynamic performance, the optimal shifting speeds can be determined
55
56 by focusing on the economic and dynamic performance at low and high APS
57
58
59 18
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24 Figure 7: Example of APS values under the WLTP driving cycle.
25
26
values, respectively. Optimization problem that integrates Eqs. (21) and (23)
27
28 can be formulated as follows:
29 
30  maximize
 fE (P) if APS ≤ 60%
P
31 P= (24)
32  minimize
 ta (P) if APS = 100%
P
33
34 Here, optimal shifting speeds between 60% and 100% APS were obtained using
35  i 
n
36 the interpolation method. Since P = vopt , · · · , vopt must consist of ascending
37 speeds, along with an increase in APS because of the drivability problem, it
38
39 should satisfy the following constraint:
40
41 i−1
vopt i
≤ vopt , i = N, · · · , 2 (25)
42
43
44 An example of optimal shifting patterns based on Eqs. (24) and (25) consid-
45 ering inertia variation, is shown in Figure 8. As the wheel acceleration (αwhl )
46
47 increases, the optimal shifting lines gradually move toward the low-speed side.
48
Since a high αwhl involves a large inertia resistance when driven by the first gear,
49
50 the up-shifting is performed rapidly with increasing vehicle speed to reduce the
51
52 inertia resistance even in areas where the powertrain efficiency (ηm ηt ) of the
53 second gear is lower than that of the first gear. In addition, the shifting speed
54
55 of the WOT condition (100% APS) is constant regardless of αwhl because this
56
speed already considers the inertia variation effect at full acceleration based on
57
58
59 19
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
Figure 8: Example of optimal shifting patterns according to wheel acceleration (r1 = 3.321,
31
32 r2 = 1.681).
33
34 Table 4: Comparison of results for economic and dynamic performances.
35
36 Performance measure
Inertia effect
37 EFF (km/kWh) ta (s)
38
39 Consideration 6.35 10.69
40
None 6.24 10.99
41
42 Improvement (%) 1.83 2.74
43
44
45 Eq. (23). Table 4 compares the results of applying the optimal shifting pattern
46
47 considering inertia variation with those not considering it. These results demon-
48
strate the importance of the inertia variation effect of gear-shifting in obtaining
49
50 optimal shifting patterns for the energy efficiency (EFF) and acceleration time
51
52 (ta ) for the economic and dynamic performances of EVs, respectively.
53
54
55
56
57
58
59 20
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9 4. Optimization results of two-speed transmission
10
11 To maximize the economic and dynamic performances of EVs, it is appropri-
12
13 ate to utilize multi-objective optimization methods that can address the trade-
14
offs between performance objectives. The critical design parameters for opti-
15
16 mizing a two-speed transmission of an EV are the gear ratios of each speed
17
18 and shifting pattern. In particular, since the optimal shifting patterns for each
19 combination of gear ratios are different, they should depend on the gear ratios.
20
21 Therefore, the multi-objective optimization problem is formulated by employing
22 the objectives of EFF and ta as follows:
23
24
25
26 minimize f¯(−EFF, ta )
r,P
27 (26)
28 subject to r ∈ Ωc
29
30 where r = [r1 , r2 ] is the vector of gear ratios and Ωc is the feasible region of
31
32 gear ratios from the dynamic constraints. Here, the dynamic constraints are
33
34 adopted with a maximum speed (150 km/h) and an ascendable gradient (40%),
35 as determined by Eqs. (12) and (14), respectively. In addition, because a large
36
37 step ratio (r1 /r2 ) results in shifting difficulties [27], the step ratio was limited
38 to three or less. In the objective function, EFF is expressed as a negative value
39
40 (−EFF) because the optimization problem is formulated to minimize the ob-
41
jective function values, although EFF should be maximized for better economic
42
43 performance.
44
45 To solve the multi-objective optimization problem in Eq. (26), this study
46 employs a non-dominated sorting genetic algorithm (NSGA) [28]. Solving multi-
47
48 objective optimization problems requires significant computational effort com-
49 pared to solving single-objective optimization problems. Therefore, an artificial
50
51 neural network (ANN)-based multi-objective optimization method is used as an
52
53 alternative to this problem [29]. The ANN algorithm trains the given sample
54 results to derive the relationship between the input (design variable) and output
55
56 (target) and builds a predictive model which estimates the objective function
57
58
59 21
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9 values from the given design variable values of transmission without analyzing
10
the EV model. Therefore, it can significantly decrease the calculation effort ow-
11
12 ing to the excessively iterative calculations in the multi-objective optimization
13
14 process.
15 To confirm the inertia variation effect on the optimal gear-shifting pattern
16
17 for economic and dynamic performance, multi-objective optimizations were con-
18
ducted in two cases: i) considering the inertia variation effect and ii) not consid-
19
20 ering it. In addition, three cases (WLTP, UDDS+HWFET, and NEDC) were
21
22 used for the driving cycles for economic performance evaluation to validate
23 the effectiveness of shifting patterns considering inertia variation. The Pareto
24
25 fronts of the two cases were compared from the optimization results for each
26 driving cycle, as shown in Figure 9. These Pareto fronts depict the trade-off
27
28 between economic and dynamic performances. For all driving cycles, the values
29
of the optimum solutions show significant differences when the inertia variation
30
31 is considered. Applying the optimal shifting pattern considering inertia varia-
32
33 tion substantially outperforms the energy efficiency (EFF) and acceleration time
34 (ta ) compared to the case in which it does not. Therefore, these results demon-
35
36 strate that the inertia variation effect should be reflected in the optimization of
37
gear-shifting patterns.
38
39 Figure 10 illustrates the optimum gear ratios for each Pareto front, as shown
40
41 in Figure 9. These results indicate the following. First, combining low and high
42 gear ratios provides economic and dynamic performance advantages. A low
43
44 gear ratio is generally beneficial for economic performance owing to its small
45 equivalent inertia. However, the optimum gear ratios in Figures 10 (b) and (c)
46
47 are not in the lowest gear ratio area. Therefore, while a low gear ratio ensures
48
low equivalent inertia, it is more economically dominant in UDDS+HWFET and
49
50 NEDC to improve the overall efficiency by changing the motor operating points,
51
52 depending on the gear ratio and shifting pattern. For dynamic performance, a
53 high gear ratio is generally helpful because it can significantly multiply the
54
55 motor torque transferred to the wheels. However, the optimum gear ratios in
56
Figure 10 are not in the highest gear ratio area. Since a high gear ratio exhibits
57
58
59 22
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22 (a) WLTP (b) UDDS+HWFET
23
24
25
26
27
28
29
30
31
32
33
34
35
36 (c) NEDC
37
38 Figure 9: Comparison of Pareto fronts for each cycle.
39
40
41 high wheel torque and high acceleration resistance owing to the large equivalent
42 inertia, considering only the wheel torque is not appropriate for the vehicle’s
43
44 acceleration ability.
45
Next, the distributions of the optimum gear ratios vary significantly de-
46
47 pending on the optimal shifting pattern, considering or without considering the
48
49 inertia variation effect. In particular, in Figures 10 (a) and (b), the optimum
50 gear ratios derived by the optimal shifting pattern without considering the in-
51
52 ertia effect are more distributed in the low gear ratio area than in the pattern
53
considering the inertia effect, whereas the optimum gear ratios considering the
54
55 effect are more distributed in the high gear ratio area than not considering the
56
57
58
59 23
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21 (a) WLTP (b) UDDS+HWFET
22
23
24
25
26
27
28
29
30
31
32
33
34
35 (c) NEDC
36
37 Figure 10: Comparison of the optimum gear ratios for each cycle.
38
39
40 effect. This means that which shifting pattern is applied affects the EV perfor-
41
mance and optimum gear ratios. For the WLTP case in Figure 10 (a), Table 5
42
43 compares the results of the best solutions for the economic and dynamic per-
44
45 formances between the cases with and without the inertia effect. To confirm
46 the effect of the optimal shifting patterns on the performance values when using
47
48 different optimal shifting patterns that consider (Case 1) and do not consider
49
(Case 2) the inertia effect, the performance values of each combination of gear
50
51 ratios are evaluated. For economic and dynamic performances, when applying
52
53 the Case 1 pattern, the EFF and ta values of the gear ratios considering the
54 inertia effect were superior to the values not considering the effect. However,
55
56 when applying the Case 2 pattern, the EFF and ta values indicated the opposite
57
58
59 24
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9
Table 5: Comparison of the best solution results based on optimal shifting pattern in WLTP
10
11 (Case 1: considering inertia effect, Case 2: not considering).
12 Value
13 Objective Best solution (r1 /r2 )
14 Case 1 Case 2
15 EFF (km/kWh) A: 2.139/0.863 7.78 7.51
16 Inertia effect (blue)
17 ta (s) C: 3.894/1.636 10.61 10.94
18
EFF (km/kWh) B: 1.649/0.863 7.67 7.55
19 None (red)
20 ta (s) D: 3.894/1.298 10.67 10.91
21
22
23 tendency.
24
25 Figure 11 shows the optimal shifting patterns for the best solutions for EFF
26 in Table 4. Solutions A and B have the same second-gear ratio (0.863), whereas
27
28 the first-gear ratio of solution A (2.139) is greater than that of solution B (1.649).
29
Due to the large step ratio of solution A, it is possible to change the motor oper-
30
31 ating points more than that of solution B by shifting the speed gear to improve
32
33 the powertrain efficiency (ηm ηt ). However, because the equivalent inertia of the
34 first gear side of solution A is larger than that of solution B, it negatively af-
35
36 fects the acceleration resistance of solution A, as expressed in Eq. (15). When
37
comparing the optimal shifting patterns in Figure 11, the shifting speeds of so-
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53 (a) Solution A (r1 = 2.139, r2 = 0.863) (b) Solution B (r1 = 1.649,r2 = 0.863)
54
55 Figure 11: Optimal shifting patterns for economic performance (A: inertia effect, B: none).
56
57
58
59 25
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9 lution A are more distributed on the left side (low speed) at 60 km/h than those
10
of solution B. This means that the second gear driving in solution A is more
11
12 frequent than that in solution B. Since it reduces the acceleration resistance by
13
14 shifting the speed gear at lower speeds, solution A is superior to solution B in
15 terms of EFF, despite the larger equivalent inertia of the first gear side.
16
17 Figure 12 shows the driving torque at WOT condition in the best solutions
18
for ta in Table 4. Solutions C and D have the same first gear ratio (3.894),
19
20 whereas the second gear ratio of solution C (1.636) is greater than that of
21
22 solution D (1.298). Since the equivalent inertia of the second gear side of solution
23 C is larger than that of solution D, it has a negative effect on the acceleration
24
25 ability of solution C. When comparing the optimal shifting speeds, solution C
26 performs gear-shifting at a speed lower than solution D. In particular, the up-
27
28 shifting of solution C is conducted before the driving torque of the first gear
29
reaches the maximum driving torque of the second gear. Therefore, the total
30
31 driving torque in the red-colored area is insufficient compared to that of solution
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55 Figure 12: Comparison of driving torque for dynamic performance (C: inertia effect, D: none).
56
57
58
59 26
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9 D. However, although the Tdrv of solution C is noticeably lower than that of
10
solution D in the red-colored area, ta of solution C is superior to that of solution
11
12 D. This demonstrates that the inertia effect on the acceleration ability is greater
13
14 than that on the static driving torque. These results are particularly remarkable
15 in designs with large gear and step ratios. In summary, the equivalent inertia
16
17 variation should be considered in the design of the gear ratios and shifting
18
patterns for the performance of EVs.
19
20
21
22 5. Conclusion
23
24 This paper proposes an effective optimization method for the multi-speed
25
26 transmission of EVs considering the equivalent inertia variation for gear-shifting
27 patterns. The equivalent inertia for each speed was analyzed from the power
28
29 flow of each driving speed gear to confirm the equivalent inertia variation. An
30 EV analysis model with variable motor and transmission efficiencies was con-
31
32 structed, and the quantitative performance criteria were determined to evaluate
33
the economic and dynamic performances. From the EV analysis, the perfor-
34
35 mance results based on the optimal shifting pattern, considering the inertia
36
37 effect, were compared. These results demonstrate the importance of the inertia
38 variation effect by gear-shifting in obtaining optimal shifting patterns for the
39
40 energy efficiency and acceleration time for the economic and dynamic perfor-
41 mances of EVs, respectively.
42
43 Due to the trade-off relationship between economic and dynamic perfor-
44
45 mances, a multi-objective optimization problem, including the design variables
46 such as gear ratios and shifting patterns and the objective functions of en-
47
48 ergy efficiency and acceleration time, was formulated. In particular, to confirm
49 the inertia variation effect on the optimal gear-shifting pattern, optimizations
50
51 were performed by dividing the optimal shifting patterns considering the iner-
52
tia effect and not considering it. In addition, various driving cycles (WLTP,
53
54 UDDS+HWFET, and NEDC) were employed for economic performance eval-
55
56 uation to determine the effectiveness of shifting patterns considering inertia
57
58
59 27
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9 variation.
10
The optimization results demonstrate the importance of considering the in-
11
12 ertia variation effect for the optimal shifting pattern. In each driving cycle,
13
14 compared to the shifting pattern without the inertia variation effect, economic
15 and dynamic performances were improved from 2.7% to 7.8% and 2.8% to 3.0%,
16
17 respectively. In addition, the gear ratio values of the optimum solutions were
18
obtained differently when the inertia variation was considered and not. Ap-
19
20 plying the optimal shifting pattern considering inertia variation substantially
21
22 outperforms the energy efficiency (EFF) and acceleration time (ta ) compared
23 with the case in which it does not. Therefore, these results demonstrate that the
24
25 inertia variation effect should be reflected in the optimization of gear-shifting
26 patterns. In conclusion, considering the equivalent inertia variation from gear
27
28 ratios and shifting can achieve superior performance in the transmission design
29
of EVs.
30
31
32
33 Funding
34
35 This research did not receive any specific grant from funding agencies in the
36
37 public, commercial, or not-for-profit sectors.
38
39
40 References
41
42 [1] X. Hong, J. Wu, N. Zhang, B. Wang, Energy efficiency optimization of
43
44 simpson planetary gearset based dual-motor powertrains for electric vehi-
45 cles, Energy 259 (2022) 124908.
46
47
48 [2] B. Gao, D. Meng, W. Shi, W. Cai, S. Dong, Y. Zhang, H. Chen, Topology
49 optimization and the evolution trends of two-speed transmission of EVs,
50
51 Renewable and Sustainable Energy Reviews 161 (2022) 112390.
52
53 [3] J. Wu, X. Hong, G. Feng, Y. Zhang, Seamless mode shift control for a new
54
55 simpson planetary gearset based dual motor powertrain in electric vehicles,
56 Mechanism and Machine Theory 178 (2022) 105056.
57
58
59 28
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9 [4] Y. Liu, D. Gao, K. Zhai, Q. Huang, Z. Chen, Y. Zhang, Coordinated
10
control strategy for braking and shifting for electric vehicle with two-speed
11
12 automatic transmission, eTransportation 13 (2022) 100188.
13
14 [5] Y. Liu, Y. Chen, Z. Li, K. Zhao, Z. Lin, Multi-objective optimal gearshift
15
16 control for multispeed transmission electric vehicles, IEEE Access 8 (2020)
17
129785–129798.
18
19
20 [6] F. A. Machado, P. J. Kollmeyer, D. G. Barroso, A. Emadi, Multi-speed
21 gearboxes for battery electric vehicles: Current status and future trends,
22
23 IEEE Open Journal of Vehicular Technology 2 (2021) 419–435.
24
25 [7] Y. Liu, J. Xie, D. Qin, Y. Zhang, Z. Chen, G. Li, Y. Zhang, Design, control
26
27 and validation of two-speed clutch-less automatic transmission for electric
28 vehicle, IEEE/ASME Transactions on Mechatronics (2021).
29
30 [8] M. R. Ahssan, M. M. Ektesabi, S. A. Gorji, Electric vehicle with multi-
31
32 speed transmission: a review on performances and complexities, SAE In-
33
ternational Journal of Alternative Powertrains 7 (2) (2018) 169–182.
34
35
36 [9] P. Wu, P. Qiang, T. Pan, H. Zang, Multi-objective optimization of gear
37 ratios of a seamless three-speed automated manual transmission for electric
38
39 vehicles considering shift performance, Energies 15 (11) (2022) 4149.
40
41 [10] Y. Li, B. Zhu, N. Zhang, H. Peng, Y. Chen, Parameters optimization of two-
42
43 speed powertrain of electric vehicle based on genetic algorithm, Advances
44 in Mechanical Engineering 12 (1) (2020) 1–16.
45
46 [11] G.-B. Sun, Y.-J. Chiu, W.-Y. Zuo, S. Zhou, J.-C. Gan, Y. Li, Trans-
47
48 mission ratio optimization of two-speed gearbox in battery electric pas-
49
senger vehicles, Advances in Mechanical Engineering 13 (6) (2021). doi:
50
51 10.1177/16878140211022869.
52
53 [12] K. Kwon, J. Jo, S. Min, Multi-objective gear ratio and shifting pattern
54
55 optimization of multi-speed transmissions for electric vehicles considering
56
variable transmission efficiency, Energy (2021) 121419.
57
58
59 29
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9 [13] B. Zhu, N. Zhang, P. Walker, X. Zhou, W. Zhan, Y. Wei, N. Ke, Gear
10
shift schedule design for multi-speed pure electric vehicles, Proceedings of
11
12 the Institution of Mechanical Engineers, Part D: Journal of Automobile
13
14 Engineering 229 (1) (2015) 70–82.
15
16 [14] X. Lin, Y. Li, B. Xia, An online driver behavior adaptive shift strategy
17
for two-speed AMT electric vehicle based on dynamic corrected factor,
18
19 Sustainable Energy Technologies and Assessments 48 (2021) 101598.
20
21 [15] J. Ruan, P. D. Walker, N. Zhang, J. Wu, An investigation of hybrid energy
22
23 storage system in multi-speed electric vehicle, Energy 140 (2017) 291–306.
24
25 [16] P. D. Walker, S. Abdul Rahman, B. Zhu, N. Zhang, Modelling, simulations,
26
27 and optimisation of electric vehicles for analysis of transmission ratio se-
28 lection, Advances in Mechanical Engineering 5 (2013) 340435.
29
30 [17] T. Liu, Y. Zou, D. Liu, et al., Energy management for battery electric
31
32 vehicle with automated mechanical transmission, International Journal of
33
Vehicle Design 70 (1) (2016) 98–112.
34
35
36 [18] S. Tan, J. Yang, X. Zhao, T. Hai, W. Zhang, Gear ratio optimization of a
37 multi-speed transmission for electric dump truck operating on the structure
38
39 route, Energies 11 (6) (2018) 1324.
40
41 [19] J. Ruan, P. D. Walker, J. Wu, N. Zhang, B. Zhang, Development of contin-
42
43 uously variable transmission and multi-speed dual-clutch transmission for
44 pure electric vehicle, Advances in Mechanical Engineering 10 (2) (2018).
45
46 doi:10.1177/1687814018758223.
47
48 [20] C. Lin, M. Zhao, H. Pan, J. Yi, Blending gear shift strategy design and
49
comparison study for a battery electric city bus with AM, Energy 185
50
51 (2019) 1–14.
52
53 [21] K. Zhao, Z. Li, Y. Chen, Y. Liu, Multi-objective optimization of gearshift
54
55 trajectory planning for multi-speed electric vehicles, Engineering Optimiza-
56
tion 54 (6) (2022) 949–972.
57
58
59 30
60
61
62
63
64
65
 1
2
3
4
5
6
7
8
9 [22] K. Kwon, J. Lee, S. Min, Motor and transmission multi-objective optimum
10
design for tracked hybrid electric vehicles considering equivalent inertia of
11
12 track system, IEEE Transactions on Transportation Electrification 7 (4)
13
14 (2021) 3110–3123.
15
16 [23] H. Liu, Y. Lei, Y. Fu, X. Li, A novel hybrid-point-line energy management
17
18 strategy based on multi-objective optimization for range-extended electric
19 vehicle, Energy 247 (2022) 123357.
20
21
[24] Z. Lei, D. Sun, Y. Liu, D. Qin, Y. Zhang, Y. Yang, L. Chen, Analysis
22
23 and coordinated control of mode transition and shifting for a full hybrid
24
25 electric vehicle based on dual clutch transmissions, Mechanism and Machine
26 Theory 114 (2017) 125–140.
27
28
[25] X. Zhou, P. Walker, N. Zhang, B. Zhu, J. Ruan, Numerical and experimen-
29
30 tal investigation of drag torque in a two-speed dual clutch transmission,
31
32 Mechanism and Machine Theory 79 (2014) 46–63.
33
34 [26] L. Li, X. Wang, R. Xiong, K. He, X. Li, AMT downshifting strategy de-
35
sign of HEV during regenerative braking process for energy conservation,
36
37 Applied energy 183 (2016) 914–925.
38
39 [27] B. Gao, Q. Liang, Y. Xiang, L. Guo, H. Chen, Gear ratio optimization and
40
41 shift control of 2-speed I-AMT in electric vehicle, Mechanical Systems and
42
43 Signal Processing 50 (2015) 615–631.
44
45 [28] K. Deb, A. Pratap, S. Agarwal, T. Meyarivan, A fast and elitist multi-
46 objective genetic algorithm: NSGA-II, IEEE transactions on evolutionary
47
48 computation 6 (2) (2002) 182–197.
49
50 [29] K. Kwon, M. Seo, S. Min, Efficient multi-objective optimization of gear
51
52 ratios and motor torque distribution for electric vehicles with two-motor
53
and two-speed powertrain system, Applied Energy 259 (2020) 114190.
54
55
56
57
58
59 31
60
61
62
63
64
65