Performance of the Transmission Parking Mechanism of a Battery Electric Vehicle Simulated with Adams Software

  • Yuan Dong
  • Yong Chen
  • Chuang Yu
  • Zhan Cao
  • Guangxin Li
  • Zhuoqiang Li
  • Genqun Cui


The electric parking mechanism is studied for an electrically controlled two-speed auto transmission that is being developed for electric vehicles. Safety requirements include low-speed safe parking, reliable self-lock and the avoidance of abnormal parking. A dynamic model of the parking mechanism is established and analyzed using Adams software. Finally, failure of the parking mechanism due to wear is observed in bench testing and compared with experimental results after optimization.


Battery electric vehicle Two-speed auto transmission Parking mechanism Safety performance 

1 Introduction

The battery electric vehicle (BEV) is the main object of development of new energy vehicles in China. The vehicle is also an important energy-saving and emission–reduction means of China’s 13th Five-Year Plan. The BEV has been rapidly developed in recent years within the framework of the Chinese government’s support and preferential policies. In terms of improving the efficiency and decreasing the speed of the motor of the electric drive system, there has been a development trend of replacing a single machine reducer with high-performance two-stage decelerator in the transmission of some advanced vehicles. At the same time, major automobile corporations are more concerned about the safety of the BEV. More and more BEVs are equipped with parking mechanisms for greater safety and standardization.

The role of the parking mechanism is to park vehicle reliably. When the automatic transmission is shifted to the park (P) position, the vehicle drive wheel is locked through locking of the shaft or differential to prevent the vehicle from moving on flat or angled ground [1, 2, 3, 4].

Parking mechanisms can be classified as those of a mechanical drive, electrohydraulic drive and electric drive. The mechanical drive was widely used in early automatic transmission and continuously variable transmission. It relies on interconnected mechanical parts transmitting the force from the shift lever to the parking pawl to complete the parking lock action by locking the ratchet [1, 5]. The electrohydraulic parking mechanism has no mechanical connections between the parking pawl and shifting lever. The electrohydraulic system completely determines whether the parking pawl is engaged or disengaged. It usually depends on the position of the shift lever or other safety factors, such as the driver’s door being open, the transmission being in a working state or the ignition key being pulled out. The electric parking mechanism has no mechanical structure or hydraulic circuit between the shift lever and the actuator and controls the execution motor to complete the parking lock action via the shift lever position signal. The electric parking mechanism allows a more flexible internal structure of the transmission and easier control of the driver’s shift force.

In the non-working state, the automatic transmission is in the P position most of the time. Reliability is an important aspect of the performance of the transmission [6, 7] and has attracted the attention of researchers. Because parking on a ramp relates to the safety of the vehicle and the parking mechanism is a complex nonlinear multi-body system [8], it is an important consideration in the design of automatic transmission.

2 Structure of the Parking Mechanism

In this paper, the parking mechanism used in the two-speed automatic transmission is driven by a motor, and there is no need to consider the effect of the actuator part on the shift force. It is easy to control the shift force [9, 10, 11] and to ensure excellent shift comfort. The mechanical part can thus be simplified, while the reliability is substantially improved. The structure and location of the parking mechanism in the gearbox are shown in Figs. 1 and 2.
Fig. 1

Structural diagram of the parking mechanism

Fig. 2

Location of the parking mechanism in the gearbox

The parking mechanism is driven by the motor to complete the lock and takeoff action. The mechanical execution part is composed of a ratchet, pawl, slider, actuating rod, guide pins, press block and other components. The fork of the motor pushes the actuating rod back and forth through the sliding block, and the actuating rod then presses the block between the pawl and guide pins until the parking action is complete. The whole mechanism relies on the actuating rod spring realizing flexible engagement between the pawl and ratchet, and disengagement is completed by the torsion spring. It is seen that the electric drive greatly simplifies the parking mechanism relative to the situations of the mechanical drive and electrohydraulic drive.

3 Design and Calculation of the Parking Mechanism

As a safety device, the parking mechanism is required to meet four performance requirements: (1) the vehicle rolling distance needs to be less than 150 mm after parking; (2) parking must be at a safe speed; (3) the vehicle must be able to be locked on a slope of 30% and the transmission must then be able to exit the P position; and (4) the parking must not appear abnormal [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16].
Fig. 3

Ramp parking in two conditions. a Forward (downhill) and b Backward (uphill)

3.1 Rolling Distance in Ramp Parking

In general, a vehicle should be able to travel on various roads in different areas, have sufficient climbing capability and have the ability to park on a ramp. Obviously, as shown in Fig. 3, there are two conditions of ramp parking. If there is no additional explanation in the following articles, the condition uphill parking in 30% slope will be taken as example:
$$\begin{aligned}&S_{\mathrm{rollback}} =\frac{r\times 2\pi }{n_{\mathrm{teeth}} \times i_{\mathrm{diff}} } \end{aligned}$$
$$\begin{aligned}&v_{wl} =\sqrt{2e_1 S_{\mathrm{rollback}} } \end{aligned}$$
where \(e_1\) is the wheel acceleration, r is the wheel radius, \(n_{\mathrm{teeth}}\) is the number of ratchet teeth, \(i_{\mathrm{diff}}\) is the main reduction ratio. Based on correlation values, get the rolling distance of 34.52 mm, and reach a speed of 1.59 km/h. \(S_{\mathrm{rollback}}\) is the rolling distance of the wheel corresponding to the tooth spacing of the ratchet, and \(v_{wl}\) is the speed reached when a vehicle parks on the ramp, which must be less than the critical parking speed (which will be explained later). If \(v_{wl}\) exceeds the critical parking speed, there is a parking failure and a possible accident.

3.2 Critical Parking Speed

The critical parking speed needs to be within a reasonable range, usually between 2 and 5 km/h. The design principle is that the pawl cannot engage with the ratchet at a high speed, even if the transmission is shifted to the P position accidentally. This is because, once engaged, the wheel will be immediately locked, causing discomfort to the occupants and even a traffic accident. Furthermore, the vehicle must be able to be safely parked at low speed. Therefore, as mentioned in Sect. 3.1, the speed of a vehicle parking on a ramp needs to be lower than the critical parking speed. As described above, neither the control strategy nor the mechanical structure design is allowed to perform the parking action. Ramp parking therefore fails if the vehicle speed is higher than the selected critical parking speed. Figure 4 shows the rotation angle of the ratchet and pawl when they are able to engage.
Fig. 4

Rotation angle of the ratchet and pawl

In theory, whether the pawl and ratchet can engage is determined as follows. The parking process begins when the pawl teeth come into contact with the ratchet teeth. It can be engaged when the root of the pawl fillet contact with the root of the ratchet fillet [6]. The rotation angle is shown on this basis in Fig. 4. In this process, the angle of the ratchet is denoted \(\Delta \varphi \), and the angle of the pawl is denoted \(\gamma \). There is smooth engagement if the pawl rotation time is less than the ratchet rotation time. The parking function cannot otherwise be achieved by the parking mechanism. The critical parking speed is obtained when the two times are equal.

Suppose the rotation time of the ratchet is denoted \(t_1\), and the rotation time of the pawl is denoted \(t_2\) in the engagement process:
$$\begin{aligned} t_1= & {} \frac{\Delta \varphi }{\omega _{\mathrm{gear}} }, \end{aligned}$$
$$\begin{aligned} t_2= & {} \sqrt{\frac{2\gamma J}{T_1 }} \end{aligned}$$
where \(\omega _{\mathrm{gear}} \) is the ratchet angular velocity, J is the pawl moment of inertia and \(T_1\) is the pawl rotation torque and its value varies with the angle. The critical vehicle speed is calculated to be 3.3 km/h.
Fig. 5

Force state of the pawl in the engaged position during the parking process

Fig. 6

Force state of the press block in the engaged position during the parking process

Taking the important position in the process of vehicle parking as an example, the analysis process is introduced as follows: The force state of the pawl in the locked position is shown in Fig. 5. According to the force balance, the torque provided by the actuating rod spring needs to overcome the pawl rotation torque, torsion spring torque and friction torque. Similarly, Fig. 6 shows the force state of the press block. According to the force balance condition,
$$\begin{aligned}&F_N \times R_a -F_F \times R_b -M_r =T_1 \end{aligned}$$
$$\begin{aligned}&F_S =F_{N1} sin\beta +F_{F1} cos\beta +F_{F2} +F_Z , \end{aligned}$$
where \(F_N\) is the vertical pressure of the press block acting on the pawl, \(F_F\) is the friction force of the press block acting on the pawl, \(R_a, R_b\) are the arms of force \(F_N\) and \(F_F\), \(M_r\) is the torque of the torsion spring; \(T_1\) is the pawl rotation torque, \(F_S \) is the force of the actuating rod spring; \(F_{N1}\) is the vertical pressure of the pawl acting on the press block; \(\beta \) is the angle between the contact surface and horizontal plane, and its value is \(10.5^{\circ }, F_{F1}\) is the friction force of the pawl acting on the press block, \(F_{F2}\) is the friction force of the guide pins acting on the press block, and \(F_Z\) is the force that the press block requires for acceleration. The spring preload force is determined to be 20 N.

The rotation time of the pawl can be adjusted using the spring force of the actuating rod. Usually, the spring force for which the pawl can engage smoothly with the ratchet is first calculated and then further modified using the simulation results.

3.3 Self-Locking Performance

After the vehicle is parked, it must be reliably kept in place, especially on a ramp. Sliding may result in occupant injury and property damage. After parking, the ratchet and pawl are engaged and the press block is in contact with the pawl and guide pins. Taking uphill parking as an example, the force state of the pawl and ratchet is shown in Figs. 7 and 8.
Fig. 7

Force state of the pawl when parking on a ramp

Fig. 8

Force state of the press block when parking on a ramp

Along the transfer path of the force, the self-locking performance is analyzed considering the pawl and press block separately. According to the force balance condition,
$$\begin{aligned}&F_{N3} \times R_3 +F_{F3} \times R_4 =F_{n1} \times R_1 -F_{f1} \times R_2 +M_r \nonumber \\\end{aligned}$$
$$\begin{aligned}&F_{\mathrm{hold}} =F_{F3}^{\prime } \times \cos \beta -F_{N3}^{\prime } \times \sin \beta +F_S +F_{F4} \end{aligned}$$
where \(R_1, R_2, R_3\), and \(R_4\) are the arm of each force, \(F_{N3}\) is the vertical pressure of the press block acting on the pawl, \(F_{F3}\) is the friction force of the press block acting on the pawl, \(F_{n1}\) is the vertical pressure of the ratchet acting on the pawl, \(F_{f1}\) is the friction force of the ratchet acting on the pawl, \(F_{F3}^{\prime }\) is the friction force of the pawl acting on the press block, \(F_{N3}^{\prime }\) is the vertical pressure of the pawl acting on the press block. The value of \(F_S\) can be calculated, \(F_{F4}\) is the friction force of the guide pins acting on the press block, \(F_{\mathrm{hold}}\) is the holding force for the press block, which is the force that the mechanism can provide to engage minus the force exerted by the pawl on the press block for pulling out.
In theory, if \(F_{\mathrm{hold}} \ge 0\), the mechanism will be self-locking. Considering special cases, \(F_{\mathrm{hold}}\) needs to be greater than a certain value for reliable self-locking. The holding force is calculated to be 45 N.
Fig. 9

Position of the pawl in the non-parking condition

3.4 Prevent Abnormal Parking

Figure 9 shows the position of the pawl when it disengages with the ratchet. When setting the installation torque of the pawl torsion spring, it is necessary to consider bumps of the vehicle in driving to prevent accidental engagement. When the vehicle travels a bumpy section of road at high speed, the acceleration of the pawl due to a heavy bump can be much larger than that in normal parking. Accidental parking in this scenario is dangerous and must be prevented by the torsion spring having sufficient installation torque:
$$\begin{aligned} M_p \times a\times b\times k-M_r \le 0 \end{aligned}$$
where \(M_p\) is the pawl mass, a is the pawl vibration acceleration, b is the distance from the center of the pawl to the center of the pivot axis, and k is a safety factor. Values of a and k are selected according to design requirements. The left side of Eq. 9 is − 202.68 N/mm, which meets the design requirements.

4 Simulation Analysis of the Parking Mechanism

A dynamic simulation analysis model was established using Pro/E and Adams, which are multi-body dynamics software [17, 18, 19], as shown in Fig. 10. There are 20 constraints, comprising 5 revolving joints, 13 fixed joints and 1 translational joint and 1 gear pair. A translational spring–damper, a rotational spring–damper and some necessary contacts are established in the model. According to the literature and empirical values, simulation parameters are given in Table 1.
Fig. 10

Simulation model

Table 1

Simulation parameter settings



Stiffness (N/mm)


Force exponent


Damping (N s/mm)


Penetration depth (mm)


Static coefficient


Dynamic coefficient


Stiction transition Vel. (mm/s)


Friction transition Vel. (mm/s)


4.1 Simulation of the Critical Vehicle Speed

Using the dynamics simulation environment that Adams provides, the initial state and the actions to be performed are set for the established simulation model. An angular acceleration is applied to the ratchet using the STEP function [17, 18, 19]. A sensor is set to stop acceleration when the ratchet speed is equivalent to 6 km/h, the speed at which the vehicle travels. The parking action is simulated by applying a drive to the fork of the motor. When the pawl is engaged with the ratchet, the ratchet stops and the simulation ends. The step in the pawl velocity curve shows the critical vehicle speed.

The Y-axis in Fig. 11 represents the angular velocity of the ratchet. The ratchet does not accelerate after reaching the desired angular velocity, and the angular velocity remains stable when the engagement gap is removed. The ratchet and pawl then begin to come into contact, but they can only partly engage and there is ejection soon after. This process repeats many times as shown by the fluctuation of the curve. The speed of the vehicle decreases gradually and approaches the critical speed. At 0.56s, the ratcheting speed drops appreciably. Although there are fluctuations, the parking is complete. At 0.56s, the angular velocity of the ratchet wheel is 660.5 deg/s and the critical speed is 3.4 km/h, which meets the design requirements.
Fig. 11

Critical vehicle speed simulation

4.2 Self-Locking Performance Simulation

To obtain the self-locking performance, the ratchet is engaged with the pawl, and the parking mechanism remains in the parking state. Torque is applied to the ratchet, whose value is equivalent to the torque generated at the ratchet shaft when the vehicle is parked on a 30% slope with full load (GVM) [6, 13].
Fig. 12

Self-locking simulation for the uphill condition

Fig. 13

Self-locking simulation for the downhill condition

In Figs. 12 and 13, the torque loading process of the ratchet is shown. The Y-axis on the left represents torque applied to the ratchet wheel, while the Y-axis on the right represents the angle of the pawl. The torque curve remains stable after reaching the desired value. With an increase in torque, the angle of the pawl is almost zero and does not change. It is seen that the actuator rod does not come out under the torque for either condition, which meets the self-locking performance requirements. In subsequent simulation, the torque is increased and the ramp parking safety factor is determined to be 2.8. As shown in Fig. 14, the pawl is completely detached from the ratchet wheel at 0.11s and no longer self-locking.
Fig. 14

Self-locking safety factor

4.3 Simulation of the Disengagement Performance

When the vehicle stops on the ramp and the transmission is in the P position, the vehicle will sometimes move a short distance. The main reason for the nonzero rolling distance is that when the transmission is shifted to the P position, the pawl is generally on top of the ratchet wheel and will engage completely after the ratchet wheel rotates a certain angle. The vehicle stops moving when the ratchet wheel and pawl come into contact and produce a torque that prevents movement of the vehicle. In some cases, it is the contact force that prevents the parking mechanism from disengaging smoothly. The aim of the simulation of the disengagement performance is to verify whether the transmission can exit the P position smoothly when needed.

Specifically, the simulation verifies whether the pawl successfully disengages from the ratchet when the vehicle is parked on a 30% slope with a full load (GVM) after the press block and actuating rod are removed.

Torque is applied to the ratchet and has a value equivalent to the torque generated at the ratchet shaft when the vehicle is parked on a 30% slope with full load (GVM). The fork is then moved to pull out the actuator rod.
Fig. 15

Disengagement simulation for the uphill condition

At 0.1–0.2s, the fork turns and pulls out the actuator rod. As shown in Fig. 15, at 0.2 s for the uphill condition, the contact force quickly reduces to zero after slight fluctuation, while the pawl quickly ejects. In Fig. 16, similarly, there is disengagement at 0.21 s for the downhill condition. Simulation shows that the disengagement performance meets requirements.
Fig. 16

Disengagement simulation for the downhill condition

4.4 Simulation of the Parking Effect on the Motion State

The main purpose of this simulation is to analyze whether the parking mechanism will be destroyed in the event of high-speed parking and to provide a reference for design and experiments.
Fig. 17

Relationship between the impact load and vehicle speed

Simulation results presented in Fig. 17 show that the impact load is higher at speeds below 10 km/h and above 60 km/h. In the low-speed stage, the coinciding area of the movement track between the pawl and ratchet is relatively large, owing to the low ratchet speed. The pawl is deep into the ratchet, and the part that comes into contact changes from the fillet of the pawl to the plane above it. The impact arm is therefore shorter and the impact force higher. In mid- and low-speed stages, with an increase in the ratcheting speed, the coinciding area of the movement track decreases rapidly, which means that the area of the pawl in the ratchet is greatly reduced. The impact force is thus lower. In the high-speed stage, an increase in speed increases the impact force. This phenomenon is more obvious above 30 km/h. The above peak impact load is in the design safety range with a safety factor of 2.

5 Experimental Optimization of the Parking Mechanism

Parking experiments are a performance experiment and a reliability and durability experiment. Some tests are completed with the aid of the test bench, excluding critical vehicle speed test. Figure 18 shows the parking mechanism test bench with an integrated base. The bench can be moved after installation and commissioning. The transmission is installed on the retainer plate, the left half-axle is locked with the locking mechanism and the torque is loaded on the right half-axle by the loading motor and reducer.
Fig. 18

Test bench of the parking mechanism

Fig. 19

Pre-optimized ratchet wear

Fig. 20

Pre-optimized pawl wear. a Front view and b Back view

Another developed automatic transmission parking mechanism is taken as an example. In the experiments, the parking mechanism initially runs well. However, there are disengagement difficulties after a long period of use and wear. The bench test program is designed to verify the durability and reliability of the parking mechanism. In the program, the slope is simulated by different loads, and ramp tests are completed in turn. The durability test is conducted for a 30% equivalent slope. Adjacent tests are kept at regular intervals.

Figures 19 and 20 show that, after the durability test, much engaging and disengaging have worn the contact rounds of the ratchet. Correspondingly, wear and scratches appear on the front and rear contact surfaces of the pawl. It is seen that the partial grinding of the ratchet wheel is serious on the left side of Fig. 18. The reason is that the stiffness of the cantilever shaft of the pawl is insufficient and deformation appears along the direction of the force. This phenomenon is eliminated in the subsequent improvement.

Generally, the larger the radius of the ratchet, the lower the contact force and the less the wear, and the greater the room for the improved design of the parking mechanism. However, under the premise that the overall size of the transmission has been determined, the ratchet radius is not allowed to increase. Therefore, the improvement direction of the program is an extension of the contact arm of the pawl, increase in the contact fillet radius, change in the friction coefficient, change in the tooth contact angle and replacement of the material.

After a series of tests, the effective and economical solution is found to be an increase in the length of the pawl contact arm, strengthening of the surface for the ratchet, no change to the friction coefficient, and a change from wire contact to surface contact by adjusting the relative positions of the ratchet and pawl. To ensure the critical vehicle speed, the contact fillets are maintained at R1.5 and R1. Figures 21 and 22 show that the wear condition of the ratchet and pawl improves greatly, which ensures normal use of the parking mechanism.

6 Conclusions

On the basis of an electrically controlled two-speed transmission being developed for the BEV, this paper presents theoretical calculations and investigates the parking mechanism in an Adams simulation of the safety performance. Finally, in a bench test, the parking mechanism is shown to face failure because of wear. The main results of the study are as follows:
  1. (1)

    An electric drive, compared with a mechanical drive or electrohydraulic drive, greatly simplifies the parking mechanism.

  2. (2)

    Safety requirements are met, and relevant performance indicators are determined through reasonable and systematic design and calculation.

  3. (3)

    A simulation model is established using multi-body dynamics software, and the correctness of theoretical calculation is verified.

  4. (4)

    A common problem faced by parking mechanisms wear failure is reproduced in experiments, and a feasible solution is presented.

Fig. 21

Optimized ratchet wear

Fig. 22

Optimized pawl wear. a Front view and b Back view

This paper provides a reference for the design and development of the parking mechanism.



The authors acknowledge financial support from the Science and Technology Research Youth Fund Project of Hebei Colleges and Universities (QN2016197).


  1. 1.
    Fei, N.Z., Li, X.C., Gu, L.Q.: A transmission parking brake optimization research. Drive Syst. Tech. 30(2), 17–21 (2016)Google Scholar
  2. 2.
    Liu, S., Zeng, J.: The performance study and simulation for optimization of Park Brake in automatic transmission. In: SAE-CHINA Congress Proceedings, vol. 1 (2015)Google Scholar
  3. 3.
    Deng, Q.B., Wang, X.J., Li, P.: Design and kinematics simulation of parking Pawl based on ADAMS. In: SAE-CHINA Congress Proceedings, vol. 1 (2015)Google Scholar
  4. 4.
    Fei, N.Z.: A dual clutch transmission parking brake optimization design and research. Shanghai Jiaotong University, Shanghai (2015)Google Scholar
  5. 5.
    Chen, H., Chen, X.B., Fu, L.L., et al.: Performance study of parking mechanism of auto transmission. Tract. Farm Transp. 4, 008 (2014)Google Scholar
  6. 6.
    Fu, L., Qiu, Z., Chen, Y., et al.: Research on safety performance for parking mechanism on a 7-speed dual clutch transmission. In: Proceedings of the FISITA 2012 World Automotive Congress, pp. 163–173. Springer, Berlin (2013)Google Scholar
  7. 7.
    Wang, M.C.: Design and integration of key components CA7CH350D wet dual-clutch automatic transmission. Automob. Technol. 3, 5–8 (2015)Google Scholar
  8. 8.
    Chen, Y., Luo, D.G., Fu, J., et al.: The development practices of mass produced dual clutch automatic transmission based on CAE simulation. Automot. Eng. 35(11), 1109–1120 (2013)Google Scholar
  9. 9.
    Zhang, Y.W., Li, H.Q.: Design of P-R shift performance of AT parking mechanism based on RecurDyn. Light Veh. Technol. 10, 34–37 (2012)Google Scholar
  10. 10.
    Hu, Z.H., Chen, Y.X., Chen, Y.: The method of multi-body dynamics evaluation for the shiftability of manual transmission. SAE-CHINA Congress (2011)Google Scholar
  11. 11.
    Chen, Y.X.: Simulation analysis on the shifting performance of mechanical transmission based on virtual prototype technology. South China University of Technology, Guangzhou (2012)Google Scholar
  12. 12.
    Yamaoka, T., Kobayashi, A., Nogami, S., et al.: Development of parking lock system behavior analysis. Honda R D Tech. Rev. 18(1), 114 (2006)Google Scholar
  13. 13.
    Wang, L.: Structure design and analysis of pure electric vehicle gear box. Jilin University, Changchun (2015)Google Scholar
  14. 14.
    Lechner, G., Naunheimer, H.: Automotive Transmissions. Springer, New York (1999)Google Scholar
  15. 15.
    Rinderknecht, S., Meier, T., Rot, I.: Electric two-drive-transmission with integrated parking lock for electric and range extender vehicles. In: 21st Aachen Colloquium Automobile and Engine Technology (2012)Google Scholar
  16. 16.
    The SAE Automatic Transmission and Transaxle Technical Standards Committee. SAE J2208. Park Standard for Automatic Transmissions. Warrendale PA: SAE Publications (2007)Google Scholar
  17. 17.
    Yu, D.Y., Qian, Y.J.: Parameter-settings for the dynamic simulation based on adams. Comput. Simul. 23(9), 103–107 (2006)Google Scholar
  18. 18.
    Chen, L., Ren, W., Zhang, Y.: Dynamic Analysis of Mechanical System and Application Tutorial of ADAMS. Publishing Company of Qinghua University, Beijing (2005)Google Scholar
  19. 19.
    Guo, W.D., Li, S.Z., Ma, L.: Adams 2013 Advanced Application Tutorial with Examples. China Machine Press, Beijing (2013)Google Scholar

Copyright information

© Society of Automotive Engineers of China (SAE-China) 2018

Authors and Affiliations

  • Yuan Dong
    • 1
  • Yong Chen
    • 1
  • Chuang Yu
    • 2
  • Zhan Cao
    • 1
  • Guangxin Li
    • 1
  • Zhuoqiang Li
    • 1
  • Genqun Cui
    • 1
  1. 1.School of Mechanical EngineeringHebei University of TechnologyTianjinChina
  2. 2.Powertrain Research InstituteGeelyHangzhou BayChina

Personalised recommendations