1. Regenerative Braking Algorithm For a HEV with CVT Ratio Control During Deceleration
Hoon Yeo, Donghyun Kim, Sungho Hwang, Hyunsoo Kim
Sungkyunkwan University
Dynamic System Design & Control Lab.
Sungkyunkwan University

2. Introduction Motor IC Engine Battery Generator + - Transmission
In regenerative braking, kinetic and potential energy of the vehicle is stored in an energy storing device instead of being dissipated as the heat by the friction braking system and the stored energy is recycled to drive the vehicle.
In a hybrid electric vehicle, the energy recuperation takes place by transforming the kinetic energy into the electric energy via a generator. + -
Battery
Regenerative Braking

3. Traction and Braking Energy Consumption
EPA 75
Urban
Austr.
ECE-15
Japan
1015
New
York
City
Traction energy KJ
Total
4000
6480
3478
1675
998
Per
km.
377
606
437
402
519
Braking energy
1934
4195
953
888
878
182
392
120
213
461
Percentage of braking energy to total traction energy (%)
48.3
64.7
27.5
53.0
88.8
Recovering vehicle kinetic or potential energy is inherent advantage of hybrid electric vehicle(HEV) which has energy storing device such as battery or ultracapacitor.
When a vehicle drives in a city, braking energy dissipated in the brakes reaches up to half of the total tractive energy.
For example, the braking energy ratio to the total traction energy is 48.3 percent for FUDS(federal urban driving schedule) and 53 percent for Japan mode

4. + the regenerative braking force is not large enough to
FRICTION
BRAKING +
the regenerative braking force is not large enough to
cover the required braking force
the regenerative braking can not be used for many
reasons such as high state of charge or high
temperature of the battery to increase the battery life
Generally, the regenerative brake works together with the conventional friction brake for the following reasons :
(1) the regenerative braking force is not large enough to cover the required braking force,
(2) the regenerative braking can not be used for many reasons such as high state of charge(SOC) or high temperature of the battery to increase the battery life.
In these cases, the conventional friction brake works to supply the required braking force.

5. Objectives Develop a regenerative braking control algorithm
Develop a prototype electro-hydraulic controlled
regenerative braking module
Propose a CVT ratio control algorithm to obtain maximum regenerative energy during deceleration
In this study, a regenerative braking control algorithm is developed
and a prototype electro-hydraulic controlled regenerative braking module is designed
and performance of the regenerative braking algorithm and hydraulic module using HIL simulator are investigated.

6. Hybrid Electric Vehicle
REGEN
Hydraulic
Module
+ -
Battery
Motor &
Generator
Pedal
Inverter
Engine
MCU
This Figure is a parallel hybrid electric vehicle structure used in this study.
The engine is connected to a 10kW motor with a single shaft.
The power from the engine and motor is transmitted to a transmission via clutch.
As a transmission, a metal belt continuously variable transmission(CVT) is used to maintain the engine operation on the minimum fuel consumption region independent of the vehicle speed.
The front wheel braking is performed by using both a regenerative brake and a hydraulic brake meanwhile the rear wheel braking is carried out only by a hydraulic brake.
In the hydraulic brake, a hydraulic pressure is supplied to a wheel cylinder so as to generate a brake force in response to a braking operation by a driver.
Whereas, in the regenerative brake, a brake force is generated by a regeneration action.
BCU
HEV-ECU

7. Regenerative Braking Hydraulic Module
This figure shows a schematic diagram of the regenerative braking hydraulic module developed in this study.
When a driver pushes a brake pedal, pedal force is transmitted to master cylinder through a vacuum booster.
The master cylinder pressure generated by the brake pedal is supplied to the rear wheel cylinder.
In the front wheel, the regenerative braking is performed corresponding to the required braking force which is calculated from a brake control unit(BCU).
The BCU calculates the regenerative braking force using the signals of the battery state of charge(SOC), CVT ratio, vehicle velocity and wheel cylinder pressure.
The regenerative hydraulic module consists of powerpack, accumulator, relief valve and proportional valve.
When the regenerative braking is applied, the hydraulic oil supply from the master cylinder to the front wheel actuator is cut off, and the reduced hydraulic pressure is supplied from the powerpack instead.

8. T N i =  W W = W W = W Regenerative Braking Algorithm (SOC)
Regenerative torque applied to the front wheel
REG R T N i • =  1 W 2 W = W
(SOC)
In order to provide an appropriate regenerative braking for the given driving conditions, a control algorithm is required to determine the magnitude of the regenerative torque with respect to the battery SOC, motor speed and etc. corresponding to the driver’s demand.
The regenerative torque applied to the front wheel TR is represented like this
where i is the CVT speed ratio, N is the final reduction gear ratio, TREG is the regenerative torque by the motor, which is determined from the motor characteristic curve for a given speed. W1, W2 are the weight factors. 1 1 W = W
(Velocity) 2 2

9. Weight Factor for Regenerative Braking
Weight factors W1 and W2 are determined by considering the battery SOC and vehicle velocity.

10. Regenerative Braking Algorithm
Regenerative braking force
Hydraulic braking force required at the front wheel
Ff HYD = Fbf – FREGEN
Front wheel cylinder pressure equivalent to FfHYD
The regenerative braking force is obtained like this
If FREGEN is larger than the required front wheel braking force Fbf, the front wheel is braked only by the regenerative brake.
Otherwise, the hydraulic brake works together with the regenerative brake. The hydraulic braking force required at the front wheel is calculated like this
The front wheel cylinder pressure equivalent to FfHYD is expressed with this equation
where Rt is the brake effective radius, a is the wheel cylinder area, is the brake friction coefficient.

11. Braking Forces on the Front and Rear Wheel
If the braking forces on the front and rear wheel follow the curve in this figure, the maximum braking stability can be achieved since the braking forces reach the adhesive limitation between the road and the tires.
In this figure, if the point P is chosen for the driver’s braking command, i.e, the given deceleration, the front and rear braking forces are obtained as Fbf and Fbr, respectively.

12. Flow Chart for Regenerative Braking
Yes No F bf
> F
REGEN
Battery SOC,
Pedal input
Velocity,
Regenerative braking
CVT ratio +
Regenerative braking
Rear pressure
Friction braking
Friction braking
Regenerative torque,
Rear braking force T P
= 0
REGEN F
bfFRICTION = F bf - F
REGEN F
iNT T =
REGEN W W
Ideal distribution of braking force R  1 2 R F p = t
bfFRICTION F
This figure shows the regenerative braking algorithm.
For the driver’s pedal input, the rear wheel cylinder pressure is obtained from the pressure sensor.
The required front braking force Fbf is determined from the given rear braking forcer. In the mean time, BCU calculates the regenerative force FREGEN for the input signals of the battery SOC, vehicle velocity and CVT ratio by considering the weight factors, W1 and W2 and the generation efficiency .
If the demanded front braking force is less than FREGEN, only the regenerative brake is applied and the magnitude of the regenerative braking force is limited not to exceed the demanded braking force.
In case that the demanded front braking force Fbf is larger than FREGEN, the hydraulic brake works together with the regenerative brake. bf
= F
REGEN f
Front braking force, F 2 rA 
Regenerative force F bf b
REGEN
Hydraulic module
bfFRICTION F

13. CVT Ratio Control Motor efficiency
This figure shows the regenerative braking algorithm.
For the driver’s pedal input, the rear wheel cylinder pressure is obtained from the pressure sensor.
The required front braking force Fbf is determined from the given rear braking forcer. In the mean time, BCU calculates the regenerative force FREGEN for the input signals of the battery SOC, vehicle velocity and CVT ratio by considering the weight factors, W1 and W2 and the generation efficiency .
If the demanded front braking force is less than FREGEN, only the regenerative brake is applied and the magnitude of the regenerative braking force is limited not to exceed the demanded braking force.
In case that the demanded front braking force Fbf is larger than FREGEN, the hydraulic brake works together with the regenerative brake.

14. CVT Ratio Control Optimal operating line in regenerative braking
This figure shows the regenerative braking algorithm.
For the driver’s pedal input, the rear wheel cylinder pressure is obtained from the pressure sensor.
The required front braking force Fbf is determined from the given rear braking forcer. In the mean time, BCU calculates the regenerative force FREGEN for the input signals of the battery SOC, vehicle velocity and CVT ratio by considering the weight factors, W1 and W2 and the generation efficiency .
If the demanded front braking force is less than FREGEN, only the regenerative brake is applied and the magnitude of the regenerative braking force is limited not to exceed the demanded braking force.
In case that the demanded front braking force Fbf is larger than FREGEN, the hydraulic brake works together with the regenerative brake.

15. CVT Ratio Control Flow chart of CVT ratio control
This figure shows the regenerative braking algorithm.
For the driver’s pedal input, the rear wheel cylinder pressure is obtained from the pressure sensor.
The required front braking force Fbf is determined from the given rear braking forcer. In the mean time, BCU calculates the regenerative force FREGEN for the input signals of the battery SOC, vehicle velocity and CVT ratio by considering the weight factors, W1 and W2 and the generation efficiency .
If the demanded front braking force is less than FREGEN, only the regenerative brake is applied and the magnitude of the regenerative braking force is limited not to exceed the demanded braking force.
In case that the demanded front braking force Fbf is larger than FREGEN, the hydraulic brake works together with the regenerative brake.

16. HEV Powertrain Model
Dynamic models of the HEV powertrain are developed based on MATLAB SIMULINK.
In this figure, MATLAB model of the HEV powertrain is shown.

17. Vehicle Data
In this Table, vehicle data used in the HILS are shown.

18. Simulation Results
This pictrue shows the hardware part of the HIL simulator.
The hardware includes the 4 wheel calipers, master cylinder, booster, pressure sensor and hydraulic module.
The brake pedal is actuated by the operator.
The brake pedal force is boosted by the vacuum pump.
The rear wheel cylinder pressure is generated from the master cylinder depending on the brake pedal force.
Pressure sensor is used to measure the master cylinder pressure.

19. Simulation Results
This pictrue shows the hardware part of the HIL simulator.
The hardware includes the 4 wheel calipers, master cylinder, booster, pressure sensor and hydraulic module.
The brake pedal is actuated by the operator.
The brake pedal force is boosted by the vacuum pump.
The rear wheel cylinder pressure is generated from the master cylinder depending on the brake pedal force.
Pressure sensor is used to measure the master cylinder pressure.

20. Simulation Results Motor Operation Trajectories for FUDS
This pictrue shows the hardware part of the HIL simulator.
The hardware includes the 4 wheel calipers, master cylinder, booster, pressure sensor and hydraulic module.
The brake pedal is actuated by the operator.
The brake pedal force is boosted by the vacuum pump.
The rear wheel cylinder pressure is generated from the master cylinder depending on the brake pedal force.
Pressure sensor is used to measure the master cylinder pressure.

21. Simulation Results
Comparison of fuel economy and final battery SOC for FUDS
This figure shows a block diagram of the HILS. For a given driving mode, the driver operates the drive pedal to follow the target velocity.
In the HILS, the driver operates the brake pedal by watching the vehicle velocity displayed on the monitor in the real time.
In the braking, the driver’s demand is estimated from the rear wheel pressure measured by the pressure sensor and is transformed into the braking force in the HEV simulator.
The vehicle velocity is reduced corresponding to the braking force in the simulator. In case of the regenerative braking, the regenerative braking force is calculated in the BCU according to the regenerative braking algorithm.

22. Conclusion Regenerative braking algorithm is proposed.
Prototype electro-hydraulic regenerative braking
module is developed.
CVT ratio control algorithm during deceleration is suggested.
Fuel economy is improved by 4 percent for FUDS
A regenerative braking algorithm and a prototype electro-hydraulic regenerative braking module are developed for a parallel hybrid electric vehicle.
In order to evaluate the regenerative braking algorithm and hydraulic module, a hardware in the loop simulation(HILS) is performed.
It is found from the hardware in the loop simulation results that the regenerative braking algorithm and hydraulic module developed in this study provide satisfactory braking performance while maintaining the battery SOC level.