Final Project: Design of an Electric Vehicle

Objective:

The objective of this project is to design, model and simulate the fundamental components of an Electric vehicle using Simulink. We will primarily focus on the powertrain components of an electric vehicle. This will help us to understand how the sizing of different components of an electric vehicle is done and their operation. In this project, a typical micro-car is modelled.

Components of EV:

The main powertrain components of an EV includes a motor, motor controller, battery (energy source) & drive-train. Figure 1 shows the block diagram of the powertrain of an electric vehicle.

Figure 1

The battery acts as the energy source to run the vehicle. Earlier, Lead-acid batteries were used in electric vehicles but now lithium-based batteries have become the popular choice for electric vehicles. Because Lithium-ion batteries offer higher performance as well as better energy density than lead-acid batteries. A motor controller is used to run the motor according to the driver’s acceleration/braking command. In case of an AC motor, it also acts as an inverter. Majority of the electric vehicle until now have been using DC motors, BLDC motors, or induction motors. The major advantage of DC or BLDC motor is their simple operation and easy to implement control. However, they are more expensive compared to an induction machine because they use rare earth magnets. Induction machines are becoming a popular choice for electric vehicles because of their low-cost and low-maintenance. With the advancement of innovation in power electronics, their control has also become a lot easier and cheaper to implement. Majority of electric vehicles use a single-speed transmission coupled to the electric motor as electric motors are capable of providing instant torque over a wide range of speed from start.

Modelling:

Figure 2. EV powertrain model in Simulink.

A Simulink model is made to model the EV powertrain, shown in figure 2. There is a ‘Vehicle’ subsystem block which implements the transmission, tire and vehicle body parameters like weight, Drag coeff., frontal area, etc. ‘Motor & Controller’ subsystem includes a DC motor and H-bridge controller to control the motor according to acceleration or deceleration command received from the driver block. Longitudinal Driver block implements a PI controller which takes vehicle speed as feedback and tries to match the reference speed by giving appropriate acceleration or braking command. ‘Battery’ subsystem models a battery which is drained with the same amount of current input to the motor and outputs battery SOC (state of charge). Now, let’s take a deeper look inside each subsystem and their configuration.

Vehicle subsystem

Figure 3. Vehicle subsystem

Figure 3 shows the implementation of a vehicle subsystem block. Vehicle subsystem includes a ‘vehicle body’ block, four ‘Tire’ blocks, & ‘simple gear’ bock. All these blocks are connected together in a way to model a rear-wheel-drive vehicle. Vehicle subsystem takes one input ‘Transmission input’ which is connected to the output shaft of the motor & outputs the vehicle speed in km/hr.

Vehicle Body:

This block represents a two-axle vehicle body in longitudinal motion. The block accounts for body mass, aerodynamic drag, road incline, and weight distribution between axles due to acceleration and road profile. A constant zero input is given as head-wind & grade to the block.

Under block parameters, mass, frontal area, & drag coefficient are configured with values 520, 1.1, & 0.28, respectively, keeping all other values as default.

Figure 4. Vehicle Body configuration.

Tire:

This block represents the longitudinal behaviour of a highway tire characterized by the tire Magic Formula. The block is built from Tire-Road Interaction (Magic Formula) and Simscape Foundation Library Wheel and Axle blocks. Optionally, the effects of tire inertia, stiffness, and damping can be included.

Under block parameters, the radius of tire is set to 0.281m & rolling resistance is configured ON for all the four tire blocks. All other parameters are kept as default.

Simple gear:

This block models a single-speed gearbox, which transfers power from the motor to wheels. An ideal mechanical rotational inertia block is also connected along with gearbox to account for the inertial load of the transmission gears. Meshing losses and viscous losses can also be modelled in Simple gear block.

Gear ratio must be calculated by considering the maximum motor torque & maximum motor speed, & making a compromise between two, such that desired wheel torque & vehicle speed is achieved.

For our purpose, the follower to base teeth ratio is configured to 9, and all other parameters are kept default.

Motor & Controller subsystem

Figure 5. Motor & Controller subsystem.

Figure 5, shows the implementation of the ‘Motor & Controller’ subsystem. The subsystem uses a DC motor to power the vehicle & an H-bridge controller. The subsystem takes acceleration and deceleration command as input. Output port ‘Motor shaft’ is the mechanical rotational conserving port associated with the DC motor rotor & output port ‘current’ gives the current through the DC motor. When brakes are applied, this same motor acts as a generator and recharges the battery.

DC motor:

This block represents the electrical and torque characteristics of a DC motor. Port connections are made as shown in figure 5. Under block parameters electrical torque tab, model parameterization is changed to ‘By rated load & speed’. No-load speed, rated speed, rated load, & rated DC supply voltage is configured with values 7000 rpm, 2100 rpm, 15 kW, & 72V respectively, and all other parameters are kept as default.

The rated power of the motor is calculated by considering the forces acting on the vehicle, i.e., rolling resistance, aerodynamic drag at top speed and the acceleration power required by the vehicle.

Mechanical Rotational reference is connected to the ‘C’ port of the DC motor block.

Figure 6. Configuration of DC motor.

A current sensor block is also connected in series between H-bridge & DC motor to measure the input and regenerative current.

H-bridge:

This block represents an H-bridge motor drive. It controls the power input to the motor according to the load requirements by controlling the input voltage to the motor.

H-bridge is configured to have its power supply internal. Simulation mode is configured as ‘Averaged’ to achieve fast simulation. Regenerative braking is also enabled.

Output voltage amplitude is set as equal to the rated voltage of the motor, i.e., 72 V. All other parameters are kept as default.

Figure 7. Configuration of H-bridge.

For braking command (BRK), a controlled voltage source block is used to produce voltage equal to the deceleration command from the driver.

Controlled PWM voltage:

This block produces PWM signal given as input to the H-bridge, according to the input voltage corresponding to the acceleration command from the driver.

A controlled voltage source block is used to produce voltage equal to the acceleration command between 0 & 1. This voltage is given as input to the ‘controlled PWM voltage’ block.

Under block parameters, simulation mode is set as ‘Averaged’ for fast simulation. Input voltage for 0% duty cycle & 100% duty cycle is set as 0 & 1, respectively. All other parameters are kept as default.

Figure 8. Configuration of Controlled PWM voltage.

Battery subsystem

Figure 9. Battery subsystem.

Battery subsystem models a battery for the EV. The implementation of the battery subsystem is shown in figure 9.

This subsystem takes current as input and gives the battery SOC % as output. A controlled current source block is used and connected with the battery as shown in figure 5. This block drains the battery with the same current drawn by the motor and also recharges during regenerative braking.

Battery:

This block is configured to model a high voltage traction battery.

Nominal voltage & internal resistance is set to 72V & 0.2 ohms, respectively. All other parameters are kept default.

Figure 10. Configuration of Battery.

SOC estimate:

This block estimates the battery SOC % using coulomb counting method. It takes current as the input and gives SOC% as output. Initial SOC of the battery is assumed to be 100%. And while calculating the SOC, battery capacity is assumed to be 100 Ah.

Figure 11. SOC estimation by coulomb counting.

Simulation:

A drive cycle source is given as input to the longitudinal driver block. The longitudinal driver also takes vehicle speed as feedback and gives out the acceleration or deceleration command to match the drive cycle source speed. A constant zero input is given as grade to the driver block.

Both the drive cycle source speed & vehicle speed is plotted using a scope block. A display block is used to display the distance covered by the vehicle. To calculate the distance travelled, speed in km/sec is integrated over time using an integrator block. Another scope is also used to plot the battery SOC %.

Case 1: FTP75 drive cycle

The FTP75 drive cycle is given as speed reference to the driver. The FTP cycle (for Federal Test Procedure) has been created by US EPA (Environmental Protection Agency) to represent a commuting cycle with a part of urban driving including frequent stops and a part of highway driving. This cycle covers a distance of 17.77 Km in 2474 seconds.

Simulation time is set as 2474 seconds.

Results:

Speed:

Figure 12. Speed plot (Case 1).

Figure 12 shows the speed plot for actual vehicle speed and the FTP75 drive cycle. We can observe that the vehicle achieves a top speed of 72.78 km/h & follows the reference speed quite well during the whole cycle. Some deviations are there because of the instant change in reference speed which would require hard braking or high acceleration. And if we look closely, these deviations are mostly during braking. This means vehicle had enough acceleration to follow the reference speed but braking power was not enough. This is because we are not applying any mechanical brakes to stop/slow down the vehicle. The only braking power available is through regenerative braking. This shows that an electric vehicle can actually be driven by a single pedal for most of the time. And therefore requires less frequent replacement of brake pads than a conventional ICE vehicle.

The actual distance covered by the vehicle is 17.59 Km which is very close to the total distance of the drive cycle.

SOC %:

Figure 13. Battery SOC % (Case 1).

After the simulation, battery SOC is reduced to 83.18 % from 100 % initial SOC. As the vehicle starts moving, battery SOC also starts to reduce and when the brake is applied we can see the SOC increasing as the battery is charged due to regenerative braking. And during the time vehicle is at rest, battery SOC also remains constant.

So, we can observe that vehicle uses about 17 % of the total battery capacity to travel a distance of 17.59 Km.

Case 2: Wide-open-throttle

Wide open throttle with following parameters is given as reference speed to the driver block to calculate 0-60 Km/hr time of vehicle.

Start time = 5s

Initial reference speed = 0 km/hr

Nominal reference speed = 60 Km/hr

Time to start deceleration = 30s

Final reference speed = 0 km/hr

WOT simulation time = 60s

Speed:

Figure 14. WOT response (case 2).

0 – 60 (km/hr) : 14.44 seconds

60 – 0 (km/hr) : 23.48 seconds

The braking time can easily be decreased by employing mechanical brakes.

SOC %:

Figure 15. Battery SOC % (Case 2).

From the SOC % plot, we can observe that from 5 sec to 30-sec battery is discharging and after 30 sec it starts charging due to regenerative braking.

Note: Please find the Simulink model file attached.