Final Project: Design of an Electric Vehicle powertrain using MATLAB Simulink

AIM : To create a MATLAB model of electric car which uses a DC motor by choosing suitable blocks from Powertrain block set.

OBJECTIVES : To prepare a technical report about the EV model including -

1. System level configurations
2. Model parameters
3. Results
4. Conclusion

THEORY :

An Electric Vehicle (EV) is a vehicle that uses one or more electric motors or traction motors for propulsion. An electric vehicle may be powered through a collector system by electricity from off-vehicle sources, or may be self-contained with a battery, solar panels, fuel cells or an electric generator to convert fuel to electricity. EVs include, but are not limited to, road and rail vehicles, surface and underwater vessels, electric aircraft and electric spacecraft.
    

Components of an EV - 

• EVs have no need for the fuel engine and gear transmission, the two of the most crucial components for internal combustion vehicles.
• An electric powered car has three primary components - Battery, motor controller, and the electric motor.

(1) Battery : The battery of an electric car can be charged through the use of ordinary grid electricity at a specialized power station. But aside from the conventional lithium-ion battery technologies, there are also other major battery technologies which can be used for electric cars.

• Lithium-Ion Batteries : This battery technology gives extra performance and range. However, it also carries the highest price tag. Lithium-ion batteries are lighter than Lead acid and Nickel metal. These are also the batteries used in digital cameras and smartphones.
• Lead Acid Batteries : This battery technology is the most popular. It is also the cheapest among the battery technologies. What’s good about it is it’s 97% recyclable.
• Nickel Metal Hydride Batteries : This battery technology provides higher output and better performance but it costs much more than lead-acid batteries.

(2) Motor Controller : The motor controller of an electric car administers its complete operation and the distribution of its power at any given moment. It acts as a floodgate between the motor and batteries. It helps monitor and regulates all key performance indications such as the vehicle’s operator, motor, battery, and accelerator pedal. It has a microprocessor which can limit or redirect current. It is used to either improve the mechanical performance of the car or suit the operator’s driving style. There are also more refined controllers which are capable of greater accuracy and thus, higher efficiency.

(3) Electric Motor : Unlike a gasoline engine with lots of moving parts, a motor only has one moving part. This makes it a very reliable source of motive power. Choosing an electric motor depends on the car’s system voltage. They can be structured to use either AC or DC current. AC motors are less expensive and lighter compared to DC motors. They are also more common and they tend to suffer from less mechanical wear and tear. However, AC technology requires a more refined or sophisticated motor controller.

PROCEDURE :

(1) Vehicle Body Subsystem :

• The vehicle is assumed to be a front axle driven (4 wheels - 2 on each axle).
• The wheels are modelled using simple Tire (Magic formula) simulink blocks.
  
• The N port of the tire corresponds to the Normal Reaction acting on it.
  The A port corresponds to the Mechanical Rotational conserving port for the wheel axle. Thus both the wheels on the same axle should form a connection through their respective A ports.
  Connection H corresponds to the Mechanical Translational conserving port for the wheel hub through which the thrust developed by the tire is applied to the vehicle. Thus, all the 4 wheels form a connection through their respective H ports and this is in turn connected to the translational hub of the vehicle body.
  The connection S represents the output port for slip of the tire.
  
  
  
  
• The tires are parameterized by Peak longitudinal force and corresponding slip.
• The other block parameters such as - rated vertical load, peak longitudinal force at rated load and slip at peak force at rated load - are kept with their default values.
• The tire radius is given as 0.3 m.
• The tire inertia is kept as 1 kg-m^2. Also, the rolling resistance is given with a constant coefficient of 0.015.
  
            
  
  
• This 2 axle (4 wheel) assebly is now connected to a Velocity Body simscape block.
      
• This block basically represents a two-axle vehicle body in longitudinal motion. The block accounts for - 
  body mass,
  aerodynamic drag,
  road incline,
  weight distribution between axles due to acceleration
  road profile.
• Here, the connection H is the mechanical translational conserving port hub.
  NF & NR correspond to the output ports for nomral reaction forces on front axle and rear axle wheels respectively.
  Connection V  represents the actual output translational velocity of the vehicle.
  beta is the road inclination angle & W corresponds to the headwind speed (headwind - direction opposite to that of vehicle).
• The gross weight is given to be 1200 kg.
• The geometric parameters of the vehicle are given as -
  
  
  
  
• To account for proper drag force on the vehicle, the related parameters are kept as -
  
   
• The pitch dynamics for the vehicle are not considered for this simulation.
  
  
• The complete subsystem is shown below -
  
• The input for this subsystem is taken as the Rotational speed of the motor. This input is fed to the front axle through a Simple Gear representing the Final Drive Ratio of 3.73
  Here, the rotational direction of the output shaft is kep the same as that of the input shaft. Also, no meshing losses are considered for the simulation.
  
  
  
  

(2) DC Motor :

• Simulink provides an inbuilt bodel block for DC motor which converts electrical input into mechanical rotational output.
  Note : Here, the blue colour corresponds to the electrical side of the motor and the green colour corresponds to the mechanical side.
• Here, the motor field type is kept as Permanent magnet and the model parameterization is done by rated load and speed.
  The armature inductance and the mechanical parameters are given their default values.
  All the other parameters are changed as shown below - 
  
  
  
  
  
  
• The connection R represents the Rotor (rotational output) while the connection C is for the Casing (stationary). Thus, the R port is connected to the input of the vehicle subsystem created earlier. The C port is connected to a mechanical rotational reference. 
• The + and - terminals of the motor are connected to the motor controller. If these terminals are directly connected to a battery, the required DC motor will run on the rated capacity of the battery. This will eventually result into no control over the DC motor and the vehicle will run at top speed throughout the simulation.
  Thus, a motor controller is very much needed.

(3) Motor Controller :

• H-Bridge circuit -
     
• This block represents an H-bridge motor drive. The block can be driven by the Controlled PWM Voltage block in PWM or Averaged mode.
• In PWM mode, the motor is powered if the PWM port voltage is above the enabled threshold voltage.
• In Averaged mode, the PWM port voltage divided by the PWM signal amplitude parameter defines the ratio of the on-time to the PWM period.
• Connection REF is for the reference input. This combined with the PWM connected will form the pulse input for the H-Bridge.
  Here, the REF input is connected to the Electrical Reference (ground).
• Connection REV corresponds to the reverse motion of the motor which essentially means, the backward motion of the vehicle.
  Here, the backward motion of the vehicle is not accounted for and so, the port is connected to the electrical reference.
• Connection BRK is for the braking of the vehicle.
• Since, the PWM mode results in a huge amount of simulation time, Averaged mode is selected for this simulation.
• Load current characteristics are considered to be smoothed here. 
• Also, the regenerative braking is anabled for the simulation. This means, when the vehicle starts decelerating, corresponding amount of charge will be fed back to the battery.
  Accordingly, the BRK input is connected to a Controlled Voltage Source. This will generate emf corresponding to the intensity of brakes applied and fed the power back to the battery.
  
  
  
  
• The Input Threshold parameters are left with their default values.
  
  
  
  
• The Output Voltage Amplitude of the H-Bridge circuit is given the same value as the rated DC supply voltage of the DC motor.
  
  
  
  
• Simulink provides an inbuilt Controlled PWM Voltage block. This block is used for providing the proper pulse inputs to the H-Brdige circuit.  
  The PWM & REF connections are for the corresponding ports on the H-Bridge circuit.
  The 2 reference inputs correspond to the throttle inputs given by the driver. The block generates corresponding pulse width as per the accelerations and brakes apllied by the driver himself.
• Here, the PWM frequency is 1000 Hz. Also, the simulation mode is Averaged as that in the H-Bridge circuit.
  
  
  
  
• The input scaling parameters given are -
  0V for 0% duty cycle and 5V for 100% duty cycle. Accordingly, the output voltage amplitude becomes 5V.
  
  
• The H-Bridge circuit along with the PWM input, together form the motor controller circuit.
  

(4) Longitudinal Driver :

• Longitudinal Driver is an inbuilt block provided by the powertrain blockset.
  
• It is a parametric longitudinal speed tracking controller for generating normalized Acceleration and Braking commands based on reference and feedback velocities.
• The VelRef port is the reference velocity port where the input drive cycle data is fed.
  The VelFdbk port corresponds to the feedback velocity. The actual velocity output given by the vehicle body is connected here. By comparing the actual (feedback) velocity with the reference velocity, the driver block generates acceleration and braking signals in order to minimise the error between the two concerned velocities.
  Grade corresponds to the grade angle. For this simulation, no inclination is considered and hence, a constant block with value 0 is connected.
  Info gives the output for the bus signal for different block calculations like difference in reference vehicle speed and vehicle speed, etc.
   AccelCmd & DecelCmd correspond to the acceleration and deceleration commands generated respectively and are connected to the corresponding ports of the Controlled PWM Voltage block.
• Here, the selected control type is PI. Acordingly, the block implements proportional-integral (PI) control with tracking windup and feed-forward gains.
  
    

(5) Battery Pack :

• To model the Battery pak, a simple Battery block is used. This is because, if the generic battery block is used instead, the simulation will run for a much longer time.
• Here, the selected option for Battery Charge Capacity is Finite. Accordingly, the block models the battery as a series internal resistance plus a charge-dependent voltage source.
  If the chosen option had been Infinite, the voltage source would have been constant.
• Also, the self-discharging of the battery and the battery dynaics are disabled.
• The Battery Nominal Voltage is given as 400V.
  The Ampere-hour rating is assumed to be 50.
  The rest all paramaters are left with their default values.
  
  
  
  
• To ensure the battery supplies current as per the requirements of the motor controller and not the rated current, a Current Sensor & Controlled Current Source pair is connected. 
  

(6) Reference Velocity (Drive Cycle) :

• The Drive Cycle Source block is used to provide the reference speed for the simulation.
• It generates a standard or user-specified longitudinal drive cycle.
  
• Here, the selected drive cycle is the standard FTP75. 
  
  
• The plot of the drive cycle is shown below -
  
  

(7) Battery State of Charge :

• A small subsystem is created to calculate the State of Charge of the battery.
• This subsystem takes the battery current as the input.
  
  
• The essential process to calculate the SOC is by subtracting the charge from the rated capacity of the battery.
• Accordingly, the current is integrated  w.r.t. time to get the charge in coulumbs. Now this coulumb value is modified into Ampere hours. 
  This value is now subtracted from the charging capacity of the battery i.e. 50 Ah.
• Finally, the remaining charge is now converted into percentage value of the charging capacity of the battery.

Model :

    

(The simulink model is attached with the report for reference)

Simulation :

• The system is ran for simulation for 2474 s which is the total time for which the input drive cycle is defined.
• A Solver Configuration is added to ensure proper solving of the mathematical equations by the model.
  
• Scopes are connected to analyze different outputs such as the SOC of the battery, the reference velocity & the actual velocity of the vehicle and the distance covered by the vehicle during the total run.
• The 2 velocities are first converted from m/s to kmph values. For this, simple Gain blocks are used.
• The distance is calculated by simply time-integration of the vehicle velocity. 

Outputs :

(1) Ref. Speed v/s Actual vehicle speed :

• As can be seen from the above picture, the actual velocity of the vehicle differs from the reference drive cycle velocity.
• This is because, the chosen battery capacity and motors are not able to propel the speed upto the required extent.
• If a battery of higher capacity is chosen, the actual velocity will tend to match with the reference velocity.
• Also, it may be noted that in certain regions, despite the drive cycle showing small decelerations, the vehicle actually is accelerating with decreasing magnitudes. Reason being the same.
  
  

Distance Covered :

• It may be noted that, since, the reference velocity of the drive cycle never becomes negative (i.e. there is no backward motion of the vehicle), the distance is increasing throughout.
• The total distance covered by the vehicle is about 3.8 km.
  
  

Battery State of Charge :

• The State-of-Charge of the battery at the end of simulation run is nearly about 77 %.
• This means that the battery is not completely drained at the end and the vehicle can run on the same drive cycle for a few more times before the battery SOC becomes 0.
• Also, the thing to be noted here is, since we had enabled the regenerative braking for the simulation, the battery SOC increases slightly whenever the vehicle undergoes heavy deceleration.
• Again, this depends upon the generating tendency of the motor and the maximum amount of current that the battery can take.

CONCLUSION : The model was succesfully ran for simulation under the standard FTP75 drive cycle. The results have been properly analyzed.

Also, learnt quite a lot of interesting things like -

1. how to model an EV and carry out a real time simulation
2. how a motor controller works and what role it plays in the control of the electric motor
3. how to calculate the SOC of the battery
4. different powertrain blocks provided by simulink which can be used to simulate a vehicle model.