MODELLING AND CONTROL OF AN
AFTERMARKET PARALLEL HYBRID
ELECTRIC VEHICLE
AUTHOR: WISDOM ENANG
RESEARCH
BACKGROUND
INFORMATION
DEFINITION OF HYBRID ELECTRIC VEHICLE
Hybrid electric vehicles (HEV) combine the internal combustion engine of
normal vehicles with battery and electric motor.
HEV ADVANTAGES
Greater operating efficiency because HEVs use regenerative braking, which
helps to minimize energy loss and recover the energy used to slow down or
stop a vehicle.
Greater fuel efficiency because hybrids consume significantly less fuel than
vehicles powered by ICE alone
Cleaner operation because HEVs can run on alternative fuels: electricity
(which have lower emissions), thereby decreasing the dependency on
fossil fuels.
HEV CONFIGURATION USED IN THIS RESEARCH
UNIQUENESS OF THIS MODEL
AFTERMARKET
PARALLEL
HYRBRID
ELECTRIC VEHICLE
Hybrid electric system added to conventional diesel engine vehicle.
Hybrid system only controls electric motor thus preserving the original
vehicle warranty.
Hybridization kits (electric motor and electric battery) are small in size and
affordable.
POWER FLOW POSSIBLE IN THE AFTERMARKET PARALLEL HEV
Motor only mode Power assist mode
Engine only mode
Recharge control mode Regenerative braking mode
RESEARCH
QUESTIONS
RESEARCH
QUESTIONS
RESEARCH AIM
To produce a robust real time controller for a parallel aftermarket HEV.
RESEARCH OBJECTIVES
To produce a parallel HEV model capable of accurately predicting fuel
consumption in real world driving scenarios.
To identify the interactions between human driver behaviour and fuel
consumption using the validated HEV model
Computation of a rule based control for the HEV
Optimal control of the HEV using Dynamic programming
Intelligent control of the HEV using GPS information
Intelligent control of the HEV using information from on-board driving
pattern learning algorithm taking in to consideration driver behaviour.
WHAT DATA IS NEEDED AND HOW IT CAN BE COLLECTED?
Chassis Dynamometer
Electric motor test rig
Engine fuel consumption map at each torque and speed operating point.
Engine transient testing of real world drive cycle for model validation.
Motor efficiency map at each torque and speed operating point.
WHAT HEV CONTROL OPTIONS ARE THERE?
WHAT HEV MODELLING OPTIONS ARE THERE?
This approach makes the assumption that the vehicle
meets the target performance, so that the vehicle speed is
supposed known a priori; thus enjoying the advantage
simplicity and low computational cost. Backward or Kinematic Approach
This approach makes use of a driver model typically a PID which compares that target vehicle speed (drive cycle
speed) with the actual speed profile, and then generates a power demand profile which is needed to follow the target
vehicle speed profile by solving the differential motion
equation of the vehicle. Quasi Static Approach
RESEARCH
PROGRESS
OVERALL RESEARCH PROGRESS
Experimental testing of Engine for fuel consumption map
and model validation
Electric motor test for motor efficiency
map
Testing
HEV modelling
Modelling
HEV model validation
Rules based HEV control
Control
Optimal HEV control using dynamic control
Intelligent HEV control using GPS
Intelligent HEV control using driver style
learning algorithm
Real time implementation of
HEV controllers
RESEARCH
RESULTS
RESULTS FROM EXPERIMENTAL TESTING
Motor testing results Engine testing results
HEV MODELLING STRUCTURE – QUASI STATIC APPROACH
Use orange switch inside to include or exclude
Engine idling when cycle speed demand is 0
Note: Time delay factor added
to the Hybrid
Controller to make the system results
more useful in real l ife
wheel torque (Nm)
shif t f lag (-)
speed_signal
Motor Torque (Nm)
Current_mode
v ehicle v elocity (m/s)
Wheel Tractiv e Force - Engine (N)
Wheel Braking f orce (N)
Vehicle Dynamics
Terminator
Scope3
Scope2
Scope1
Scope
Motor_power
Power_demand
Engine_power
Plots
Engine Power
Plot Analysis
Manual Switch
Initialize Model Parameters
SOC
Hy brid Switch
Pdemand
motor_speed (RPM)
Engine_speed (RPM)
Current Mode
Motor Power
Hybrid Control System
[shift_flag]
Goto9
[Engine_torque]
Goto8
[engine_speed]
Goto7
[current_mode]
Goto6
[Engine_Power]
Goto5
[Motor_Torque]
Goto4
-T-
Goto3
[Motor_Power]
Goto20
[wheel_torque]
Goto2
[SOC]
Goto18
[Power_demand]
Goto15
[motor_speed]
Goto14
[speed_signal]
Goto13
[idle_flag]
Goto10
[gear_demand]
Goto1[vehicle_velocity]
Goto
Fuel Savings %
Fuel Consumption g
[Engine_Power]
From9
[Power_demand]
From8
[Motor_Torque]
From7
[Motor_Power]
From6
-T-
From5
[wheel_torque]
From4
[gear_demand]
From3
[current_mode]
From22
[Motor_Power]
From20
[vehicle_velocity]
From2
[engine_speed]
From19
[Engine_torque]
From18
[motor_speed]
From17
[Power_demand]
From16
[idle_flag]
[SOC]
From14
[speed_signal]
From13
[shift_flag]
From12
[shift_flag]
From11
[engine_speed]
From10
[vehicle_velocity]
From1
chassis_dyno_speed_dmd
cycle_gear_demand
cycle_speed_demand
[vehicle_velocity]
From
idle f lag (-)
Engine_torque (Nm)
Enginespeed (rpm)
Fuel Consumption (g)
Fuel Sav ings (%)
Engine
cy cle_gear_demand
cy cle_speed_demand (km/h)
v ehicle_v elocity (m/s)
gear_demand
wheel torque (Nm)
Shif t_f lag [-]
Speed Signal
Power demand (w)
Driver Subsystem
gear_demand
v ehicle v elocity (m/s)
Wheel tractiv e f orce - Engine (N)
Shif t f lag [-]
Engine Torque (Nm)
engine speed (RPM)
idle f lag [-]
Engine Power (KW)
Drive Train
0
Constant2
1
Constant
v ehicle v elocity (m/s)
Motor Power demand (W)
Motor Torque (Nm)
SOC (%)
Motor Speed (RPM)
Battery and Electric Motor Subsystem
Hy brid_Switch
This approach makes use of a driver model typically a PID which compares that target vehicle speed (drive cycle speed) with the actual speed profile, and then generates a power demand profile which is needed to follow the target vehicle speed profile by solving the differential motion equation of the vehicle.
HEV MODEL VALIDATION
Model validation carried out over the NEDC (New European Drive Cycle)
NEDC testing results proves it to be highly repeatable and hence why it has been chosen for the model validation
Level of accuracy achieved: 99% model accuracy
HIGHLIGHTS FROM MODEL VALIDATION
RULE BASED CONTROL STRUCTURE
Overview of the control structure Traction mode control structure
Braking mode control structure
RULE BASED CONTROL RESULTS
Drive cycle speed time profile Power split profile Instantaneous Fuel consumption
profile comparison
Engine operating point Battery state of Charge profile Cumulative fuel consumption
profile comparison
State of charge boundaries: Highest allowable (80%) and lowest allowable (20%)
Fuel savings achieved over the NEDC 12.58%
Lowest state of charge encountered 27%
Future Work
Optimal HEV control using
dynamic control
Intelligent HEV control using
GPS
Intelligent HEV control using driver
style learning algorithm
Real time implementation
of HEV controllers
PhD RESEARCH PROJECT GANTT CHART
QUESTIONS
PLEASE?
THANK YOU FOR
LISTENING