Optimization and prediction of battery electric vehicle driving range using adaptive fuzzy technique

Despite the realization of zero-emission design of Battery Electric Vehicles (BEVs) for transportation to replace the traditional fuel-based vehicles, the real challenge is the minimal driving range. The conflict between battery power supply and consumption for motor and auxiliaries such as heati...

Full description

Saved in:
Bibliographic Details
Main Author: Abulifa, Abdulhadi Abdulsalam
Format: Thesis
Language:English
Published: 2022
Subjects:
Online Access:http://psasir.upm.edu.my/id/eprint/104050/1/FK%202022%2094%20IR.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Despite the realization of zero-emission design of Battery Electric Vehicles (BEVs) for transportation to replace the traditional fuel-based vehicles, the real challenge is the minimal driving range. The conflict between battery power supply and consumption for motor and auxiliaries such as heating, ventilation and air conditioning system (HVAC) implies the need to increase the battery energy. Since BEV battery storage is the sole source of energy, the travel distance for one charge significantly affects the driver’s confidence. Hence the goal of increasing the travel distance in BEV, and maintaining the driver’s confidence will be achieved by Energy Management System (EMS). EMS is an effective method to boost power efficiency by reallocating the electrical power flow inside HVAC system to obtain optimum effectiveness. This research aims to propose solutions to achieve a more efficient power control of the HVAC consumption by using adaptive fuzzy technique for optimization and prediction of EMS in BEV. The model was based on the actual parameters using software MATLAB-Simulink and ADVISOR. The vehicle was configured according to backward facing model and the design incorporated the technical specifications of a Malaysia local car, PROTON IRIZ (BEV). An optimal solution was proposed by integrating fuzzy logic technique with brute force algorithm that gave the best system optimization, where the decision was based on the Satisfaction Ratio (SR) and State of Charge (SoC). The study also developed an algorithm for predictive EMS using fuzzy model predictive control technique based on regression algorithm. The available parameters of speed, SoC, and power consumption had no pattern and the sample diversity was limited, therefore using simple regression was accommodating. The study also proposed a new measure named Wise Performance Measure (WPM) to achieve the balance between SoC and auxiliaries needs of energy by setting the threshold level for SoC drop and SR, then counting any breaks of this threshold every time intervals. For benchmarking, the results were compared with different designs based on previous studies using New European Driving Cycle (NEDC), Urban Dynamometer Driving Schedule (UDDS), Japanese 10-15 Mode Driving Cycle (Japan 10-15), and Highway Fuel Economy Test Driving Cycle (HWFET). The results showed that the basic fuzzy EMS was able to improve the power consumption by 11.7% to 12.4%; the optimized fuzzy EMS had improved by 18.8% to 26.6%, and the predictive fuzzy EMS had improved by 16.9% to 29.1%. Predictive EMS improved the city driving experience by 0.8% to 2.5% but was unable to improve the non-city driving experience. The proposed solutions show a significant improvement in the driving range. It is clear that implementing the fuzzy logic strategy, optimization, and predictive can iimprove the power consumption of HVAC system and retain the power capacity for motor torque and speed. The work in this thesis is expected to be the best approach in formulating an adaptive fuzzy technique based on brute force and regression algorithms for optimization and prediction of EMS in BEV application.