Development of a comprehensive energy model to simulate the energy efficiency of a battery electric vehicle to allow for prototype design optimisation and validation.

Abstract

This dissertation describes the development of an energy model of a battery electric vehicle (BEV) to assist designers in evaluating the impact of overall energy efficiency on vehicle performance. Energy efficiency is a crucial metric for BEVs as it defines the driving range of the vehicle and optimises the limited amount of energy available from the on-board battery pack, typically the most expensive component of the vehicle. Energy modelling also provides other useful information to the designer, such as the range of the vehicle according to legislative drive cycles and the maximum torque required from the motor. An accurate, fast and efficient model is therefore required to simulate BEVs in the early stages of design and for prototype validation. An extensive investigation into BEV modelling and the mechanisms of energy losses within BEVs was conducted. Existing literature was studied to characterise the effect of operating conditions on the efficiency of each mechanism, as well as investigating existing modelling techniques used to simulate each energy loss. A complete vehicle model was built by considering multiple domain modelling methods and the flow of energy between components in both mechanical and electrical domains. Simscape™, a MathWorks MATLAB™ tool, was used to build a physics based, forward facing model comprising a combination of custom coded blocks representing the flow of energy from the battery pack to the wheels. The acceleration and speed response of the vehicle was determined over a selected drive cycle, based on vehicle parameters. The model is applicable to normal driving conditions where the power of the motor does not exceed its continuous rating. The model relies on datasheet or non-proprietary parameters. These parameters can be changed depending on the architecture of the BEV and the exact components used, providing model flexibility. The primary model input is a drive cycle and the primary model output is range as well as the dynamic response of other metrics such as battery voltage and motor torque. The energy loss mechanisms are then assessed qualitatively and quantitatively to allow vehicle designers to determine effective strategies to increase the overall energy efficiency of the vehicle. The Mamba BEV, a small, high-power, commercially viable electric vehicle with a 21 kWh lithium-ion battery was simulated using the developed model. As the author was involved in the design and development of the vehicle, required vehicle parameters were easily obtained from manufacturers. The range of the vehicle was determined using the World-Harmonised Light Duty Vehicles Test Procedure and provided an estimated range of 285.3 km for the standard cycle and 420.8 km for the city cycle.