1. Introduction
The development of the traditional automobile industry is currently limited by the shortage of resources [
1] and environmental pollution [
2]. Under the concept of sustainable development, new-energy vehicles have gradually gained public recognition. New-energy vehicles are mainly divided into three categories, namely, hybrid vehicles, pure electric vehicles, and fuel-cell vehicles [
3]. After years of research and development, pure electric buses have taken a dominant position in the field of passenger cars. With the advantages of no pollution, low noise, high energy efficiency, and convenient maintenance, they have been popularized in various cities and have gradually replaced traditional fuel buses. They have achieved great advantages in protecting the environment, improving the country’s energy structure, and reducing urban noise pollution [
4].
However, there are still some problems in the development of the power system of the pure electric bus. (1) The selection of the motor [
5]: For example, when the traffic is congested, the pure electric vehicle has a smaller loss than the fuel vehicle. However, due to the characteristics of the motor’s low-speed constant torque and high-speed constant power characteristics, when driving at high speed, the electric vehicle’s energy consumption growth is much larger than that of conventional vehicles [
6]. (2) The selection of power batteries [
7]: The selection of batteries directly affects the cruising range [
8] and safety of buses. At present, lithium iron phosphate batteries and ternary batteries occupy the power battery market for pure electric buses. Both have certain problems. The former has good cycle stability but large mass and low energy density. The latter has high energy density but poor safety. It is easy to spontaneously ignite when the temperature is high, and it is difficult to control [
9].
Many researchers have studied the power system of pure electric buses. Hu et al. [
10] developed the power distribution control strategy of the dual-motor coupling mode considering the dynamic characteristics of the motor by constructing a vehicle model including the motor dynamic control model and simulated it in the MATLAB platform. The results showed that the established dynamic model of the motor was closer to the actual operating conditions. The control strategy effectively reduced the power demand of the electric drive system, reduced the energy consumption of the electric drive system, and extended the vehicle mileage. Vora et al. [
11] proposed a novel framework that enables the parametric design optimization of hybrid electric vehicles, taking into account battery degradation and its impact on the vehicle life cycle. The framework captures the impact of battery degradation on fuel consumption and battery replacement by integrating a battery model capable of predicting degradation and performance degradation into drive cycle simulations. Raga et al. [
12] analyzed the different factors that affect the optimal design of a power system (including the minimum power provided by the fuel cell, the storage of recovered energy in the regenerative braking stage, the battery technology, and the change in the maximum charging state of the battery) and analyzed how these design factors affect the quality, volume, and cost of the optimal distribution architecture, as well as how to consider these factors in the design. Hoshing et al. [
13] compared the life-cycle cost of series and parallel architectures for PHEVs, saving fuel costs and enabling a faster return on investment for transit bus scenarios compared with medium-duty trucks, driving the early viability of transit bus applications. Soltani et al. [
14] studied the effectiveness of the hybrid energy storage system in protecting the battery from high power loss during charging and discharging, reducing the negative impact of power peaks related to urban driving cycles, and improving the battery life by 16%. Jin et al. [
15] evaluated state-of-the-art methods in Li-ion battery degradation models, including accuracy, computational complexity, and ease-of-control algorithm development. Based on a comparison of simulation results and experimental data, key differences in aging factors captured by each model were summarized. However, there is a problem in these research processes, that is, the fact that they only considered the influence of single factors, such as battery or motor, on the design of the power system, without considering whether the matching of battery, motor, and other components was the best, which may have led to the dynamic performance of the vehicle not being fully reflected. Therefore, the analysis of the selection and matching of different components plays an important role in the design of pure electric buses.
The orthogonal test is another design method to study multiple factors and levels. It selects some representative points from the comprehensive test according to orthogonality. These representative points have the characteristics of “uniform dispersion, neat and comparable”. Orthogonal test design is the main method in fractional factorial design. Xu et al. [
16] optimized the spin-forming parameters of bimetallic composite pipes using the finite element method and the experimental method and analyzed the torque and residual contact pressure in the forming process. Through the orthogonal test method to optimize the spinning process parameters of the bimetal composite pipe and through the deep drawing experiment of the composite pipe, the accuracy of the numerical simulation value was verified. Quan et al. [
17] used the orthogonal experimental design method combining experiments and numerical calculations to optimize the design structure of a vortex pump impeller. Through an orthogonal experimental design, the design cycle of the vortex pump could be effectively shortened; the design level of the vortex pump could be improved; and a hydraulic model with superior performance could be obtained. Yu et al. [
18] studied the effects of load, frequency, duration, and concentration on the friction properties of additives on an SRV reciprocating wear tester using an orthogonal experimental design. This paper was not only aimed at the single factor of battery or motor but also at the multi-factor and multi-level research of different batteries, motors, and other influencing factors. Therefore, the orthogonal test was very consistent with the design of the experimental scheme.
ADVISOR is a series of models, data, and script files in the MATLAB and Simulink software environment. It can quickly analyze the fuel savings, power, and emissions of traditional vehicles, pure electric vehicles, and hybrid electric vehicles using the parameters of various parts of the vehicle under the given road cycle conditions. Sun et al. [
19] used ADVISOR and iSIGHT software to jointly optimize the transmission ratio and used ADVISOR simulation software to analyze the savings and power of the optimized transmission ratio. The optimized pure electric passenger vehicle achieved a balance between power and economic performance, thus giving play to its performance advantages. Wang et al. [
20] established the energy model of a vehicle system using ADVISOR and MATLAB software. An energy control strategy for the dynamic wireless charging of electric vehicles was designed. The influence of dynamic wireless power transmission on energy storage system loss and electric vehicle range under different operating scenarios and different battery minimum charging states was studied. Sun et al. [
21] selected configuration parameters such as vehicle body, battery, tires, and transmission in ADVISOR for simulation and selected vehicle mass, rolling resistance coefficient, and auxiliary power as control variables to conduct a cost–benefit analysis based on the vehicle’s full life-cycle mileage, so as to reduce the short-term profit limit of optimization evaluation, which has practical guiding significance for the formulation of an energy optimization scheme for electric commercial vehicles. The biggest feature of ADVISOR software is that the simulation model and software source code are completely open all over the world and can be downloaded free of charge from the website. On the basis of the original simulation model, researchers can modify or reconstruct some simulation models and adjust or redesign the control strategy as needed to make it closer to the actual situation, and the simulation results are more reasonable. Through the secondary development of the battery model and motor model in ADVISOR, this paper made the simulation process more reasonable and more practical.
This paper combined MATLAB/Simulink and ADVISOR software to analyze and model the whole vehicle, battery, motor, etc., of a medium-sized bus [
22]. Around the needs of the actual vehicle model [
23], according to the orthogonal test table, multiple groups of combined parameters, such as different numbers of battery packs, motor power models, wheel rolling resistance coefficients, and wind resistance coefficients, were determined. The influence of different combined parameters on the power performance and economic performance of a pure electric medium-sized bus was simulated and analyzed [
24]. The optimization scheme of the power system of an electric bus was selected through comparison, and the feasibility of the optimization scheme was verified. It can provide a certain reference for the design of shuttle buses for Nantong’s public transport system and provides a new idea for the design of the power system of a pure electric bus.
Figure 1 presents a flow chart showing the structure of this paper.