Energy Consumption Model of Battery Electric Buses Using Real-World Data and Machine Learning

The energy consumption (EC) of the battery-electric buses (BEBs) is the backbone of the e-buses research, including optimization, battery capacity/performance, charging station's spatial distribution, hybrid vehicles, transit emissions and feasibility, and transit systems robustness studies. These studies depend on the accurate estimation of the EC rates within the BEB system, which cannot consider the bus as a stand-alone element, but focusing on the entire BEB system due to the interactions between the bus and its environment(He et al., 2018; Kivekas et al., 2017; Liu et al., 2017; X. Qi et al., 2018; Teoh et al., 2018; Wang et al., 2017). The EC rates differ significantly based on various parameters that can be classified into; 1) bus parameters such as mass, frontal area, motor power, drag coefficient, and rolling resistance coefficient; 2) battery parameters such as battery capacity, battery temperature, and state of charge; 3) operational parameters such as traffic condition, travel time, passenger loading, average speed; 4) route parameter such as grade, route length, speed limit, and spacing between stops; 5) environmental parameters such as weather, ambient temperature, air density, wind speed, and road condition; and 6) auxiliary parameters such as HVAC power, auxiliary power, and regenerative brake energy (Gallet et al., 2018; Kivekäs et al., 2018; Lajunen, 2018; Vepsäläinen, Ritari, et al., 2018). Through the literature, three main approaches have been used to predict the EC rates based on the aforementioned parameters: simulation, experimental (real-world), and equation. First, the simulation-based approach aims to mimic the performance of the BEBs in the real-world. Toward that end, it estimates accurate values of the EC using different MATLAB software's applications such as; Simulink that uses complex models graphically using block diagrams that are used to model, simulate, and analyze the entire powertrain to predict the EC(Hahn et al., 2019; Rupp et al., 2019; Vepsäläinen, Kivekäs, et al., 2018).Autonomie, in comparison, assumes different scenarios to analyze the automotive performance of the BEB component and predicting the EC(Gao et al., 2017). Advanced Vehicle Simulator (ADVISOR) is a system analysis tool that provides various models for predicting the EC, bus performance, and emissions. It also analyzes the performance of the vehicle's components such as fuel cells, electric motors, hybrid engines, and batteries (Lajunen, 2014).

• Authors:
  • Abdelaty, Hatem
  • Al-Obaidi, Abdullah
  • Mohamed, Moataz
  • 0000-0002-1345-7240
  • Farag, Hany E Z
• Conference:
  • CTRF 56th Annual Conference - Ensuring Resilience in Transportation Systems: Anticipating and Responding to Pandemic, Climate, Demographic and Economic Changes
• Publication Date: 2021-5

Language

• English

Media Info

• Media Type: Digital/other
• Features: Figures; References; Tables;
• Pagination: 5p

Subject/Index Terms

• ITRD Terms: 1272: Bus; 8018: Canada; 8525: Conference; 1282: Electric vehicle; 213: Energy; 132: Forecast; 9103: Simulation; 8645: Software
• Subject Areas: I91: Vehicle Design and Safety;

Filing Info

• Accession Number: 01888506
• Record Type: Publication
• Source Agency: Transportation Association of Canada (TAC)
• Files: ITRD, TAC
• Created Date: Jul 21 2023 5:21PM