Developing an electric vehicle urban driving cycle to study differences in energy consumption

Area of Science:

• Automotive Engineering
• Transportation Science
• Energy Systems

Background:

• Accurate estimation of electric vehicle (EV) energy consumption and driving range is crucial for real-world applications.
• Existing international driving cycles may not accurately reflect diverse urban driving conditions, leading to estimation errors.

Purpose of the Study:

• To develop a methodology for constructing a representative urban driving cycle for electric vehicles (EVs) in Xi'an.
• To analyze the differences in energy consumption and driving range estimates using the developed cycle versus international standards.

Main Methods:

• Designing a test route based on Xi'an's urban topology and traffic flow data.
• Utilizing wavelet decomposition/reconstruction for data preprocessing.
• Applying Principal Component Analysis (PCA) for kinetic parameter reduction.
• Employing Fuzzy C-Means (FCM) clustering for driving segment classification.
• Constructing a representative EV urban driving cycle (XA-EV-UDC) based on segment proportions and correlation coefficients.

Main Results:

• The constructed Xi'an EV urban driving cycle (XA-EV-UDC) was developed.
• Estimates using international driving cycles showed significant errors: 9.65–21.17% for energy consumption and 20.10–38.14% for driving range.
• The study highlights the inadequacy of generic international cycles for specific urban environments.

Conclusions:

• A novel methodology for creating representative urban EV driving cycles has been established.
• Using region-specific driving cycles is essential for accurate EV energy consumption and driving range predictions.
• The findings underscore the need for localized driving cycle development to improve EV performance assessments.