Validation of a statistical-dynamic framework for predicting energy consumption: A study on vehicle energy conservation equation
Bin Sun,Qijun Zhang,Hongjun Mao,Zhijun Li
DOI: https://doi.org/10.1016/j.enconman.2024.118330
IF: 10.4
2024-05-01
Energy Conversion and Management
Abstract:Predictive models for vehicle energy consumption are crucial for sustainable development in urban road traffic systems. This paper comprehensively reviews classic predictive models and develops a novel statistical-dynamical energy consumption prediction framework called Vehicle Energy Conservation Equation (VECE). VECE is constructed based on the principles of vehicle energy flow and regression analysis, employing a continuous and concise mathematical formulation. Its coefficients possess clear physical interpretations, allowing for application in various vehicle categories. To validate VECE, this study collected energy consumption data from 28 vehicles, including 9 diesel vehicles, 16 gasoline vehicles, 1 ethanol gasoline vehicle, and 2 battery electric vehicles. A rigorous data processing procedure was designed. Data analysis revealed that the VECE coefficients are correlated with vehicle type, speed, and acceleration. VECE's predictive performance is minimally impacted by the number of vehicle categories, effectively modeling energy consumption for vehicles of the same fuel type or size category. Comparative analysis with E-EcoGest, VT-Micro, PERE, and CMEM demonstrates the moderate accuracy of VECE in predicting instantaneous vehicle energy consumption while excelling in predicting cumulative energy consumption. The minimum relative percentage error for the cumulative predicted values is 4.2%. Overall, VECE demonstrates outstanding performance in computational simplicity, coefficient interpretability, adaptability, and extensibility, making it a crucial tool for achieving energy-efficient road transport systems.
energy & fuels,thermodynamics,mechanics