Experimental Calibration and Validation of an Average-Speed Fuel Consumption Model Based on Synthetic Driving-Cycles

Vehicle consumption models are a key element in the development of sustainable traffic monitoring and signal control systems. Leveraging innovative Intelligent Traffic Systems (ITS) technologies, we develop and validate an average-speed fuel consumption model for various vehicle types, including a 2020 Mercedes-Benz GLE 350, which is also used for experimental validation. Toward this aim, we first show and identify an instantaneous-speed fuel consumption model as an intermediate step. We then develop a unique method for generating synthetic driving-cycles by merging existing driving-cycles in the frequency domain and randomizing their features. In this way, a comprehensive database of synthetic driving-cycles, capturing diverse conditions and scenarios, is generated, enabling a robust foundation for fuel consumption analysis and comparison. Combining these results with the instantaneous-speed model, an average driving-cycle speed to fuel consumption model can be consequently calibrated. By applying this approach to different vehicle classes, the developed models facilitate accurate fuel consumption predictions, and support the optimization of traffic signal control strategies to reduce emissions and enhance energy efficiency.