Research On The Fuel Economy Of Automatic Transmission Vehicles For Running On The Hilly Roads

Vehicle fuel economy is affected by various factors, including the vehicle-related factors, the roadway-related factors, driver-related factors and so on. The road gradient within the roadway-related factors has a great impact on both vehicle loading state and driver's driving behavior. Driving across the hilly road with aggressive acceleration, deceleration or unreasonable shifting will cause an increase in vehicle fuel consumption. Therefore, to improve the vehicle fuel economy, a fuel economy optimization problem of running on hilly roads for automatic transmission vehicles was defined and researched. Its main objects include the fuel consumption modelling and fuel economy optimizing for running on hilly roads.Fuel consumption modelling has emerged as an effective tool to help develop and assess advanced vehicle technologies and to help predict vehicle instantaneous fuel consumption. Based on the analysis of the main factors affecting the vehicle fuel consumption, transient fuel consumption models that have the form of “steady-state prediction + transient correction” were developed using data collected by the Argonne National Laboratory(ANL). These models were characterized by two sub-modules, the steady-state module whose inputs are engine torque and engine speed and the transient correction module whose inputs are vehicle speed and acceleration. According to the different implementation methods of each steady-state module and each transient correction module, two transient fuel consumption models were developed, namely the “2D-interpolation + transient difference correction” model BIT-TFCM-1 and the “Polynomial fitting + transient ratio correction” model BIT-TFCM-2. These two models were verified using ANL's testing data and their results showed that the prediction accuracy of fuel consumption models can be improved significantly by introducing appropriate transient correction measures based on the steady-state fuel consumption models. The Mean Absolute Percentage Error values of the BIT-TFCM-1 model and BIT-TFCM-2 model were approximately 29.43%-34.48% and 11.44%-19.28% lower than their corresponding steady-state module. Additionally, compared to the classical transient fuel consumption model VT-Micro, the new developed transient fuel consumption models have a higher prediction accuracy.An eco-driving algorithm that can utilize the changing road gradient information was developed for automatic transmission vehicles based on the Bellman Dynamic Programming(DP). The goal of the eco-driving algorithm was to optimize vehicle fuel economy through properly planning vehicle speed trajectory while driving across the hilly roads. The applicability and effectiveness of the new developed eco-driving algorithm were verified through the co-simulation of MATLAB/Simulink and CarSim. The validation results showed that the vehicle controlled by the new eco-driving algorithm will increase its velocity before entering an uphill so that high acceleration can be avoided during going up; the vehicle will decrease its velocity before entering a downslope so that it can take full advantage of downslope to accelerate. Moreover, compared to the traditional cruise control, the energy saving potential for the new eco-driving algorithm ranged from 5.59% to 40.14% depending on the different application scenarios.