Prediction of fuel efficiency and CO2 emission characteristics of Light Duty Vehicles (LDVs) based on vehicle dynamics modeling

Abstract

In this dissertation, thirteen types of LDV models (gasoline, diesel, LPG fuel type LDVs, HEVs) were generated based on vehicle dynamic based modeling method. The main structure of this method is to determine total traction force needed to follow a target speed at each time step. And then, demanded engine power and fuel consumption is calculated at each time step. In case of HEV model, fuzzy logic is added to control an engine, motor and battery pack module. Before using these LDVs models into various studies, prediction accuracies are need to be validated by using chassis dynamometer test data. To advance prediction accuracy, relationship between pedal sensor position and engine torque are investigated and implemented in gear shifting program. From this procedure, it was validated that predicted results such as FE, CO2 emission rate and engine operating conditions have strong correlation with test results. In case of ICEVs, most of relative error of FE and CO2 emission converged in 5 percent. Also, predicted engine operation points are well match with experimental data. In case of HEV model, relative error of predicted FE, CO2 emission rate also converged in acceptable level and SOC is well controlled in +/-3% range. After validating prediction accuracy of LDV models, four types of studies (section 3.5.2 to 3.5.5) were conducted as follows. In this investigation, the effects of changing vehicle specifications on fuel efficiency and CO2 emission specifically analyzed. From this study, it is revealed that 10% of GVW reduction is advantageous to reduce 2-7% of total fuel consumption and CO2 emission in low speed modes. The frontal area, air drag coefficient and rolling resistance also expected to be highly impact on FE and CO2 emission in high speed modes. (5-6%) Meanwhile, effect of changing full load characteristic curve is merely shown in results with except for small gasoline LDV. The CO2 correction method based on an energy deviation analysis was applied in various simulation cased to reduce a relative error can be occurred in same test conditions. Deviation of actual vehicle speed against target speed and inaccurate road load settings were selected as main factors and 189 cases are simulated to validate the effectiveness of CO2 correction method. From the results, it is revealed that total quantity of CO2 emission can be changed 2.8-5% even if LDVs operated in same conditions and driving mode because of acceptable speed error range (+/-3.2km/h) inaccurate road load setting. After applying CO2 correction method, 28.9-93.0% reduction of the relative error was observed regardless of driving mode and LDV type. Five types of middle LDVs (gasoline, diesel, gasoline-electric, diesel-electric and LPG) are simulated in certification driving modes and real driving routes. Compared to ICEVs, HEVs shows stable performances in all real driving routes. To conduct more detailed analysis, relationship between acceleration and CO2 emission rate was investigated. From the results, it is revealed that LPG fuel type LDVs and HEVs emit a 50-90% of CO2 emission compared to gasoline and diesel fuel type LDVs in all driving conditions. Based on power binning method, general CO2 emission rates of LDVs in real driving conditions were predicted. Total amount of CO2 emission generated in Korea LDV sector is predicted bases on 6 types of blue map scenarios. In scenario 1, 5.2% of CO2 emission reduction is expected with increasing portion of small vehicle up to 18%. In scenario 3 and 4 (assumed that HEVs occupying 15-30% of total penetration rate), 4.6-7.9% of CO2 emission reduction is expected. In scenarios 5 and 6, 10.5-14.5% of CO2 reduction was it expected by reducing 8-12% of total number of LDVs. From the scenario results, it is revealed that 5-15% of total CO2 reduction generated in LDV sector is enabled by enacting active policies.