Search Results

The increasing numbers of hybrid electric and full electric vehicle models currently in the market or in the pipeline of automotive OEMs require creative testing mechanisms to drive down development costs and optimize the efficiency of these vehicles. In this paper, such a testing mechanism that has been successfully implemented at the US Environmental Protection Agency National Vehicle and Fuel Emissions Laboratory (EPA NVFEL) is described. In this testing scheme, the units-under-test consist of a battery pack and its associated battery management system (BMS). The remaining subsystems, components, and environment of the vehicle are virtual and modeled in high fidelity.

Engine dynamometer testing was performed comparing fuels having different octane ratings and ethanol content in a Ford 3.5L direct injection turbocharged (EcoBoost) engine at three compression ratios (CRs). The fuels included midlevel ethanol “splash blend” and “octane-matched blend” fuels, E10-98RON (U.S. premium), and E85-108RON. For the splash blends, denatured ethanol was added to E10-91RON, which resulted in E20-96RON and E30-101 RON. For the octane-matched blends, gasoline blendstocks were formulated to maintain constant RON and MON for E10, E20, and E30. The match blend E20-91RON and E30-91RON showed no knock benefit compared to the baseline E10-91RON fuel. However, the splash blend E20-96RON and E10-98RON enabled 11.9:1 CR with similar knock performance to E10-91RON at 10:1 CR. The splash blend E30-101RON enabled 13:1 CR with better knock performance than E10-91RON at 10:1 CR. As expected, E85-108RON exhibited dramatically better knock performance than E30-101RON.

In order to improve the accuracy of vehicle simulation under transient cycle conditions and thus predict performance and fuel consumption, consideration of the complete system engine/drivetrain/vehicle is necessary. The coupling of otherwise independent simulation programs is therefore necessary for the vehicle and engine. The description of thermally transient processes enables the calculation of the heat balance of the engine, which in turn enables the simulation of warming up operation. Through consideration of the engine warming up process, the quality of the prediction of fuel consumption and emissions is improved. The combination of the simulation programs CRUISE and BOOST to determine the engine heat balance has proven to be successful for the analysis of transient drive cycles.

The cost of fuel for commercial trucks is second only to labor in the total vehicle operating costs. Therefore, technologies that reduce fuel consumption can have a significant impact on the bottom line for both trucking fleets and owner/operators. Quantifying the fuel savings associated with different technologies, however, is complicated by many factors, and short-term testing often cannot adequately quantify small changes in fuel consumption that, over time, can add up to substantial cost savings on a vehicle. For example, fuel economy gains of less than one percent may not be reliably measurable using fuel tests, and variable environmental and use factors can cast some doubt on the appropriateness of short-term testing.

Increasing powertrain complexity and the growing number of vehicle variants are putting a strain on current calibration development processes. This is particularly challenging for vehicle drivability calibration, which is traditionally completed late in the development cycle, only after mature vehicle hardware is available. Model-based calibration enables a shift in development tasks from the real world to the virtual world, allowing for increased system robustness while reducing development costs and time. A unique approach for drivability calibration was developed by incorporating drivability analysis software with online optimization software into a virtual engine test cell environment. Real-time, physics-based engine and vehicle simulation models were coupled with real engine controller hardware and software to execute automated drivability calibration within this environment.

This paper outlines the characterization of a Li-Ion Iron Phosphate battery pack with nominal voltage of 700V as well as the modeling of this pack as an equivalent electrical circuit (EEC) for the purpose of vehicle simulations. For a higher level of fidelity and accuracy, the equivalent circuit is initially modeled as an R-2RC circuit which consists of a voltage source with one resistor (R) and two resistor-capacitor (RC) branches. In this modeling effort, first, several open circuit voltage (OCV) determination methods in the literature are benchmarked and state-of-charge (SOC) dependent OCV curve which is used in the voltage source of the EEC model is derived. Then, two methods of parameter estimation of the EEC are developed for both step current and dynamic current profiles. The first estimation method is applicable to discharge or charge step currents and relies mostly on the relaxation portion of the battery response and involves some manual calibration.