Search Results

This work represents an advanced engineering research project partially funded by the U.S. Department of Energy (DOE). Ford Motor Company, FEV North America, and Oak Ridge National Laboratory collaborated to develop a next generation boosted spark ignited engine concept. The project goals, specified by the DOE, were 23% improved fuel economy and 15% reduced weight relative to a 2015 or newer light-duty vehicle. The fuel economy goal was achieved by designing an engine incorporating high geometric compression ratio, high dilution tolerance, low pumping work, and low friction. The increased tendency for knock with high compression ratio was addressed using early intake valve closing (EIVC), cooled exhaust gas recirculation (EGR), an active pre-chamber ignition system, and careful management of the fresh charge temperature.

The rise of greenhouse gas emissions has reached historic levels, with 37 billion tons of CO2 released into the atmosphere in 2018 alone. In the European Union, 32% of these emissions come from transportation, with 73.3% of that percentage coming from vehicles. To address this problem, solutions such as cleaner fuels and more efficient engines are necessary. Artificial Intelligence can also play a crucial role in climate analysis and verification to move towards a more sustainable future. By utilizing connected vehicle data, automakers can analyze real-time vehicle performance data to identify opportunities for improvement and reduce carbon emissions. This approach benefits the environment, improves vehicle quality, and reduces engineering work time, making it a win-win solution. Connected vehicle data offers a wealth of information on vehicle performance, such as fuel consumption and carbon emissions.

This is an extension of simple fuel consumption modeling toward HEV. Previous work showed that in urban driving the overhead of running an ICEV engine can use as much fuel as the traction work. The bidirectional character and high efficiency of electric motors enables HEVs to run as a BEV at negative and low traction powers, with no net input from the small battery. The ICE provides the net work at higher traction powers where it is most efficient. Whereas the network reduction is the total negative work times the system round-trip efficiency, the reduction in engine running time requires knowledge of the distribution of traction power levels. The traction power histogram, and the work histogram derived from it, provide the required drive cycle description. The traction power is normalized by vehicle mass, so that the drive trace component becomes invariant, and the road load component nearly invariant to vehicle mass.

Several factors stimulate the development of new materials in the industry. From specific physical-chemical characteristics to strategic market advantages, technology companies seek to diversify their raw materials. In the automotive sector, the current trend of electrification in vehicles and the increase of government and market demand for reducing the emission of greenhouse gases makes lighter materials more and more necessary. As electric vehicles use heavy batteries, the vehicle weight is directly related to its power demand and level of autonomy. The same applies to internal combustion vehicles where the vehicle weight directly impacts fuel consumption and emissions. In this context, there is a lot of research on special alloys and composites to replace traditional materials. Aluminum is a good alternative to steel due to its density which is almost five times smaller while that material still has good mechanical properties and has better impact absorption capability.

This study investigated the potential of generating reactive chemical species (including radicals) through pilot fuel injection in negative valve overlap for improving the combustion and emissions performances of spark ignition gasoline engines under low load and low speed operating conditions. Several Ford sub-models were used for simulating the physics and chemistry processes of injecting a small amount of fuel in NVO (negative valve overlap). Effects of different NVO degrees and different pilot injection timings, factors for fuel conversion were simulated and investigated. CO and H2 conversions during NVO, CO and H2 amounts before spark timing were used for comparing different schemes.

The previously developed power-based fuel consumption theory for Internal Combustion Engine Vehicles (ICEV) is extended to Battery Electric Vehicles (BEV). The main difference between the BEV model structure and the ICEV is the bi-directional character of traction motors and batteries. A traction motor model was developed as a bi-linear function of positive and negative traction power. Another difference is that the accessories and cabin heating are powered directly from the battery, and not from the powertrain. The resulting unified model for ICEV and BEV energy consumption has linear terms proportional to positive and negative traction power, accessory power, and overhead, in varying proportions. Compared to the ICEV, the BEV powertrain has a high marginal efficiency and low overhead. As a result, BEV energy consumption data under a wide range of driving conditions are mainly proportional to net traction power, with only a small offset.

Historical driver behavior and drive style are crucial inputs in addition to V2X connectivity data to predict future events as well as fuel consumption of the vehicle on a trip. A trip is a combination of different maneuvers a driver executes to navigate a route and interact with his/her environment including traffic, geography, topography, and weather. This study leverages big data analytics on real-world customer driving data to develop analytical modeling methodologies and algorithms to extract maneuver-based driving characteristics and generate a corresponding maneuver distribution. The distributions are further segmented by additional categories such as customer group and type of vehicle. These maneuver distributions are used to build an aggressivity distribution database which will serve as the parameter basis for further analysis with traffic simulation models.

Fuel power consumption for light duty vehicles has previously been shown to be proportional to vehicle traction power, with an offset for overhead and accessory losses. This allows the fuel consumption for an individual powertrain to be projected across different vehicles, missions, and drive cycles. This work applies the power-based model to commercial vehicles and demonstrates its usefulness for projecting fuel consumption on both regulatory and customer use cycles. The ability to project fuel consumption to different missions is particularly useful for commercial vehicles, as they are used in a wide range of applications and with customized designs. Specific cases are investigated for Light and Medium Heavy- Duty work trucks. The average power required by a vehicle to drive the regulatory cycles varies by nearly a factor 10 between the Class 4 vehicle on the ARB Transient cycle and the loaded Class 7 vehicle at 65 mph on grade.

The automotive industry is subject to stringent regulations in emissions and growing customer demands for better fuel consumption and vehicle performance. Engine calibration, a process that optimizes engine performance by tuning engine controls (actuators), becomes challenging nowadays due to significant increase of complexity of modern engines. The traditional sweep-based engine calibration method is no longer sustainable. To tackle the challenge, this work considers two powerful global optimization methods: genetic algorithm (GA) and Bayesian optimization for steady-state engine calibration for single speed-load point. GA is a branch of meta-heuristic methods that has shown a great potential on solving difficult problems in automotive engineering. Bayesian optimization is an efficient global optimization method that solves problems with computationally expensive testing such as hyperparameter tuning in deep neural network (DNN), engine testing, etc.

Previously fuel consumption on a drive cycle has been shown to be proportional to traction work, with an offset for powertrain losses. This model had different transfer functions for different drive cycles, performance levels, and applied powertrain technologies. Following Soltic it is shown that if fuel usage and traction work are both expressed in terms of cycle average power, a wide range of drive cycles collapse to a single transfer function, where cycle average traction power captures the drive cycle and the vehicle size. If this transfer function is then normalized by weight, i.e. by working in cycle average power/weight (P/W), a linear model is obtained where the offset is mainly a function of rated performance and applied technology. A final normalization by rated power/weight as the primary performance metric further collapses the data to express the cycle average fuel power/rated power ratio as a function of cycle average traction power/rated power ratio.

This paper proposes a transmission ratio scheduling and control methodology for a vehicle with a Continuous Variable Transmission (CVT) and a downsized gasoline engine. The methodology is designed to deliver the optimal vehicle fuel economy within drivability and performance constraints. Traditionally, the Optimum Operating Line (OOL) generated from an engine brake specific fuel consumption map is considered to be the best option for ratio scheduling, as it defines the points at which engine efficiency is maximized. But the OOL does not consider transmission efficiency, which may be a source of significant losses. To develop a CVT ratio schedule that offers the best fuel economy for the complete powertrain, an empirical approach was used to minimize fuel consumption by considering engine efficiency, CVT efficiency, and requested vehicle power. A backward-looking model was used to simulate a standard driving cycle (FTP-75) and develop a new powertrain-optimal operating line (P-OOL).

The new European test procedure, called the worldwide harmonized light vehicle test procedure (WLTP), deviates in some details from the current NEDC-based test which will have an impact on the determination of the official EU fuel consumption values for the new vehicles. The adaptation to the WLTP faces automakers with new challenges for meeting the stringent EU fuel consumption and CO2 emissions standards. This paper investigates the main changes that the new test implies to a mid-size sedan electrified vehicle design and quantifies their impact on the vehicles fuel economy. Three common electrified powertrain architectures including series, parallel P2, and powersplit are studied. A Pontryagin’s Minimum Principle (PMP) optimization-based energy management control strategy is developed to evaluate the energy consumption of the electrified vehicles in both charge-depleting (CD) and charge-sustaining (CS) modes.

Medium duty vehicles come in many design variations, which makes testing them all for CO2 impractical. As a result there are multiple ways of reporting CO2 emissions. Actual tests may be performed, data substitution may be used, or CO2 values may be estimated using an analytical correction. The correction accounts for variations in road load force coefficients (f0, f1, f2), weight, and axle ratio. The EPA Analytically Derived CO2 equation (EPA ADC) was defined using a limited set of historical data. The prediction error is shown to be ±130 g/mile and the sensitivities to design variables are found to be incorrect. Since the absolute CO2 is between 500 and 1,000 g/mi, the equation has limited usefulness. Previous work on light duty vehicles has demonstrated a linear relationship between vehicle fuel consumption, powertrain properties and total vehicle work. This relationship improves the accuracy and avoids co-linearity and non-orthogonality of the input variables.

The fuel consumption of a vehicle is shown to be linearly proportional to (1) total vehicle work required to drive the cycle due to mass and acceleration, tire friction, and aerodynamic drag and (2) the powertrain (PT) mechanical losses, which are approximately proportional to the engine displaced volume per unit distance travelled (displacement time gearing). The fuel usage increases linearly with work and displacement over a wide range of applications, and the rate of increase is inversely proportional to the marginal efficiency of the engine. The theoretical basis for these predictions is reviewed. Examples from current applications are discussed, where a single PT is used across several vehicles. A full vehicle cycle simulation model also predicts a linear relationship between fuel consumption, vehicle work, and displacement time gearing and agrees well with the application data.

Vehicle certification data are used to study the effectiveness of the major powertrain technologies used by car manufacturers to reduce fuel consumption. Methods for differentiating vehicles effectively were developed by leveraging theoretical models of engine and vehicle fuel consumption. One approach normalizes by displacement per unit distance, which puts both fuel used and vehicle work in mean effective pressure units, and is useful when comparing engine technologies. The other normalizes by engine rated power, a customer-relevant output metric. The normalized work/power is proportional to weight/power, the most fundamental performance metric. Certification data for 2016 and 2017 U.S. vehicles with different powertrain technologies are compared to baseline vehicles with port fuel injection (PFI) naturally aspirated engines and six-speed automatic transmissions.

Powertrain and vehicle technology is rapidly changing to meet the ever increasing demands of customers and government regulations. In some cases technologies that are designed to improve one attribute may impact others or interact with other design decisions in unexpected ways. Understanding the interactions and optimizing the transient performance at the vehicle level may require controls and calibration that is not available until late in the vehicle development process, after hardware changes are no longer possible. As a result, an efficient, up front, CAE process for assessing the interaction of various design choices on transient vehicle behavior is desirable. Building, calibrating and validating a vehicle system model with full controls and a mature calibration is very time consuming and often requires significant experimental data that is not available until it is too late to make hardware changes.

To assess the fuel consumption of vehicles, three sets of input data are required; drive cycles, vehicle parameters, and environmental conditions. As the first part of a series of studies on real-world fuel consumption, this study focuses on the drive cycles. In principle, drive cycles should represent real-world usage. Some of them aim at a specific usage such as a city driving condition or an aggressive driving style. However, the definition of city or aggressive driving is very subjective and difficult to quantitatively correlate with the real-world usage. This study proposes a methodology to quantify the speed and dynamics of drive cycles, or vehicle speed traces in general, against the real-world usage. After reviewing parameter sets found in other studies, relative cubic speed (RCS) and positive kinetic energy (PKE) are selected to represent the speed and dynamics through energy flow balance at the wheels.

An integral part of combustion system development for previous NA gasoline engines was the optimization of charge motion towards the best compromise in terms of full load performance, part load stability, emissions and, last but not least, fuel economy. This optimum balance may potentially be different in GTDI engines. While it is generally accepted that an increased charge motion level improves the mixture preparation in direct injection gasoline engines, the tradeoff in terms of performance seems to become less dominant as the boosting systems of modern engines are typically capable enough to compensate the flow losses generated by the more restrictive ports. Nevertheless, the increased boost level does not come free; increased charge motion generates higher pumping- and wall heat losses. Hence it is questionable and engine dependent, whether more charge motion is always better.

Low Temperature Combustion (LTC) engines are promising to improve powertrain fuel economy and reduce NOx and soot emissions by improving the in-cylinder combustion process. However, the narrow operating range of LTC engines limits the use of these engines in conventional powertrains. The engine’s limited operating range can be improved by taking advantage of electrification in the powertrain. In this study, a multi-mode LTC-SI engine is integrated with a parallel hybrid electric configuration, where the engine operation modes include Homogeneous Charge Compression Ignition (HCCI), Reactivity Controlled Compression Ignition (RCCI), and conventional Spark Ignition (SI). The powertrain controller is designed to enable switching among different modes, with minimum fuel penalty for transient engine operations.

The drive cycle average powertrain efficiency of current US vehicles is studied by applying a first principles model to the EPA Test Car List database. The largest group of vehicles has naturally aspirated engines and six speed planetary automatic transmissions, and defines the base technology level. For this group the best cycle average powertrain efficiency is independent of vehicle size and is achieved by the lowest power-to-weight vehicles. For all segments of the EPA test, the fuel required per unit of vehicle work (the inverse of powertrain efficiency), is found to increase linearly with a basic powertrain matching parameter. The parameter is (D/M)(n/V), where D is engine displacement, M vehicle mass, and (n/V) the top gear engine speed over the vehicle speed. The fuel consumption penalties in the City segments due to powertrain warm-up, aftertreatment warm-up, stop-and-go operation, and power-off operation are estimated.