Search Results

The development of autonomous driving generally requires enormous annotated data as training input. The availability and quality of annotated data have been major restrictions in industry. Data synthesis techniques are then being developed to generate annotated data. This paper proposes a 2D data synthesis pipeline using original background images and target templates to synthesize labeled data for model training in autonomous driving. The main steps include: acquiring templates from template libraries or alternative approaches, augmenting the obtained templates with diverse techniques, determining the positioning of templates in images, fusing templates with background images to synthesize data, and finally employing the synthetic data for subsequent detection and segmentation tasks. Specially, this paper synthesizes traffic data such as traffic signs, traffic lights, and ground arrow markings in 2D scenes based on the pipeline.

An appropriate energy management strategy can further reduce the fuel consumption of P2 hybrid electric vehicles (HEV) with simple hybrid configuration and low cost. The rule-based real-time energy management strategy dominates the energy management strategies utilized in commercial HEVs, due to its robustness and low computational loads. However, its performance is sensitive to the setting of parameters and control actions. To further improve the fuel economy of a P2 HEV, the energy management strategy of the HEV has been re-designed based on the globally optimal control theory. An optimization strategy model based on the longitudinal dynamics of the vehicle and Bellman’s dynamic programming algorithm was established in this research and an optimal power split in the dual power sources including an internal combustion engine (ICE) and an electric machine at a given driving cycle was used as a benchmark for the development of the rule-based energy management strategy.

Pumping Mean Effective Pressure (PMEP) is the main factor limiting the improvement of thermal efficiency in a spark-ignition (SI) engine under low load. One of the ways to reduce the pumping loss under low load is to use Cylinder DeActivation (CDA). The CDA aims at reducing the firing density (FD) of the SI engine under low load operation and increasing the mass of air-fuel mixture within one cycle in one cylinder to reduce the throttling effect and further reducing the PMEP. The multi-stroke cycles can also reduce the firing density of the SI engine after some certain reasonable design, which is feasible to improve the thermal efficiency of the engine under low load in theory. The research was carried out on a calibrated four-cylinder SI engine simulation platform. The thermal efficiency improvements of the 6-stroke cycle and 8-stroke cycle to the engine performance were studied compared with the traditional 4-stroke cycle under low load conditions.

With the rapid development of autonomous driving technologies, the proportion of autonomous vehicles (AVs) will increase and influence road capacity. In this study, the simulation of mixed traffic flow was studied using an improved cellular automata model. Safety inter-vehicle spacing, the length of vehicles and reaction time are introduced into the cellular automata model. We delete the acceleration, deceleration and randomization rule for ideal conditions. Numerical simulations are utilized to analyze road capacity with different proportions of AVs. Road capacity is about 2200 pcu/h/lane for pure manual vehicle (MV) traffic flow and about 3600 pcu/h/lane for pure AV traffic flow. The capacity increases by 19.2% when there are 50% AVs in the traffic flow. And the capacity increases by around 63.6% due to the pure AV traffic flow.

The Gasoline direct-injection (GDI) engine can emit high levels of particulate matter and unburned Hydrocarbons when operating under stratified charge combustion mode. Injecting late in the compression stroke means the fuel has insufficient time to atomise and evaporate. This could cause fuel film accumulation on the piston surface and combustion liner. Locally fuel rich diffusion combustion could also result in the formation of soot particles. Employing a split-injection strategy can help tackle these issues. The first injection is initiated early in the intake stroke and could ensure a global homogeneous charge. The second injection during the compression stroke could help form a fuel-rich charge in the vicinity of the spark plug. Many studies have established the crucial role that a split-injection strategy plays in the stratified charge operation of GDI engines.

Engine downsizing is an effective technology to lower automotive CO2 emissions. However, the high load low speed regions are plagued with knocking combustion that are usually overcome by retarding the ignition. This interferes with the efficiency gains due to very late combustion. This paper reports the use of Exhaust Gas Recirculation (EGR) on a Ford Ecoboost 1l downsized gasoline turbocharged direct injection (GTDI) engine to improve efficiency by optimising combustion phasing unlocked by the improved knock resistance with EGR dilution. Further ignition system upgrades are tested for impact towards further efficiency improvements. 75mJ (standard) and 120mJ (high energy) ignition systems were compared. The experimental results showed that the brake specific fuel consumption (BSFC) can be improved by 5.6% with EGR dilution at 25%. When considering combined effects of EGR and high energy ignition upon engine fuel economy, the BSFC gain improves to 7.9%.

Controlled Auto-Ignition (CAI), also known as Homogeneous Charge Compression Ignition (HCCI), can improve the fuel economy of gasoline engines and simultaneously achieve ultra-low NOx emissions. However, the difficulty in combustion phasing control and violent combustion at high loads limit the commercial application of CAI combustion. To overcome these problems, stratified mixture, which is rich around the central spark plug and lean around the cylinder wall, is formed through port fuel injection and direct injection of gasoline. In this condition, rich mixture is consumed by flame propagation after spark ignition, while the unburned lean mixture auto-ignites due to the increased in-cylinder temperature during flame propagation, i.e., stratified flame ignited (SFI) hybrid combustion.

Metal foam with their high porosity and heat storage capacity can be combined with phase change materials to be a powerful heat storage device. Numerical simulations of metal foam behavior can be challenging due to their complex geometric patterns necessitating high mesh requirements. Furthermore, simulations of the inner workings of a metal foam heat exchanger comprising of a large number of individual metal foam canisters can be impossible. The objective of the current work is to develop a computational model using a proprietary CFD tool Simerics-MP/Simerics-MP+® to simulate the workings of a metal foam heat exchanger with phase change element. A heat transfer coefficient capturing this heat transfer between wax and metal is used to formulate the “simplified” mixture model. The versatility of the proposed model is in the universality of its application to any shape or structure of metal foam. The computational model developed is tested to replicate the results of the 3D simulation.

Engine downsizing is one of the most effective means to improve the fuel economy of spark ignition (SI) gasoline engines because of lower pumping and friction losses. However, the occurrence of knocking combustion or even low-speed pre-ignition at high loads is a severe problem. One solution to significantly increase the upper load range of a 4-stroke gasoline engine is to use 2-stroke cycle due to the double firing frequency at the same engine speed. It was found that a 0.7 L two-cylinder 2-stroke poppet valve gasoline engine equipped with a two-stage serial boosting system, comprising a supercharger and a downstream turbocharger, could replace a 1.6 L naturally aspirated 4-stroke gasoline engine in our previous research, but its fuel economy was close to that of the 4-stroke engine at upper loads due to knocking combustion.

In order to meet increasingly stringent emissions standards and lower the fuel consumption of heavy-duty (HD) vehicles, significant efforts have been made to develop high efficiency and clean diesel engines and aftertreatment systems. However, a trade-off between the actual engine efficiency and nitrogen oxides (NOx) emission remains to minimize the operational costs. In addition, the conversion efficiency of the diesel aftertreatment system decreases rapidly with lower exhaust gas temperatures (EGT), which occurs at low load operations. Thus, it is necessary to investigate the optimum combustion and engine control strategies that can lower the vehicle’s running costs by maintaining low engine-out NOx emissions while increasing the conversion efficiency of the NOx aftertreament system through higher EGTs.

In an automotive cooling circuit, the wax melting process determines the net and time history of the energy transfer between the engine and its environment. A numerical process that gives insight into the mixing process outside the wax chamber, the wax melting process inside the wax chamber, and the effect on the poppet valve displacement will be advantageous to both the engine and automotive system design. A fully three dimensional, transient, system level simulation of an inlet controlled thermostat inside an automotive cooling circuit is undertaken in this paper. A proprietary CFD algorithm, Simerics-Sys®/PumpLinx®, is used to solve this complex problem. A two-phase model is developed in PumpLinx® to simulate the wax melting process. The hysteresis effect of the wax melting process is also considered in the simulation.

This paper describes a novel design and verification process for analytical methods used in the development of advanced combustion strategies in internal combustion engines (ICE). The objective was to improve brake thermal efficiency (BTE) as part of the US Department of Energy SuperTruck program. The tools and methods herein discussed consider spray formation and injection schedule along with piston bowl design to optimize combustion efficiency, air utilization, heat transfer, emission, and BTE. The methodology uses a suite of tools to optimize engine performance, including 1D engine simulation, high-fidelity CFD, and lab-scale fluid mechanic experiments. First, a wide range of engine operating conditions are analyzed using 1-D engine simulations in GT Power to thoroughly define a baseline for the chosen advanced engine concept; secondly, an optimization and down-select step is completed where further improvements in engine geometries and spray configurations are considered.

Modern heavy duty diesel engines can well extend the goal of 50% brake thermal efficiency by utilizing waste heat recovery (WHR) technologies. The effect of an ORC WHR system on engine brake specific fuel consumption (bsfc) is a compromise between the fuel penalty due to the higher exhaust backpressure and the additional power from the WHR system that is not attributed to fuel consumption. This work focuses on the fuel efficiency benefits of installing an ORC WHR system on a heavy duty diesel engine. A six cylinder, 7.25ℓ heavy duty diesel engine is employed to experimentally explore the effect of backpressure on fuel consumption. A zero-dimensional, detailed physical ORC model is utilized to predict ORC performance under design and off-design conditions.

As the emission regulations for internal combustion engines are becoming increasingly stringent, different solutions have been researched and developed, such as dual injection systems (combined port and direct fuel injection), split injection strategies (single and multiple direct fuel injection) and different intake air devices to generate an intense in-cylinder air motion. The aim of these systems is to improve the in-cylinder mixture preparation (in terms of homogeneity and temperature) and therefore enhance the combustion, which ultimately increases thermal efficiency and fuel economy while lowering the emissions. This paper describes the effects of dual injection systems on combustion, efficiency and emissions of a downsized single cylinder gasoline direct injection spark ignited (DISI) engine. A set of experiments has been conducted with combined port fuel and late direct fuel injection strategy in order to improve the combustion process.

Engine downsizing can effectively improve the fuel economy of spark ignition (SI) gasoline engines, but extreme downsizing is limited by knocking combustion and low-speed pre-ignition at higher loads. A 2-stroke SI engine can produce higher upper load compared to its naturally aspirated 4-stroke counterpart with the same displacement due to the double firing frequency at the same engine speed. To determine the potential of a downsized two-cylinder 2-stroke poppet valve SI gasoline engine with 0.7 L displacement in place of a naturally aspirated 1.6 L gasoline (NA4SG) engine, one-dimensional models for the 2-stroke gasoline engine with a single turbocharger and a two-stage supercharger-turbocharger boosting system were set up and validated by experimental results.

Engine downsizing of the spark ignition gasoline engine is recognized as one of the most effective approaches to improve the fuel economy of a passenger car. However, further engine downsizing beyond 50% in a 4-stroke gasoline engine is limited by the occurrence of abnormal combustion events as well as much greater thermal and mechanical loads. In order to achieve aggressive engine downsizing, a boosted uniflow scavenged direct injection gasoline (BUSDIG) engine concept has been proposed and researched by means of CFD simulation and demonstration in a single cylinder engine. In this paper, the intake port design on the in-cylinder flow field and gas exchange characteristics of the uniflow 2-stroke cycle was investigated by computational fluid dynamics (CFD). In particular, the port orientation on the in-cylinder swirl, the trapping efficiency, charging efficiency and scavenging efficiency was analyzed in details.

An experimental study has been carried out with the end goal of minimizing engine-out methane emissions with Premixed Micro Pilot Combustion (PMPC) in a natural gas-diesel Dual-Fuel™ engine. The test engine used is a heavy-duty single cylinder engine with high pressure common rail diesel injection as well as port fuel injection of natural gas. Multiple variables were examined, including injection timings, exhaust gas recirculation (EGR) percentages, and rail pressure for diesel, conventional Dual-Fuel, and PMPC Dual-Fuel combustion modes. The responses investigated were pressure rise rate, engine-out emissions, heat release and indicated specific fuel consumption. PMPC reduces methane slip when compared to conventional Dual-Fuel and improves emissions and fuel efficiency at the expense of higher cylinder pressure.

Controlled Auto Ignition (CAI), also known as Homogeneous Charge Compression Ignition (HCCI), is one of the most promising combustion technologies to reduce the fuel consumption and NOx emissions. Most research on CAI/HCCI combustion operations have been carried out in 4-stroke gasoline engines, despite it was originally employed to improve the part-load combustion and emission in the two-stroke gasoline engine. However, conventional ported two-stroke engines suffer from durability and high emissions. In order to take advantage of the high power density of the two-stroke cycle operation and avoid the difficulties of the ported engine, systematic research and development works have been carried out on the two-stroke cycle operation in a 4-valves gasoline engine. CAI combustion was achieved over a large range of operating conditions when the relative air/fuel ratio (lambda) was kept at one as measured by an exhaust lambda sensor.

Vehicle weight reduction has become one of the essential research areas in the automotive industry. It is important to perform design optimization of Body-in-White (BIW) at the concept design phase so that to reduce the development cost and shorten the time-to-market in later stages. Finite Element (FE) models are commonly used for vehicle design. However, even with increasing speed of computers, the simulation of FE models is still too time-consuming due to the increased complexity of models. This calls for the development of a systematic and efficient approach that can effectively perform vehicle weight reduction, while satisfying the stringent safety regulations and constraints of development time and cost. In this paper, an efficient BIW weight reduction approach is proposed with consideration of complex safety and stiffness performances. A parametric BIW FE model is first constructed, followed by the building of surrogate models for the responses of interest.

SI-CAI hybrid combustion, also known as spark-assisted compression ignition (SACI), is a promising concept to extend the operating range of CAI (Controlled Auto-Ignition) and achieve the smooth transition between spark ignition (SI) and CAI in the gasoline engine. In order to investigate the effect of the thermal boundary condition on the hybrid combustion, the experiments with different coolant temperatures are performed to adjust the chamber wall temperature in a gasoline engine. The experimental results indicate that increasing wall temperature would advance the combustion phasing, enlarge the peak heat release rate and shorten the combustion duration. While the capacity of the wall temperature effect on the hybrid combustion characteristics are more notable in the auto-ignition dominated hybrid combustion.