Search Results

To advance vehicle lightweighting, chopped carbon fiber sheet molding compound (SMC) is identified as a promising material to replace metals. However, there are no effective tools and methods to predict the mechanical property of the chopped carbon fiber SMC due to the high complexity in microstructure features and the anisotropic properties. In this paper, a Representative Volume Element (RVE) approach is used to model the SMC microstructure. Two modeling methods, the Voronoi diagram-based method and the chip packing method, are developed to populate the RVE. The elastic moduli of the RVE are calculated and the two methods are compared with experimental tensile test conduct using Digital Image Correlation (DIC). Furthermore, the advantages and shortcomings of these two methods are discussed in terms of the required input information and the convenience of use in the integrated processing-microstructure-property analysis.

SI engines fueled with hydrogen represent a promising powertrain solution to meet the ambitious target of carbon-free emissions at the tailpipe. Therefore, fast and reliable numerical tools can significantly support the automotive industry in the optimization of such technology. In this work, a 1D-3D methodology is presented to simulate in detail the combustion process with minimal computational effort. First, a 1D analysis of the complete engine cycle is carried out on the user-defined powertrain configuration. The purpose is to achieve reliable boundary conditions for the combustion chamber, based on realistic engine parameters. Then, a 3D simulation of the power-cycle is performed to mimic the combustion process. The flow velocity and turbulence distributions are initialized without the need of simulating the gas exchange process, according to a validated technique.

In recent years, with the increase of traffic accidents and traffic jams, lane change, as one of the most important and commonly automatic driving operations for autonomous vehicles, is receiving attention in academia. It is considered to be one of the important solutions that play an important role in improving road traffic safety and efficiency. However, most existing lane-changing models are rule-based lane-changing models. These models only assume a one-direction impact of surrounding vehicles on the lane-changing vehicle. In fact, lane change is a process of mutual interaction between vehicles due to the complexity and uncertainty of the traffic environment. Moreover, the safety and efficiency of existing lane-changing decision algorithms need to be improved. In this paper, we proposed a multivehicle cooperative control approach with a distributed control structure to control the model.

Ammonia has attracted the attention of a growing number of researchers in recent years. However, some properties of ammonia (e.g., low laminar burning velocity, high ignition energy, etc.) inhibit its direct application in engines. Several routes have been proposed to overcome these problems, such as oxygen enrichment, partial fuel cracking strategy and co-combustion with more reactive fuels. Improving the reactivity of ammonia from the oxidizer side is also practical. Ozone is a highly reactive oxidizer which can be easily and rapidly generated through electrical plasma and is an effective promoter applicable for a variety of fuels. The dissociation reaction of ozone increases the concentration of reactive radicals and promotes chain-propagating reactions. Thus, obtaining accurate rate constants of reactions related to ozone is necessary, especially at elevated to high pressure range which is closer to engine-relevant conditions.

The prediction of agents' future trajectory is a crucial task in supporting advanced driver-assistance systems (ADAS) and plays a vital role in ensuring safe decisions for autonomous driving (AD). Currently, prevailing trajectory prediction methods heavily rely on high-definition maps (HD maps) as a source of prior knowledge. While HD maps enhance the accuracy of trajectory prediction by providing information about the surrounding environment, their widespread use is limited due to their high cost and legal restrictions. Furthermore, due to object occlusion, limited field of view, and other factors, the historical trajectory of the target agent is often incomplete This limitation significantly reduces the accuracy of trajectory prediction. Therefore, this paper proposes ETSA-Pred, a mapless trajectory prediction model that incorporates enhanced temporal modeling and spatial self-attention.

Fuel film deposits on combustion chamber walls are understood to be the main source of particle emissions in GDI engines under homogenous charge operation. More precisely, the liquid film that remains on the injector tip after the end of injection is a fuel rich zone that undergoes pyrolysis reactions leading to the formation of poly-aromatic hydrocarbons (PAH) known to be the precursors of soot. The physical phenomena accompanying the fuel film deposit, evaporation, and the chemical reactions associated to the injector film are not yet fully understood and require high fidelity CFD simulations and controlled experimental campaigns in optically accessible engines. To this end, a simplified model based on physical principles is developed in this work, which couples an analytical model for liquid film formation and evaporation on the injector tip with a stochastic particle dynamics model for particle formation.

Precise trajectory prediction of surrounding vehicles is critical for decision-making of autonomous vehicles, and learning-based approaches are well recognized for the robustness. However, state-of-the-art learning-based methods ignore (1) the feasibility of the vehicle’s multimodal state information for prediction and (2) the mutually exclusive relationship between the global traffic scene receptive fields and the local position resolution when modeling vehicles’ interactions, which may influence prediction accuracy. Therefore, we propose a “vehicle descriptor”-based long short-term memory (LSTM) model with the dilated convolutional social pooling (VD+DCS-LSTM) to cope with the above issues.

Accurate and fast positioning of large aircraft component is of great importance for Automated Alignment System. The Ball joint is a widely-used mechanical device connecting the aircraft component and positioners. However, there are some shortcomings for the device in man-machine engineering, such as the entry state of the ball-head still needs to be confirmed by the workers and then switched to the locking state manually. To solve above problems, a new positioning mechanism is present in this paper, which consists of a ball-head and a ball-socket. The new device is equipped with a monocular vision system, in which a calibrated industrial camera is used to collect the images of the ball-head. And then, the 3-D coordinate of the ball-head center is calculated by a designed algorithm, guiding the positioner to capture the ball-head. Once the ball-head gets into the ball-socket, the pneumatic system will drive the pistons to move to the specified location.

To achieve higher efficiencies and lower emissions, dual-fuel strategies have arisen as advanced engine technologies. In order to fully utilize engine fuels, understanding the combustion chemistry is urgently required. However, due to computation limitations, detailed kinetic models cannot be used in numerical engine simulations. As an alternative, approaches for developing reduced reaction mechanisms have been proposed. Nevertheless, existing simplified methods neglecting the real engine combustion processes, which is the ultimate goal of reduced mechanism. In this study, we propose a novel simplified approach based on fuel reactivity. The high-reactivity fuel undergoes pyrolysis first, followed by the pyrolysis and oxidation of the low-reactivity fuel. Therefore, the simplified mechanism consists of highly lumped reactions of high-reactivity fuel, radical reactions of low-reactivity fuel and C0-C2 core mechanisms.

Conventionally, the engines are calibrated under the assumption that engines will be made exactly to the prints, and all the engines from the same batch will be identical. However, engine-to-engine variations do exist which will affect the engine performances, and part-to-part variations, i.e., the tolerance, is an important factor leading to engine-to-engine variations. There are researches conducted on the influence of dimensional tolerances on engine performance, however, the impact of straightness, which is an important geometric tolerance, on lubrication is an unsolved issue. This study presents a systematic method to model the straightness and to analyze its effects on the friction loss. The bearing model is built based on elastohydrodynamic (EHD) theory. Meanwhile a novel modeling method to represent any form of straightness in three-dimensional space is proposed.

A novel normal measurement device for robotic drilling and countersinking has been developed. This device is mainly composed of three contact displacement sensors and a spherically compliant clamp pad. The compliance of the clamp pad allows it to be perpendicular to the part when the Multi-Function End Effector (MFEE) drives it to clamp the part surface prior to drilling, while the displacement sensors are used to measure the movement of the clamp pad relative to the MFEE. Once the sensors’ position is calibrated, the rotation angle of the clamp pad can be calculated by the displacement of the sensors. Then, the normal adjustment of MFEE is obtained, and the adjustment process can be achieved by the Rotation Tool Center Point (RTCP) function of robot. Thus, an innovative method based on laser tracker to identify the position of sensors is proposed.

The 4th Workshop of the Engine Combustion Network (ECN) was held September 5-6, 2015 in Kyoto, Japan. This manuscript presents a summary of the progress in experiments and modeling among ECN contributors leading to a better understanding of soot formation under the ECN “Spray A” configuration and some parametric variants. Relevant published and unpublished work from prior ECN workshops is reviewed. Experiments measuring soot particle size and morphology, soot volume fraction (fv), and transient soot mass have been conducted at various international institutions providing target data for improvements to computational models. Multiple modeling contributions using both the Reynolds Averaged Navier-Stokes (RANS) Equations approach and the Large-Eddy Simulation (LES) approach have been submitted. Among these, various chemical mechanisms, soot models, and turbulence-chemistry interaction (TCI) methodologies have been considered.

An existing one-dimensional (1D) spray model, which successfully captures inert spray processes, has been extended to enable prediction of ignition delay and lift-off length under reacting conditions. For that purpose, an additional transport equation for the progress variable has been incorporated, which includes detailed chemistry effects by means of a tabulation method based upon an external flamelet solver. The transport equation for the progress variable is solved in a quasi-1D fashion, along presumed mixture fraction trajectories, while the 1D approach is retained for the mixture fraction and axial velocity fields. The paper includes the model development, as well as the validation against Spray A measurements from the Engine Combustion Network. In spite of the simplified approach, the model captures some of the experimental trends of the lift-off length and ignition delay with a quite low computational cost.

Aiming at the problem of weak communication, strong interference, cross-domain, and large-scale environment, it is difficult to achieve efficient decision-making and planning in the collaborative operation of intelligent groups. Based on the SOM algorithm, this paper proposes a dual-selection allocation and distributed vectorized trajectory planning. Form a collaborative planning algorithm that can be updated with high frequency and a rational decision-making mechanism. Provide technical support for collaborative search and detection of intelligent groups. At the same time, based on the principle of minimum consistency, this paper proposes a clock synchronization model under spatial coordination and conducts simulation experiments to verify it. The result proves the efficiency and practicability of the collaborative intelligent decision-making plan proposed in this paper.

The energy management strategies (EMS) for hybrid electric vehicles (HEV) have a great impact on the fuel economy (FE). The Pontryagin's minimum principle (PMP) has been proved to be a viable control strategy for HEV. The optimal costate of the PMP control can be determined by the given information of the driving conditions. Since the full knowledge of future driving conditions is not available, this paper proposed a dynamic optimization method for PMP costate without the prediction of the driving cycle. It is known that the lower fuel consumption the method yields, the more efficiently the engine works. The selection of costate is designed to make the engine work in the high efficiency range. Compared with the rule-based control, the proposed method by the principle of Hamiltonian, can make engine working points have more opportunities locating in the middle of high efficiency range, instead of on the boundary of high efficiency range.

The Engine Combustion Network (ECN) is a coordinate effort from research partners from all over the world which aims at creating a large experimental database to validate CFD calculations. Two injectors from ECN, namely Spray C and D, have been compared in a constant pressure flow vessel, which enables a field of view of more than 100 mm. Both nozzles have been designed with similar flow metrics, with Spray D having a convergent hole shape and Spray C a cylindrical one, the latter being therefore more prone to cavitation. Although the focus of the study is on reacting conditions, some inert cases have also been measured. High speed schlieren imaging, OH* chemiluminescence visualization and head-on broadband luminosity have been used as combustion diagnostics to evaluate ignition delay, lift off length and reacting tip penetration. Parametric variations include ambient temperature, oxygen content and injection pressure variations.

Advanced Computational Fluid Dynamics (CFD) modeling of reacting sprays provides access to information not available even applying the most advanced experimental techniques. This is particularly evident if the combustion model handles detailed chemical kinetic models efficiently to describe the fuel auto-ignition and oxidation processes. Complex chemistry also provides the temporal evolution of key species closely related to emissions formation, such as polycyclic aromatic hydrocarbons (PAHs) that are well-known as soot precursors. In this framework, present investigation focuses on the analysis of the so-called Spray-A combustion characteristics using two different flamelet-based combustion models. Both Reynolds-Averaged Navier-Stokes (RANS) and Large-Eddy Simulation (LES) predictions are combined to study not only the averaged spray characteristics, but also the relevance of different realizations in this particular problem.

Most studies about common-rail diesel injection consider the fuel flow along the injector as isothermal. This hypothesis is arguable given the small diameter of the orifices along which the fuel flows, together with the expansions that take place across them. These phenomena may provoke variations in the fuel temperature, which in turn modify the fuel properties (i.e. viscosity, density, speed of sound…), thus influencing injector dynamics as well as the fuel atomization and mixing processes. The present investigation accounts for these effects by means of a 1D model for the fuel flow along a common-rail ballistic injector. Local variations of fuel temperature and pressure are considered by the model thanks to the implementation of the adiabatic flow hypothesis.

For the purposes of on-line control, e.g., in an automatic driving system, or of closed-loop directional control simulation, an optimal preview artificial neural network (ANN) driver model based on error elimination algorithm(EEA) is built. Then the optimal preview times are discussed in high frequency range in this system. The simulation results of optimal preview ANN driver model and Error Elimination Algorithm driver model are compared under the condition of different vehicle speeds and paths, which shows that the proposed approach is efficient and reliable enough, particularly for driver-vehicle closed-loop system.

The improvement mechanism of fuel consumption at partial and full loads of a boosted direction-injection gasoline engine with the elevated geometrical compression ratio and Miller cycle by either early or late intake valve closing (EIVC or LIVC) are analyzed based on the first law of thermodynamics and one dimensional engine simulation. An increase in geometric compression ratio increases the theoretical thermal efficiency for all the operating loads, but deteriorates the fuel economy at full loads, owing primarily to the full-load knock limit. Use of Miller cycle improves the fuel economy for both the partial and full load operations by reducing the pumping loss and optimizing the combustion phasing, respectively. A comparison between EIVC and LIVC on the influencing factors on the thermal efficiency at the partial load shows that EIVC leads to higher mechanical efficiency and less heat transfer loss than LIVC, and hence its efficiency improvement is superior over LIVC.