Search Results

This works presents a Reinforcement Learning (RL) agent to implement a Cooperative Adaptive Cruise Control (CACC) system that simultaneously enhances energy efficiency and comfort, while also ensuring string stability. CACC systems are a new generation of ACC which systems rely on the communication of the so-called ego-vehicle with other vehicles and infrastructure using V2V and/or V2X connectivity. This enables the availability of robust information about the environment thanks to the exchange of information, rather than their estimation or enabling some redundancy of data. CACC systems have the potential to overcome one typical issue that arises with regular ACC, that is the lack of string stability. String stability is the ability of the ACC of a vehicle to avoid unnecessary fluctuations in speed that can cause traffic jams, dampening these oscillations along the vehicle string rather than amplifying them.

Predictive combustion models are useful tools towards the development of clean and efficient engines operating with alternative fuels. This work intends to validate two different combustion models on compression-ignition engines fueled with Dimethyl Ether. Both approaches give a detailed characterization of the combustion kinetics, but they substantially differ in how the interaction between fluid-dynamics and chemistry is treated. The first one is single-flamelet Representative Interactive Flamelet, which considers turbulence-kinetic interaction but cannot correctly describe the stabilization of the flame. The second, named Tabulated Well Mixed, correctly accounts for local flow and mixture conditions but does not consider interaction between turbulence and chemistry. An experimental campaign was carried out on a heavy-duty truck engine running on DME at a constant load considering trade-off of EGR and SOI.

Fuel cell electrified powertrains are currently a promising technology towards decarbonizing the heavy-duty transportation sector. In this context, extensive research is required to thoroughly assess the hydrogen economy potential of fuel cell heavy-duty electrification. This paper proposes a real-time capable energy management strategy (EMS) that can achieve improved hydrogen economy for a fuel cell electrified heavy-duty truck. The considered heavy-duty truck is modelled first in Simulink® environment. A baseline heuristic map-based controller is then retained that can instantaneously control the electrical power split between fuel cell system and the high-voltage battery pack of the heavy-duty truck. Particle swarm optimization (PSO) is consequently implemented to optimally tune the parameters of the considered EMS.

Diesel combustion is a very complex process, involving a reacting, turbulent and multi-phase flow. Furthermore, heavy duty engines operate mainly at medium and high loads, where injection durations are very long and cylinder pressure is high. Within such context, proper CFD tools are necessary to predict mixing controlled combustion, heat transfer and, eventually, flame wall interaction which might result from long injection durations and high injection pressures. In particular, detailed chemistry seems to be necessary to estimate correctly ignition under a wide range of operating conditions and formation of rich combustion products which might lead to soot formation. This work is dedicated to the identification of suitable methodologies to predict combustion in heavy-duty diesel engines using detailed chemistry.

Dimethyl ether (DME) represents a promising fuel for heavy-duty engines thanks to its high cetane number, volatility, absence of aromatics, reduced tank-to-wheel CO2 emissions compared to Diesel fuel and the possibility to be produced from renewable energy sources. However, optimization of compression-ignition engines fueled with DME requires suitable computational tools to design dedicated injection and combustion systems: reduced injection pressures and increased nozzle diameters are expected compared to conventional Diesel engines, which influences both the air-fuel mixing and the combustion process. This work intends to evaluate the validity of two different combustion models for the prediction of performance and pollutant emissions in compression-ignition engines operating with DME. The first one is the Representative Interactive Flamelet while the second is the Approximated Diffusive Flamelet.

The potential of internal EGR (iEGR) and external EGR (eEGR) in reducing the engine-out NOx emissions in a heavy-duty diesel engine has been investigated by means of a refined 1D fluid-dynamic engine model developed in the GT-Power environment. The engine is equipped with Variable Valve Actuation (VVA) and Variable Geometry Turbocharger (VGT) systems. The activity was carried out in the frame of the CORE Collaborative Project of the European Community, VII FP. The engine model integrates an innovative 0D predictive combustion algorithm for the simulation of the HRR (heat release rate) based on the accumulated fuel mass approach and a multi-zone thermodynamic model for the simulation of the in-cylinder temperatures. NOx emissions are calculated by means of the Zeldovich thermal and prompt mechanisms.

Traditional CACC systems utilize inter-vehicle wireless communication to maintain minimal yet safe inter-vehicle distances, thereby improving traffic efficiency. However, introducing communication delays generates system uncertainties that jeopardize string stability, a crucial requirement for robust CACC performance. To address these issues, we introduce a decentralized Model Predictive Control (MPC) approach that incorporates Kalman Filters and state predictors to counteract the uncertainties posed by noise and communication delays. We validate our approach through MATLAB Simulink simulations, using stochastic and mathematical models to capture vehicular dynamics, Wi-Fi communication errors, and sensor noises. In addition, we explore the application of a Reinforcement Learning (RL)-based algorithm to compare its merits and limitations against our decentralized MPC controller, considering factors like feasibility and reliability.

Pressure-based and model-based techniques for the control of MFB50 (crank angle at which 50% of the fuel mass fraction has burned) have been developed, assessed and tested by means of rapid prototyping (RP) on a FPT F1C 3.0L Euro VI diesel engine. The pressure-based technique requires the utilization of a pressure transducer for each cylinder. The transducers are used to perform the instantaneous measurement of the in-cylinder pressure, in order to derive its corresponding burned mass fraction and the actual value of MFB50. It essentially consists of a closed-loop approach, which is based on a cycle-by-cycle and cylinder-to-cylinder correction of the start of injection of the main pulse (SOImain), in order to achieve the desired target of MFB50 for each cylinder.

Natural gas is a promising alternative fuel for internal combustion engine application due to its low carbon content and high knock resistance. Performance of natural gas engines is further improved if direct injection, high turbocharger boost level, and variable valve actuation (VVA) are adopted. Also, relevant efficiency benefits can be obtained through downsizing. However, mixture quality resulting from direct gas injection has proven to be problematic. This work aims at developing a mono-fuel small-displacement turbocharged compressed natural gas engine with side-mounted direct injector and advanced VVA system. An injector configuration was designed in order to enhance the overall engine tumble and thus overcome low penetration.

Computational fluid dynamics represents a useful tool to support the design and development of Heavy Duty Engines, making possible to test the effects of injection strategies and combustion chamber design for a wide range of operating conditions. Predictive models are required to ensure accurate estimations of heat release and the main pollutant emissions within a limited amount of time. For this reason, both detailed chemistry and turbulence chemistry interaction need to be included. In this work, the authors intend to apply combustion models based on tabulated kinetics for the prediction of Diesel combustion in Heavy Duty Engines. Four different approaches were considered: well-mixed model, presumed PDF, representative interactive flamelets and flamelet progress variable. Tabulated kinetics was also used for the estimation of NOx emissions.

The modeling of fuel sprays under well-characterized conditions relevant for heavy-duty Diesel engine applications, allows for detailed analyses of individual phenomena aimed at improving emission formation and fuel consumption. However, the complexity of a reacting fuel spray under heavy-duty conditions currently prohibits direct simulation. Using a systematic approach, we extrapolate available spray models to the desired conditions without inclusion of chemical reactions. For validation, experimental techniques are utilized to characterize inert sprays of n-dodecane in a high-pressure, high-temperature (900 K) constant volume vessel with full optical access. The liquid fuel spray is studied using high-speed diffused back-illumination for conditions with different densities (22.8 and 40 kg/m3) and injection pressures (150, 80 and 160 MPa), using a 0.205-mm orifice diameter nozzle.

The present paper is concerned with part of the work performed by Renault, IFPEN and Politecnico di Torino within a research project founded by the European Commission. The project has been focused on the development of a dedicated CNG engine featuring a 25% decrease in fuel consumption with respect to an equivalent Diesel engine with the same performance targets. To that end, different technologies were implemented and optimized in the engine, namely, direct injection, variable valve timing, LP EGR with advanced turbocharging, and diluted combustion. With specific reference to diluted combustion, it is rather well established for gasoline engines whereas it still poses several critical issues for CNG ones, mainly due to the lower exhaust temperatures. Moreover, dilution is accompanied by a decrease in the laminar burning speed of the unburned mixture and this generally leads to a detriment in combustion efficiency and stability.

Sustainable mobility has become a major issue for internal combustion engines and has led to increasing research efforts in the field of alternative fuels, such as bio-fuel, CNG and hydrogen addition, as well as into engine design and control optimization. To that end, a thorough control of the air-to-fuel ratio appears to be mandatory in SI engine in order to meet the even more stringent thresholds set by the current regulations. The accuracy of the air/fuel mixture highly depends on the injection system dynamic behavior and to its coupling to the engine fluid-dynamic. Thus, a sound investigation into the mixing process can only be achieved provided that a proper analysis of the injection rail and of the injectors is carried out. The present paper carries out a numerical investigation into the fluid dynamic behavior of a commercial CNG injection system by means of a 0D-1D code.

A contemporary approach for improving and developing the understanding of heavy-duty Diesel engine combustion processes is to use a concerted effort between experiments at well-characterized boundary conditions and detailed, high-fidelity models. In this paper, combustion processes of n-dodecane fuel sprays under heavy-duty Diesel engine conditions are investigated using this approach. Reacting fuel sprays are studied in a constant-volume pre-burn vessel at an ambient temperature of 900 K with three reference cases having specific combinations of injection pressure, ambient density and ambient oxygen concentration (80, 150 & 160 MPa - 22.8 & 40 kg/m3-15 & 20.5% O2). In addition to a free jet, two different walls were placed inside the combustion vessel to study flame-wall interaction.

Reducing computational time has become a critical issue in recent years, particularly in the transportation field, where the complexity of scenarios demands lightweight controllers to run large simulations and gather results to study different behaviors. This study proposes two novel formulations of the Optimal Control Problem (OCP) for the Energy Management System of a Plug-in Hybrid Electric Vehicle (PHEV) and compares their performance with a benchmark found in the literature. Dynamic Programming was chosen as the optimization algorithm to solve the OCP in a Matlab environment, using the DynaProg toolbox. The objective is to address the optimality of the fuel economy solution and computational time. In order to improve the computational efficiency of the algorithm, an existing formulation from the literature was modified, which originally utilized three control inputs.

This paper deals with the energy efficiency of cooperative cruise control technologies when considering vehicle strings in a realistic driving environment. In particular, we design a cooperative longitudinal controller using a state-of-the-art model predictive control (MPC) implementation. Rather than testing our controller on a limited set of short maneuvers, we thoroughly assess its performance on a number of regulatory drive cycles and on a set of driving missions of similar length that were constructed based on real driving data. This allows us to focus our assessment on the energetic aspects in addition to testing the controller’s robustness. The analyzed controller, based on linear MPC, uses vehicle sensor data and information transmitted by the vehicle driving the string to adjust the longitudinal trajectory of the host vehicle to maintain a reduced inter-vehicular distance while simultaneously optimizing energy efficiency.

A model-based approach to control BMEP (Brake Mean Effective Pressure) and NOx emissions has been developed and assessed on a FPT F1C 3.0L Euro VI diesel engine for heavy-duty applications. The controller is based on a zero-dimensional real-time combustion model, which is capable of simulating the HRR (heat release rate), in-cylinder pressure, BMEP and NOx engine-out levels. The real-time combustion model has been realized by integrating and improving previously developed simulation tools. A new discretization scheme has been developed for the model equations, in order to reduce the accuracy loss when the computational step is increased. This has allowed the required computational time to be reduced to a great extent.

Direct injection (DI) of natural gas (NG) at high pressure conditions has emerged as a high-potential strategy for improving SI engine performance. Besides, DI allows an increase in the fuel economy, due to the possibility of a significant engine dethrottling at partial load. The high-pressure gas injection can also increase the turbulence level of mixture and thus the overall fuel-air mixing. Since direct NG injection is an emerging technology, there is a lack of experience on the optimum configuration of the injection system and the associated combustion chamber design. In the last few years, some numerical investigations of gas injection have been made, mainly oriented at the development of reliable numerical investigation tools. The present paper is concerned with the development and application of a numerical Star-CD based model for the investigation of the direct NG injection process from a poppet-valve injector into a bowl-piston engine combustion chamber.

This paper presents the results of part of the research activity carried out by the Politecnico di Torino and AVL List GmbH as part of the European Community InGAS Collaborative Project. The work was aimed at developing a combustion system for a mono-fuel turbocharged CNG engine, with specific focus on performance, fuel economy and emissions. A numerical and experimental analysis of the jet development and mixture formation in an optically accessible, single cylinder engine is presented in the paper. The experimental investigations were performed at the AVL laboratories by means of the planar laser-induced fluorescence technique, and revealed a cycle-to-cycle jet shape variability that depended, amongst others, on the injector characteristics and in-cylinder backpressure. Moreover, the mixing mechanism had to be optimized over a wide range of operating conditions, under both stratified lean and homogeneous stoichiometric modes.

Range anxiety in current battery electric vehicles is a challenging problem, especially for commercial vehicles with heavy payloads. Therefore, the development of electrified propulsion systems with multiple power sources, such as fuel cells, is an active area of research. Optimal speed planning and energy management, referred to as eco-driving, can substantially reduce the energy consumption of commercial vehicles, regardless of the powertrain architecture. Eco-driving controllers can leverage look-ahead route information such as road grade, speed limits, and signalized intersections to perform velocity profile smoothing, resulting in reduced energy consumption. This study presents a comprehensive analysis of the performance of an eco-driving controller for fuel cell electric trucks in a real-world scenario, considering a route from a distribution center to the associated supermarket.