Search Results

The paper deals with the application of two techniques for the selection of the training data set used for the identification of Neural Network black-box engine models; the research starts from previous studies on Sequential Experimental Design for regression based engine models. The implemented methodologies rely on the Active Learning approach (i.e. active selection of training data) and are oriented to drive the experiments for the Neural Network training. The methods allow to select the most significant examples leading to an improvement of model generalization with respect to a heuristic choice of the training data. The data selection is performed making use of two different formulation, originally proposed by MacKay and Cohn, based on the Shannon's Statistic Entropy and on the Mean Error Variance respectively.

A computer code oriented to S.I. engine control and powertrain simulation is presented. The model predicts engine and driveline states, taking into account the dynamics of air and fuel flows into the intake manifold and the transient response of crankshaft, clutch, transmission gearing and vehicle. The whole model is integrated in the code O.D.E.C.S., now in use at Magneti Marelli, and is based on a hierarchical structure composed of different classes of models, ranging from black-box Neural Network to grey-box mean value models. By adopting the proposed approach, a satisfactory accuracy is achieved with limited computational demand, which makes the model suitable for the optimization of engine control strategies. Furthermore, in order to simulate the driver behavior during the assigned vehicle mission profile, two drive controllers have been implemented for throttle and brakes actuation, based on classical PID and fuzzy-logic theory.

Road topography has a remarkable impact on vehicle fuel consumption for both passenger and heavy duty vehicles. In addition, erroneous or non-optimized scheduling of both velocity set-point and gear shifting may be detrimental for fuel consumption and performance. Recent technologies have made road data, such as elevation or slope, either available or measurable on board, thus making possible the exploitation of this additional information in innovative controllers. The aim of this paper is the development of a smart, fuel-economy oriented controller adapting cruising speed and engaged gear to current road load (i.e. local slope). Unlike traditional cruise controllers, the velocity set-point is not constant, but it is set by applying a mathematical transformation of the current slope, accounting for the mission time duration as well.

Due to the greater need to reduce exhaust emissions of harmful gases (GHG, NOx, PM, etc.), to promote the decarbonisation process and to overcome the drawbacks of electric vehicles (low range, high cost, impact of electricity production on CO2 emissions…), the hybrid-electric vehicles are still considered as the more feasible path through sustainable mobility. However, one of the main tasks to be accomplished to get maximum benefit from hybrid-electric powertrain is the management of the energy flows between the different power sources, namely internal combustion engine, electric machine(s) and battery pack. In this paper a methodology for the experimental testing of HEVs energy management strategies at the engine test bed is presented. The experimental set-up consists in an eddy-current dyno and a light-duty common-rail Diesel engine.

The computer code ODECS (Optimal Design of Engine Control Strategies) for the design of Spark Ignition engine control strategies is presented. This code has been developed starting from the author's activity in this field, availing of some original contributions about engine stochastic optimization and dynamical models. This code has a modular structure and is composed of a user interface for the definition, the execution and the analysis of different computations performed with 4 independent modules. These modules allow the following calculations: (i) definition of the engine mathematical model from steady-state experimental data; (ii) engine cycle test trajectory corresponding, to a vehicle transient simulation test such as ECE15 or FTP drive test schedule; (iii) evaluation of the optimal engine control maps with a steady-state approach.

In the global quest for preventing fossil fuel depletion and reducing air pollution, hybridization plays a fundamental role to achieve cleaner and more fuel-efficient automotive propulsion systems. While hybrid powertrains offer many opportunities, they also present new developmental challenges. Due to the many variants and possible architectures, development issues, such as the definition of powertrain concepts and the optimization of operating strategies, are becoming more and more important. The paper presents model-based fuel economy analyses of different hybrid vehicle configurations, depending on the position of the electric motor generator (EMG). The analyses are intended to support the design of powertrain architecture and the components sizing, depending on the driving scenario, with the aim of reducing fuel consumption and CO2 emissions.

In the last few years, several studies have proved the benefits of exploiting information about the road topography ahead of the vehicle to adapt vehicle cruising for fuel consumption reduction. Recent technologies have brought on-board more road information enabling the optimization of the driving profile for fuel economy improvement. In the present paper, a cruise controller able to lowering vehicle fuel consumption taking into account the characteristics of the road the vehicle is traveling through is presented. The velocity profile is obtained by minimizing via discrete dynamic programming the energy spent to move the vehicle. In order to further enhance vehicle fuel efficiency, also the gear shifting schedule is optimized, allowing to avoid useless gear shifts and choose the most suitable gear to match current road load and keeping the engine in its maximum efficiency range. Despite the optimality of the solution provided, dynamic programming entails high computational time.

Diesel engines are attractive thanks to good performance in terms of fuel consumption, drivability, power output and efficiency. Nevertheless in the last years, increasing restrictions have been imposed to particulate emissions, concerning both mass (PM) and number (PN). Different technologies have been proposed to meet emissions standards and the wall-flow Diesel Particulate Filter (DPF) is currently the most common after-treatment system used to trap PM from the exhaust gases. This technology exhibits good features such that it can be regenerated to remove any accumulation of PM. However, this process involves oxidation of the filtered PM at a high temperature through after and post fuel injection strategies, which results in an increase of fuel consumption and may lead to physical damages of the filter in the long term. This work deals with the experimental testing of a catalytic silicon carbide (SiC) wall flow DPF, aiming at decreasing the soot oxidation temperature.

Conventional electric vehicles adopt either single-speed transmissions or direct drive architecture in order to reduce cost, losses and mass. However, the integration of optimized multiple-speed transmissions is considered as a feasible method to enhance EVs performances, (i.e. top speed, acceleration and grade climbing), improving powertrain efficiency, saving battery energy and reducing customer costs. Perfectly in line with these objectives, this paper presents a patented fully integrated electric traction system, as scalable solution for electrifying light duty passenger and commercial vehicles (1.5-4.2 tons), with a focus on minibuses (<20 seats). The adoption of high-speed motor coupled to multiple-speed transmission offers the possibility of a relevant efficiency improvement, a 50% volume reduction with respect to a traditional transmission, superior output torque and power density.

This work aims at addressing the challenge of reconciling the surge in road transportation with the need to reduce CO2 emissions. The research particularly focuses on exploring the potential of fuel cell technology in long-distance road haulage, which is currently a major solution proposed by relevant manufacturers to get zero local emissions and an increased total payload. Specifically, a methodology is applied to enable rapid and accurate identification of techno-economically effective fuel cell hybrid heavy-duty vehicle (FCH2DV) configurations. This is possible by performing model-based co-design of FCH2DV powertrain and related control strategies. Through the algorithm, it is possible to perform parametric scenario analysis to better understand the prospects of this technology in the decarbonization path of the heavy-duty transportation sector, changing in an easy way all the parameters involved.

The last environmental regulations on passenger vehicles’ emissions harden constraints on designing powertrains. A promising solution consists in vehicle electrification leading to hybrid configurations: the tank-to-wheel pollutant emissions can be drastically reduced combining features of typical battery electric vehicles adding an Internal Combustion Engine (ICE) controlled as a Range Extender (REX). Furthermore, HC and CO/CO2 emissions can be avoided using green hydrogen as fuel for the ICE; moreover, in absence of a mechanical coupling between REX and wheels the best operating conditions in terms of maximum ICE efficiency may be easily achieved. In this work, a light quadricycle (EU L6e, classification) series hybrid vehicle with four in-wheel motors is studied for the application of a range extender system.

This article presents a novel approach for predicting fuel consumption in vehicles through a recurrent neural network (RNN) that uses only speed, acceleration, and road slope as input data. The model has been developed for real-time vehicle monitoring, route planning optimization, cost and emissions reduction and it is suitable for fleet-management purposes. To train and test the RNN, chosen after addressing several structures, experimental data have been measured on-board of a heavy-duty truck representative of a heavy-duty transportation company. Data have been acquired during typical daily missions, making use of an advanced connectivity platform, which features CANbus vehicle connection, GPS tracking, 4G/LTE - 5G connectivity, along with on-board data processing. The experimental data used for RNN train and test have been treated starting from on-board acquired raw data (e.g., speed, acceleration, fuel consumption, etc.) along with road slope downloaded from map providers.