Search Results

Automotive engines and control systems are more and more sophisticated due to increasingly restrictive environmental regulations. Particularly in both diesel and SI lean-burn engines NOx emissions are the key pollutants to deal with and sophisticated Engine Management System (EMS) strategies and after-treatment devices have to be applied. In this context, the in-cylinder oxygen mass fraction plays a key-role due its direct influence on the NOx formation mechanism. Real-time estimation of the intake O₂ charge enhances the NOx prediction during engine transients, suitable for both dynamic adjustments of EMS strategies and management of aftertreatment devices. The paper focuses on the development and experimental validation of a real-time estimator of O₂ concentration in the intake manifold of an automotive common-rail diesel engine, equipped with turbocharger and EGR system.

The paper focuses on the experimental identification and validation of recurrent neural networks (RNN) for virtual sensing of NO emissions in internal combustion engines (ICE). Suited training procedures and experimental tests are proposed to improve RNN precision and generalization in predicting NO formation dynamics. The reference Spark Ignition (SI) engine was tested by means of an integrated system of hardware and software tools for engine test automation and control strategies prototyping. A fast response analyzer was used to measure NO emissions at the exhaust valve. The accuracy of the developed RNN model is assessed by comparing simulated and experimental trajectories for a wide range of operating scenarios. The results evidence that RNN-based virtual NO sensor will offer significant opportunities for implementing on-board feedforward and feedback control strategies aimed at improving the performance of after-treatment devices.

The paper focuses on the experimental identification and validation of recurrent neural networks (RNN) for real-time prediction and control of air-fuel ratio (AFR) in spark-ignited engines. Suited training procedures and experimental tests are proposed to improve RNN precision and generalization in predicting both forward and inverse AFR dynamics for a wide range of operating scenarios. The reference engine has been tested by means of an integrated system of hardware and software tools for engine test automation and control strategies prototyping. The comparison between RNNs simulation and experimental trajectories showed the high accuracy and generalization capabilities guaranteed by RNNs in reproducing forward and inverse AFR dynamics. Then, a fast and easy-to-handle procedure was set-up to verify the potentialities of the inverse RNN to perform feed-forward control of AFR.

This paper deals with modeling and analysis of the integration of ThermoElectric generators (TEG) into a conventional vehicle, specifically aimed at recovering waste heat from exhaust gases. The model is based on existing and commercial thermoelectric materials, specifically Bi2Te3, having ZTs not exceeding 1 and efficiency below 5%, but a trade-off between cost and performance that would be acceptable for automotive applications. TEGs operate on the principle of thermoelectric energy conversion via Seebeck effect, utilizing thermal gradients to generate electric current, with exhaust gases at the hot side and coolant at the cold side. In the simulated configuration the TEG converters are interfaced with the battery/alternator supporting the operation of the vehicle, reducing the energy consumption due to electrical accessories and HVAC.

The paper deals with the use of Recurrent Neural Networks (RNNs) for the Air-Fuel Ratio (AFR) control in Spark Ignition (SI) engines. Because of their features, Neural Networks can perform an adaptive control more efficiently than classical techniques. In the paper, a review of the most useful control schemes based on Neural Networks is presented and the potential use in the field of engine control is analyzed. A preliminary controller has been implemented making use of a Direct Inverse Modeling approach. The controller compensates for the wall wetting dynamics and estimates the right amount of fuel to be injected to meet the target AFR during engine transients. The Direct Inverse Controller has been tested within an engine/vehicle simulator. The simulation tests have been performed by imposing a set of throttle transients at different engine speeds. The results show that the Inverse Model can satisfactorily bound the AFR excursions around the target value.

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.

The measurement of the various pollutant species (HC, CO, NO, etc.) has become one of the main issues in internal combustion engine research. This interest concerns not only their quantitative measurement but also the study of the mechanism of their formation. In fact, pollutant species concentration can be used as an indicator for the combustion characteristics. For instance, it enables the determination of a lean or a rich combustion, the percentage of EGR, etc. The purpose of this research is to investigate the behavior of pollutant gas particles in the first part of an engine exhaust system through a detailed study of the unsteady flow in the exhaust pipe. The results are intended to designate the appropriate sensor positions which ensure accurate measurement results. This investigation wants to track an inert component in the exhaust system, namely the NO gas.

An integrated system of phenomenological models is applied in conjunction with identification techniques to simulate SI engine performance and emissions. In the framework of a hierarchical model architecture, the model structure provides the steady state engine data required for the design and validation of synthetic engine models. This approach allows limiting the recourse to the experimental data and speeds up the engine control strategies prototyping. The model structure is composed of a multi-zone thermodynamic engine model linked to a 1-D commercial fluid-dynamic model for intake and exhaust gas flow and to a physical model for NOx exhaust emissions. In order to improve model accuracy and generalization, an identification methodology is applied to estimate the optimal parameters for the turbulent combustion model. Due to the built-in physical content, the proposed methodology requires a relatively limited amount of experimental data for characterizing the under-study engine.

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.

The development of more affordable sensors together with the enhancement of computation features in current Engine Management Systems (EMS), makes the in-cylinder pressure sensing a suitable methodology for the on-board engine control and diagnosis. Since the 1960’s the in-cylinder pressure signal was employed to investigate the combustion process of the internal combustion engines for research purposes. Currently, the sensors cost reduction in addition to the need to comply with the strict emissions legislation has promoted a large-scale diffusion on production engines equipment. The in-cylinder pressure signal offers the opportunity to estimate with high dynamic response almost all the variables of interest for an effective engine combustion control even in case of non-conventional combustion processes (e.g. PCCI, HCCI, LTC).

The paper deals with the steady-state optimal tuning of control variables for an automotive turbocharged Diesel engine. The optimization analysis is based on an engine simulation model, composed of a control oriented model of turbocharger integrated with a predictive multi-zone combustion model, which allows accounting for the impact of control variables on engine performance, NOx and soot emissions and turbine outlet temperature. This latter strongly affects conversion efficiency of after treatment devices therefore its estimation is of great interest for both control and simulation of tailpipe emissions. The proposed modeling structure is aimed to support the engine control design for common-rail turbocharged Diesel engines with multiple injections, where the large number of control parameters requires a large experimental tuning effort.

The paper describes suited methodologies for developing Recurrent Neural Networks (RNN) aimed at estimating NOx emissions at the exhaust of automotive Diesel engines. The proposed methodologies particularly aim at meeting the conflicting needs of feasible on-board implementation of advanced virtual sensors, such as neural network, and satisfactory prediction accuracy. Suited identification procedures and experimental tests were developed to improve RNN precision and generalization in predicting engine NOx emissions during transient operation. NOx measurements were accomplished by a fast response analyzer on a production automotive Diesel engine at the test bench. Proper post-processing of available experiments was performed to provide the identification procedure with the most exhaustive information content. The comparison between experimental results and predicted NOx values on several engine transients, exhibits high level of accuracy.

The paper deals with the development of a multi-zone phenomenological model for the combustion process in a common rail multi-injection Diesel engine. The model simulates the fuel jet and its interaction with surrounding gases by dividing the jet core into many parcels in order to describe the thermal gradient and the chemical composition within the combustion chamber. This is mandatory for the simulation of the NO pollutant formation, carried out via the Zeldovich mechanism. The air entrainment into the fuel jet is modeled by means of the momentum balance applied to each zone and to the air zone. The stratification of the chemical composition within the cylinder and the details of the spray and its interaction with the air zone are simulated to estimate the spray penetration and speed, the mass of entrained air and the equivalence ratio in each zone. The combustion model is based on the laminar-and-turbulent characteristic-time approach.

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 paper addresses the potentialities of Recurrent Neural Networks (RNN) for modeling and controlling Air-Fuel Ratio (AFR) excursions in Spark Ignited (SI) engines. Based on the indications provided by previous studies devoted to the definition of optimal training procedures, an RNN forward model has been identified and tested on a real system. The experiments have been conducted by altering the mapped injection time randomly, thus making the effect of fuel injection on AFR dynamics independent of the other operating variables, namely manifold pressure and engine speed. The reference engine has been tested by means of an integrated system of hardware and software tools for engine test automation and control strategies prototyping. The developed forward model has been used to generate a reference AFR signal to train another RNN model aimed at simulating the inverse AFR dynamics by evaluating the fuel injection time as function of AFR, manifold pressure and engine speed.

The paper deals with the development of a system of phenomenological models for the simulation of combustion and NOx-Soot emissions in Common-Rail Multi-Jet Diesel engines. The system has been built by following a modular modeling approach and is suitable for the implementation in the framework of Hardware In the Loop (HIL) ECU rapid prototyping. A single-zone model simulates the ignition delay and the combustion during a sequence of pilot, pre and main fuel injections for a production 1,9 liters Diesel engine equipped with High Pressure Injection system, electronically controlled. The heat release model is based on the synthetic description of both premixed and diffusive combustion. The Zeldovich mechanism has been used to simulate the formation of NO emissions while the Soot model is based on the approach proposed by Hiroyasu. The models have been tested vs. a wide set of experimental data with a good accuracy in predicting pressure cycle and heat release.

A two zones combustion model suitable for the engine control design of common rail multi-jet Diesel engines is presented. The modeling approach is based on a semi-empirical two-zone combustion model coupled with identification analysis in order to implement a predictive tool for simulating the effects of control injection strategies on combustion and exhaust emissions. Fuel jet formation and combustion for both premixed and diffusive regimes are predicted, by dividing the combustion chamber into two control volumes; these account for the fuel jet and the surrounding air, composed by fresh air and residual gases; the fuel jet is divided into two zones to separate liquid and vapor phases. The simulation results have shown that the model predicts the effects of different injection parameters in case of single and multiple injection in a short computational time, suitable for the accomplishment of intensive simulations or optimization analyses over generic engine driving cycles.

In the last decades, NOx emissions legislations for Diesel engines are becoming more stringent than ever before and the selective catalytic reduction (SCR) is considered as the most suitable technology to comply with the upcoming constraints. Model-based control strategies are promising to meet the dual objective of maximizing NOx reduction and minimizing NH3 slip in urea-selective catalytic reduction. In this paper, a control oriented model of a Cu-zeolite urea-SCR system for automotive diesel engines is presented. The model is derived from a quasi-dimensional four-state model of the urea-SCR plant. To make it suitable for the real-time urea-SCR management, a reduced order one-state model has been developed, with the aim of capturing the essential behavior of the system with a low computational burden. Particularly, the model allows estimating the NH3 slip that is fundamental not only to minimize urea consumption but also to reduce this unregulated emission.

Main aim of this paper was to exploit the well-known Willans line method in a twofold manner: indeed, beyond the usual identification of Willans line parameters to enable internal combustion engine scaling, it is also proposed to infer further information from identified parameters and correlations, particularly aiming at characterizing mechanical and frictional losses of different engine technologies. The above objectives were pursued relying on extended experimental performance data, which were gathered on different engine families, including turbo-charged Diesel and naturally aspirated gasoline engines. The matching between Willans line scaled performance and experimental ones was extensively tested, thus allowing to reliably proceed to the subsequent objective of characterizing mechanical losses on the basis of identified Willans parameters.

A computer code oriented to S.I. engine control and powertrain simulation is presented. The model, developed in Matlab-Simulink® environment, 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, transmission gearing and vehicle. The model, derived from the code O.D.E.C.S. for the optimal design of engine control strategies now in use at Magneti Marelli, is suitable both for simulation analysis and to achieve optimal engine control strategies for minimum consumption with constraints on exhaust emissions and driveability via mathematical programming techniques. The model is structured as an object oriented modular framework and has been tested for simulating powertrain system and control performance with respect to any given transient and control strategy.