Search Results

Increasing demands on emissions reduction and efficiency encouraged a progressive introduction of cleaner combustion concepts. "Advanced" diesel combustions offer a high potential for simultaneous reduction of both NOx and soot within the engine through high inlet charge dilution and mixture homogenization. However, the potential benefits of these combustions in terms of emissions are counterbalanced by their high sensitivity to in-cylinder thermodynamic conditions. This sensitivity makes the engines require closed loop combustion control with real-time information about combustion quality. The parameter widely considered as the most important for the evaluation of the combustion quality in internal combustion engines is the cylinder pressure. However, this kind of measure involves an intrusive approach to the cylinder, expensive sensors and a special mounting process.

The present study proposes the use of a MLP neural network to model the relationship between the engine crankshaft speed and parameters derived from the in-cylinder pressure cycle. This allows to have an indirect measure of cylinder pressure permitting a real time evaluation of combustion quality. The structure of the model and the training procedure is outlined in the paper. The application of the model is demonstrated on a single-cylinder engine with data from a wide range of speed and load. Results confirm that a good estimation of combustion pressure parameters can be obtained by means of a suitable processing of crankshaft speed signal.

Electronic engine controls based on non-intrusive diagnostics can significantly help in complying with the stricter and stricter regulations on pollutants emissions and fuel consumption. The aim of this paper is the use of a low-cost linear capacitive accelerometer placed on the engine block for non-intrusive diagnosis of combustion process in spark ignition engines. In particular, good correspondences between the engine block vibrations and the combustion pressure signal were obtained. The angular position of pressure peak evaluated by accelerometer data can be used in a closed-loop control system for real time control of spark advance.

In this work, we propose an innovative coil driver kit for new generation ABS/ESC control unit. It is composed by STMicroelectronics L9374-L9375 coil drivers and a VCCL (Virtual Current Control Loop) implemented on the micro-controller of the control unit that allows the L9374-L9375 to work as having 6 current regulated channels. This SW library, implementing the virtual current control loop, has been tested in different working conditions of the coil drivers. Basically, the results show a satisfactory behavior of the virtual current control loop in all the test conditions taken into account. It comes out an accuracy of the virtual current control loop that is similar the nominal accuracy of the 2 current regulated channels of the L9374.

In this work, an adaptive control strategy for electro-mechanical brakes based on an optimized generalized predictive control is proposed. The control strategy is so defined. A parametric identification of the system transfer function is real-time implemented in order to provide the controller with the needful parameters for the control action. Moreover, an off-line optimization of design parameters (e.g. control horizon, prediction horizon, weights associated to different parts of the J functional, etc.) driven by stochastic searching algorithms (such as Evolution Strategies, Particle Swarm Optimization Algorithm, Differential Evolutions, etc.) is used to improve the control system robustness when system deviations from nominal operating conditions occur. Our approach has been validated on ABS (Anti-lock Brake System) braking patterns which are considered the most stressing patterns for electro-mechanical brakes.

In the feedforward part of SI engine Air/Fuel control system, the in-cylinder mass air flow rate has to be accurately estimated in order to determine the fuel amount to be injected. Generally, this evaluation is performed either with a dedicated sensor (MAF sensor) or with an indirect evaluation based on the speed-density method. In this paper we propose a soft computing mass air flow estimator for a single-cylinder gasoline engine which is able to estimate, by using the combustion pressure signal, the incoming mass air flow both in steady states and in transient conditions.

The aim of this study is to realize a virtual combustion sensor, that is, a “grey-box” model able to forecast the rate of heat release (ROHR) in a common rail diesel engine, supplied by a modulated injection rate. The model has the following inputs: engine speed, injection pressure, environment conditions and control parameters of the fuel split injection. The idea behind model development is to research ROHR's discriminating features on a Wiebe functions basis using evolutionary algorithms. After this we used a clustering algorithm to find the optimal data set with which we trained the neural network which represents our “grey-box” model. The ROHR model could be used as a virtual combustion sensor in a model based control system for the real-time updating of control parameters. Moreover, it can be used to develop hardware-in-the-loop diesel engine simulation systems. The ROHR model is global, portable, and multi-resolution.