Mario Lavorgna

Neuro-fuzzy Networks

One of the most important research themes, in the sense of intelligent processing techniques hybridization, is the neuro-fuzzy approach. The birth of this kind of system is mostly connected with the attempt to unify the advantages of neural and fuzzy techniques using one hybrid architecture only, often referred to as fuzzy neural networks (FNN).

Fuzzy Cellular Neural Networks

In this, as in the previous chapter, we will deal with information processing systems that were originally inspired by the concepts underlying soft computing. The integration of fuzzy logic concepts in a widely spread architecture such as that of cellular neural networks in fact led to the birth of fuzzy CNNs.

GAs for fuzzy modeling of noise pollution

A growing problem in town areas is noise pollution due to the increasing number of vehicles that daily cross cities. A classical approach to model this kind of system is based on numerical regression, but its performance is not satisfactory due to the nonlinearity of the considered model. A suitable approach can be therefore to determine a fuzzy model of the system. There has been a considerable number of studies on fuzzy identification, where fuzzy implications are used to express rules, in this paper the Tagaki-Sugeno approach has been adopted applying a genetic algorithm during the optimization phase. The obtained models are compared with traditional ones showing the suitability of the proposed method.

Modelling on-off virtual lambda sensors based on multi-spread probabilistic neural networks

In this work, we have explored a novel model of learning machine which seems to be able to emulate effectively the way of functioning of the traditional on-off lambda sensors (i.e. O2 sensor). These sensors are a low cost solution used in the SI (spark ignition) engines to monitor the air-fuel ratio and so to maintain a strict control of the air-fuel mixture close the stoichiometric condition. The idea behind this work is to suggest a scheme of air/fuel control system for SI engines in which there is not need of a lambda sensor. The last is replaced by a model, named as virtual lambda sensor (VLS), trained in order to predict the air-fuel ratio values in function of features suitably selected by the in-cylinder pressure sensor signal

Use of in-Cylinder Pressure and Learning Circuits for Engine Modeling and Control

The parameter widely considered as the most important for the diagnosis of combustion process in internal combustion engines is the cylinder pressure and numerous control algorithms based on pressure measurement as a feedback signal have been proposed. Use of real-time cylinder pressure in control architectures for both SI and Diesel engines allows to replace many other sensors present in engines and offers a variety of significant advantages in terms of improved engine performances and reduced toxic emissions. The present chapter provides an overview of the main applications of cylinder pressure signal analysis in engine modeling and control.

Non-interfering Diagnostics for the Study of Thermo-Fluid Dynamic Processes

The conversion of chemical energy into mechanical power, operated by internal combustion engines, involves a great number of complex phenomena that often occur in transient thermo-fluid dynamic conditions. The majority of these phenomena are affected by nonlinear dynamics, thus requiring appropriate compensation techniques. The analysis and comprehension of these nonlinear processes is a basic requirement for the design of effective control solutions, able to optimize the combustion processes in terms of engine power, efficiency, and emissions. In this chapter, we present some advanced non-interfering optical diagnostics that allow to study in detail the reasons and the effects of the nonlinear behavior of many processes occurring in internal combustion engines.

Modeling of Particle Size Distribution at the Exhaust of Internal Combustion Engines

Nowadays, the interest in the effect of exhaust emissions from road vehicles on public health is stronger than ever. Great attention is paid to particulate matter (PM) both for its impact on the environment and for the adverse effect on human health. The internal combustion engines (ICEs) are a major source of PM emissions in the urban area. Particles are usually classified according to their diameter in coarse particles, diameter larger than 10 μm (PM10), fine particles, diameter smaller than 2.5 μm (PM2.5). The present chapter will firstly describe the characteristics of engine emitted partices and some of the mechanisms involved in their formation process. Then, it will be introduced a soft computing model, developed by the authors, devoted to the real-time prediction of particle size distribution at the exhaust of internal combustion engines on the basis of some specific inputs, such as engine speed, engine load, and amount of exhaust recirculated gases.

Identification and Compensation of Nonlinear Phenomena in Gasoline Direct Injection Process

Latest emission regulations strongly push toward a reduction of fuel consumption in order to reduce CO2 emissions. To achieve this goal, gasoline direct injection engines are one of the best candidate. GDI engines, in fact, can work in stratified operations allowing stable combustions with ultra-lean mixtures that allow a strong reduction of toxic emissions coupled towith fuel consumption reduction. GDI stratified operation needs the use of multiple fuel injections, splitting the quantity of injected fuel into several and shorter shots in order to reduce the cylinder wall impingement. However, small injections force solenoid injectors to work in their ballistic mode, with a highly nonlinear correlation between electrical command pulse width and the actual amount of injected fuel. In the present chapter, the nonlinear phenomena correlated with the injection process in GDI engines are analyzed and an effective compensation method is proposed.

Diagnosis and Control of Engine Combustion Using Vibration Signals

In other parts of this book, the importance of non-intrusive diagnostic techniques to evaluate the combustion quality in internal combustion engines has been highlighted. Among non-intrusive diagnostic techniques, those based on the analysis of the engine vibration seem to be the most promising. The present chapter proposes a method for “advanced” combustion diagnosis and control without using in-cylinder pressure transducers. The method includes a vibration signal processing in order to separate the combustion phenomena from all the other noise signatures on the signal. The correlation between the filtered block vibration signal and some combustion parameters has been also demonstrated. Finally, the chapter contains some possible combustion control strategies based on vibration signal analysis.

Artificial Intelligence for Modeling and Control of Nonlinear Phenomena in Internal Combustion Engines

Artificial intelligence techniques allow to solve highly nonlinear problems offering an alternative way to deal with complex and dynamic systems with good flexibility and generalization capability. Because of their good ability to model nonlinear phenomena together with their relatively simple application procedure, artificial intelligence systems have found an increasing usage in the modeling, diagnosis, and control of internal combustion engines. The present chapter aims to describe the use of artificial intelligence, especially Artificial Neural Networks and Fuzzy Logic techniques, in some engine applications where the inherent nonlinear nature of the process dynamics requires alternative approaches to guarantee a more accurate control action.