Modeling and Control of a Compressed Natural Gas Injection System

The application of new technologies in the automotive field has highly improved the performances of internal combustion engines. In particular, applying electronic control to the fuel injection system has improved dynamic performances and reduced noxious emissions, noise and fuel consumption. To achieve this goal it is necessary to exactly meter the injected fuel by controlling both the injection pressure and the injectors opening time. On the other hand, to develop a suitable control strategy for the injection system, it is necessary a mathematical representation, describing the dynamics of most significant variables. In this paper we propose a fast and simple model of a Compressed Natural Gas injection system to predict the system dynamics with a good accuracy for different operating conditions. We also present a comparison between simulation and experimental data to validate the results. Then we use this model to design a fuzzy logic controller of the injection pressure. Simulation experiments show that the fuzzy logic controller can cope with system nonlinearities, guaranteeing better performances than classical PI controllers. Key-Words: Internal combustion engines, CNG, Injection systems, Fuzzy control.