Search Results

Spark-ignited engines equipped by a three-way catalyst require a precise control of the air fuel ratio fed to the combustion chamber. A stoichiometric mixture is necessary for the proper working of the catalyst in order to meet the legislation requirement. A critical part of the air fuel ratio control is the feed-forward compensation of the fuel dynamics. Conventional strategies are based on a simplified model of the wall-wetting phenomena whose parameters are stored in off-line computed look-up tables. Unfortunately, errors in the parameters calibration over the whole engine map deteriorate the control performances in terms of emissions. In this paper an automatic procedure for a rapid and efficient identification of the wall-wetting parameters is presented. The whole procedure has been experimentally tested on a vehicle by using a test bench.

Based on a state space dynamic model of the powertrain system, a new control technique for the dry clutch engagement process is proposed. The feedback controller is designed following a linear quadratic approach by using the crankshaft speed and the clutch disk speed as state variables. The controller guarantees fast engagement, minimum slipping losses and comfortable lock-up. The critical standing start operating conditions are considered and a comparison with a classical open loop control strategy is presented. The numerical results, carried out by a Simulink/Stateflow simulation scheme and a realistic set of parameters, show the good performance of the closed loop system.

A key point in three-way catalytic converter modeling problems is the definition of a possible chemical scheme able to represent the catalyzed process inside the converter, especially during transients. The lack of precise kinetic measurements during the transient thermal phase makes hard the choice of the kinetic expressions and, overall, of the chemical parameter values. To solve this problem here we propose the use of neural networks (NN) to model the reaction kinetics since a NN structure can provide enough degrees of freedom to capture all the significant features of the real system. Since the NN is embedded into the overall TWC dynamics, it cannot be trained through one of the standard method and some difficulties arise when dealing with the parameter tuning of this model, that are circumvented using a genetic algorithm (GA).