Air-Fuel Ratio Control Based On State Observer

Automobile is becoming more and more important in transportation nowadays, and people get more reliant on it. As the heart of vehicle, engine's performance would directly affect vehicle's performance on power, economy and environmental protection. To achieve low-emission, low power consumption and high impetus performance at the same time, we need not only to control the injection volume and ignition time, but also to control the air-fuel ratio precisely by adjusting the temperature and suction pressure. It is the precise control on air-fuel ratio that is the key aspect of improving the performance of gasoline engine. An awful control on air-fuel ratio will lead to ill-equipped gasoline engine power and the poor economic performance and also increase the emission of harmful gas. In terms of the existing main request to the engine performance or the introduction of new technologies, the modeling and control of the air-fuel ratio seem to be extremely important.In this dissertation, the real-time simulation software of engine, enDYNA, which is a production of Tesis in Germany, is used as a virtual engine to get parameters and operation data of engine. And then based on the mean value theory, the model of gasoline engine is carefully established, which coefficients are fitted by using the data from the virtual engine model—enDYNA. Finally, a controller based on state observer is formulated, and the controller is proved effective not only to the average model but also to enDYNA model. The main tasks completed are as follows.1. The modeling of the air-fuel ratio of gasoline engineBased on the mean value theory, the air-fuel ratio of gasoline engine system is modeled. The air-fuel ratio model consists of the intake air channel dynamics model and the fuel channel dynamics model. A non-linear model of the air-fuel ratio is set up using Matlab/simulink library. With real-time simulation software of engine, enDYNA, treated as a virtual engine, from which the model parameters and experimental parameters are collected, the parameters based on the mean value model are fitted. Finally in the same working condition and the same input, using of Matlab/simulink library set up the air-fuel ratio model based on the mean value theory is check out by a more precise engine model from enDYNA to verify its validity.2. The design of an air-fuel ratio controller based on the state observerBased on the analysis of the mechanism the engine follows and the model established,a model-based control strategy of the air-fuel ratio is proposed, namely: when the engine works on the steady-state condition, a closed-loop state observer on intake volume is built through the wide domain air-fuel ratio oxygen sensor to achieve closed-loop predictive control of the air-fuel ratio and meet the necessary accuracy; when the engine works on the transient-state condition, a state observer on intake volume is built through the pressure sensor, combined with the dynamic compensator of the fuel, to achieve the open-loop control of the air-fuel ratio to meet the requirements of real-time control on the transient-state condition. Both of the steady-state and the transient-state observers are built, and also the flow of two schemes.3. The simulation of the rapid control prototypingThe Kalman filter theory, as well as the common derivation of Kalman filter algorithm is introduced, and aiming at the divergence problem in the Kalman filter a common solution is recommended.The experimental simulation platform which is built using the experimental equipment, dSPACE and the real-time simulation system for engine, xPC Target is briefly introduced first. Aiming at the air-fuel ratio controller based on the state observer above-mentioned, the air-fuel ratio model based on the mean value theory for gasoline engine and the accurate gasoline engine model from enDYNA are all verified by the off-line test and rapid control prototyping experiment respectively. The experiments can be the further evidence which proves the control algorithm proposed is effective and useful in practical applications.In this paper, the air-fuel ratio controller is designed based on a certain working condition on which the engine speed is a constant value and the parameters of the dynamic compensator are also set to be constant. In order to strengthen the control ability on the transient-state condition, the next goal of our group is to design another air-fuel ratio controller considering the variable engine speed, with parameters of dynamic compensator adding to the algorithm in the form of MAP chart to improve the application scope of the algorithm and serve the practical application better.