Electronic Throttle Control Based On Backstepping

Throttle is an appliance for controlling airflow in vehicle engines. Compared with conventional mechanical throttle, electronic throttle can regulates the airflow in the intake manifold more accurately and optionally. This is of great significance for many vehicle control tasks such as Idle Speed Control, Cruise Control, Anti-lock Braking Control, Vehicle Stability Control, Automatic Transmission Control. Because of its superiority, Electronic Throttle Control System is quickly coming into application widely, and that makes research on electronic throttle control a significant and exigent mission. The purpose of this thesis is to develop an effective algorithm to give a new method on electronic throttle control.The thesis mainly includes the following three parts:The first part, electronic throttle modeling. A continuous state space model of electronic throttle which has one input, duty cycle, and one output ,throttle plate position, is built in part. According to the component of electronic throttle, several submodels, including the D.C.motor model, the gear model, the limp home spring model and the throttle plate model, are built respectively. These submodels compose the whole electronic throttle model. Some other work have to be done to acquire model parameters. Firstly, the gear ratio and the throttle sensor's output need to be demarcated. Secondly, the voltage-current characteristic of the armature circuit need to be measured and be described by a graph. Thirdly, with solving the balance equations for the least squares solution, some other model parameters can be obtained. Before the model is used for controller design, the model validation should be done, and the result indicate that the model can describe the electric throttle hardware in enough specific. With all those work above, the electronic throttle model is ready for use. The second part, controller design and system performance analysis. Based on the Backstepping method and the electronic throttle model, a control law which contains feedforward and feedback is obtained. The close loop system would have global asymptotic stability on condition that there is no model error or other disturbs. But that can't be true because model error or other disturb always exist in an actual system. The input state stability theory is used with the conclusion that the close loop system have robust stability. An estimate of the boundary of system static error is given, and this is propitious to controller regulating.The third part, experiment validation. The simulation shows these facts: 1. The close loop system is stable and can meet the requirement on precision. 2. Extra experiment that contains real electronic throttle in loop is needed. Then the RCP experiment is carried out. The experiment result comes out to meet the performance requirements very well. Besides, the controller shows nice robust performance.Further research work needs to be done since some problems remain to be solved, for example, the feedforward basing on model is open loop in fact. If the feedforward is combined with adaptive control, the controller would be more practical.