Robust Control Research On Automotive Engine Idle-Speed

As one of automotive engine's important working states, idle-speed has much relation to the engine's fuel consumption and exhaust-gas pollution level. During idle-speed state, there may be some gusty disturbance torque coming up, such as air-condition's turning on or off, drive redirector, automatic transmission, which would affect the engine speed output, even leading to flameout. According to the variety of the working environment including air pressure, temperature model parameters'change and so on, the stability of idle-speed may be influenced. This thesis does analysis and research on the robust controller to these two control problems mainly.This thesis talks about the idle-speed modeling methods and builds a mean-value model. With the software of enDYNA from German TESIS Corporation, some engine parameters are picked up and the validity of the model is proved. The disturbances which mainly come from equipments are uncertain and have limited energy. The differential geometry method finds new state variables for the system which will make the nonlinear model to be linear. The physical meaning of the new state variables is analyzed and the linear model of the idle speed model with disturbances is given.With the theory of LMI, H∞controller subject to constrained system is designed. With engine idle-speed model in enDYNA as object, PID, Backstepping and H∞controllers are compared. A H∞state feedback strategy subject to constrained system in moving horizon is developed. With the some parameters and working environment, the H∞state feedback controller in moving horizon exceeds other controllers in transition time and output constraints. Its validity in output constraint and robust performance to model parameters'uncertainty are proved, and the improvement in idle-speed stability is obvious. The simulation results show that this controller is very efficient.