Nonlinear modeling and control of a spark ignition engine idle speed / Hazem Mohamed

Hazem, Mohamed (1998) Nonlinear modeling and control of a spark ignition engine idle speed / Hazem Mohamed. Masters thesis, University of Malaya.

PDF (Master Thesis)
Download (33Mb) | Preview


    The main objective of this research work is to model and control the idle mode of a four cylinder spark ignition engine. The modeling approach is to handle the engine as a multi input multi output nonlinear black box model, where the inputs are the throttle angle and the spark timing and the outp'uts are the speed and manifold pressure. The input output data was collected from an experimental four cylinder spark ignition engine, then this data is fitted to a NARMAX model using a radial basis function network trained by the orthogonal least squares algorithm. After that the developed model is validated using nonlinear correlation tests and it could efficiently reproduce the plant dynamics. Tlie control teclmique used is the fuzzy control. The fuzzy controller is formulated as a radial basis function network trained by the orthogqnalleast squares algorithm to estimate the controller parameters. The train~ng data of the network is the entries of an optimal control table derived by analyzing the evolution of the state trajectories of the system under different initial conditions using the cell to cell mapping technique. The developed controller can efficiently stabilize the plant and compensate for different uncertainties and nonlinearities of the plant.

    Item Type: Thesis (Masters)
    Additional Information: Thesis (M.Eng.)- Faculty of Engineering, University of Malaya, 1998.
    Uncontrolled Keywords: Fuzzy Controller; Electronic Control Unit; Radial Basis Functions Networks
    Subjects: T Technology > TJ Mechanical engineering and machinery
    Divisions: Faculty of Engineering
    Depositing User: Mr Prabhakaran Balachandran
    Date Deposited: 26 Aug 2017 14:59
    Last Modified: 26 Aug 2017 14:59

    Actions (For repository staff only : Login required)

    View Item