NONLINEAR CONTROL USING PARAMETER ESTIMATION FROM FORWARD NEURAL MODEL
Anna Jadlovská
A neural network application for system identification and control of
nonlinear process is described in this paper. The nonlinear identification is
mostly using feedforward neural networks as a useful mathematical tool to build
a nonparametric model between the input and output of a real nonlinear process.
The possibility of online estimation of the actual parameters from an offline
trained neural model of the nonlinear process using the gain matrix is considered
in this paper. This linearization technique is used in the algorithm of online
tuning of the controller parameters based on a pole placement control design for
the nonlinear SISO process.
Keywords: neural model, NARMAX structure, nonlinear parameter estimation
