In power plant control system, the capability to achieve an optimal
tracking property of the nonlinear multi-input multi-output (MIMO) units
has been an important task. This paper proposes a direct adaptive wavelet
neural network controller of boiler-turbine system for improving the
performance and efficiently achieving the good tracking property to meet
the load demands under load changes, large disturbances and change of
system operating points. This paper describes the application of a multiloop < br />direct adaptive wavelet neural network for a drum boiler; three
important outputs were controlled using a direct adaptive controller.
WNNs are rapidly trained with adaptive learning rates (ALRs) which have
been derived from the discrete Lyapunov stability theorem and used to
guarantee the convergence of the WNN controllers. Simulation results
show that the robustness and the good performance of the proposed
control system to satisfy stable tracking of the boiler-turbine system.