Direct neural network control for a fuel-steam pressure system based on differentiator

To achieve adaptive control without modeling information for boiler steam pressure in the fuel-steam pressure system, a direct neural network controller is designed on the basis of a priviously developed sliding-mode controller based on high-gain differentiator, which does not require measurement of speed and acceleration signals. The closed system is proved to be stable by Lyapunov analysis. Simulation results indicate that good tracking performance of boiler steam pressure can be guaranteed by using neural network controller based on differentiator.