Abstract: This paper puts forward a kind of feed-forward combination controller for a kind of nonlinear large-lag objects in industry.The feed-forward controller can optimize the estimated value of parameters and get the static inverse model of the controlled object.Identification of inverse model is realized by a neural network that is separated from that which is used for inverse control,therefore they can adjust weight value of the network online.The basic fuzzy controller can be switched between fuzzy PID control algorithm and fuzzy PD control algorithm so that it can be adapted to all kinds of active states of the system.The control strategy simulates human intelligence and can adjust combination of control algorithm according to various stages of the output error.It can self-optimize too.A simulation case shows that the controller is effective.