Abstract: A new nonlinear adaptive control strategy of the hot spot temprature in an industrial fexed bed reactor is proposed.A reduced order nonlinear dynamic model is firstly developed through lumping via OCM.Then based on this reduced order model,a state estimator of the reactor is developed using an open loop observer,a new nonlinear adative control algorithm is employed to design the nonlinear controller.Finally the nonlinear adaptive control system for the hot spot temperature,which is an important median quality index of the fixed bed reactor,is designed.Simulation results show that the control system provides good track performance,strong ability to reject disturbance and significant robustness.