Affiliations: Department of Computer Science and Information Engineering, Taoyuan Innovation Institute of Technology, Chung-Li, Taoyuan, Taiwan, R. O. C. | Department of Electrical Engineering, Yuan Ze University, Chung-Li, Taoyuan, Taiwan, R. O. C.
Note: [] Corresponding author. Yi-Jen Mon, Department of Computer Science and Information Engineering, Taoyuan Innovation Institute of Technology, Chung-Li, Taoyuan, 320 Taiwan, R. O. C. Tel.: +86 3 4361070 6937; Fax: +86 4361070 6999; E-mail: monbuy@tiit.edu.tw
Abstract: An intelligent control methodology for a long-term ecological systems is developed in this paper. This intelligent control methodology is called as robust recurrent fuzzy neural network control (RRFNNC). This control methodology is used to deal with multi-biomass ecological system which is an uncertain nonlinear system subject to unpredictable but bounded disturbances. This RRFNNC system is comprised of a recurrent fuzzy neural network (RFNN) controller and a robust controller. The RFNN controller is used to approximate an ideal controller; and the robust controller is designed to compensate for the approximation error between the RFNN controller and the ideal controller. The proposed RRFNNC system is applied to keep the multi-biomasses of ecological system within a stay small neighborhood of the unique nontrivial optimal equilibrium state of the undisturbed exploited ecosystem. For the simulation results of accumulative yield of harvest, more harvest can be obtained by applying the proposed RRFNNC system when compared with state feedback control.
