4,3 ANFIS: hệ thống thích ứng Neuro-mờ suy luận Hình 4,29 Mặt trận bảng điều khiển của các ví dụ ANFIS bằng cách sử dụng chức năng thành viên chuông Ở bên phải của hình. 4,29 là đồ thị đầu ra ANFIS đại diện cho chức năng Trainer (sinc chức năng) và đầu ra thực tế của các ANFIS đào tạo. Lỗi đồ thị đại diện cho các giá trị lỗi tại mỗi thời đại. Như chúng ta biết, chức năng chuông trong khoảng Å'0; 1A là đại diện toán học như: 1 f x / D: (4,25) xc 2b C1 Sau. | ANFIS Adaptive Neuro-fuzzy Inference Systems 115 Fig. Front panel of the ANFIS example using bell membership functions On the right side of Fig. is the ANFIS Output graph that represents the Trainer function sinc function and the actual Output of the trained ANFIS. The Error graph represents the error values at each epoch. As we know the bell function in the range 0 1 is represented mathematically as f x 1 V f C 1 Then this VI will adapt the parameters a b and c from the above equation. The minimum error and maximum iterations are proposed in Table . In this example the VI delimits the sinc function in the interval 0 200 . Running the program we can look at the behavior of the training procedure as in Figs. . Remember to switch on the Train and Cts buttons. Example . We want to control a 12V DC motor in a fan determined by some ambient conditions. If the temperature is less than 25 C then the fan is switched off. If the temperature is greater than 35 C then the fan has to run as fast as possible. 116 4 Neuro-fuzzy Controller Theory and Application Table ANFIS example 1 MFs 5 Min E 1E-5 MaxI 10 000 Ctea Cteb Ctec Training function Sinc Fig. Initial step in the training procedure for ANFIS If the temperature is between 25 C and 35 C then the fan has to follow a logistic function description. In this way we know that the velocity of rotation in a DC motor is proportional to the voltage supplied. Then the logistic function is an approximation of the voltage that we want to have depending on the degrees of the environment. The function is described by . f x e-al 1 8a 2 R ANFIS Adaptive Neuro-fuzzy Inference Systems 117 Fig. Training procedure for ANFIS at 514 epochs A simple analysis offers that the range of the logistic function is 0 12 and for limiting the domain of the function suppose an interval 0 60 . Select a . Using the hybrid learning method train an ANFIS system selecting four .
