Vulcanization Control of Rubber Fender Based on Neural Network PID Method

Because vulcanization process of rubber fender has characteristics of nonlinearity, long time-delay, multi-variable and so on, traditional forms of proportional-integral-differential (PID) controller can not work well. This paper proposes a method which is a combinative arithmetic including BP (backpropagation), RBF (radial basis function) and PID control algorithm to tackle this problem. The relationship of proportional coefficient, integral coefficient and differential coefficient of conventional PID is usually linear, so the output of PID controller can not control nonlinear process perfectly. The combinative arithmetic is efficient to adapt the nonlinear vulcanization process. We can get nonlinear relationship of these three coefficients through the outputs of BP Neural Network which can be nonlinear. Simulation result shows that this method has got the adaptive characteristic and is feasible.

[1]  Zhang Qing Fuzzy temperature control for microwave vulcanization , 2001 .

[2]  Alan F. Murray,et al.  IEEE International Conference on Neural Networks , 1997 .

[3]  L. Autrique,et al.  Vulcanization process: a stochastic method for identification and control problem , 1994, 1994 Proceedings of IEEE International Conference on Control and Applications.

[4]  Wang Jianguo,et al.  Characteristics on energy-absorption and outer insulation of semiconductor silicon rubber , 2003, Proceedings of the 7th International Conference on Properties and Applications of Dielectric Materials (Cat. No.03CH37417).

[5]  Chen Hong-yan Human simulation intelligent control of rubber vulcanization process , 2003 .

[6]  Hui Li,et al.  Application based on neural network of the boiler optimal control in power station , 2005, 2005 International Conference on Machine Learning and Cybernetics.

[7]  A. Sideris,et al.  A multilayered neural network controller , 1988, IEEE Control Systems Magazine.