Multivariate Decoupling Control Based on Neural Network

Because the adoption of multivariable coupling system can not get good effect in the process of industrial production, we put forward a scheme that neural network PID decoupling control. This paper shows that how can the use of the golden section rate help the genetic algorithm to adjusting the parameters of the PID control, and how to realize the Jacobian information identification of the controlled object. through the RBF ( the network identification NNI). Finally realize the intPelligent dissolve accidentally control for the system. We also simulate the control process of a group of two variables strong coupling of time-varying system by the computer. The results show that the proposed control schemes have good effect, fast response strong adaptability and robustness.