Decoupling Control for Non - Model MIMO System by Decentralized Single Neural Controllers

Based on the principle of decoupling and neural-network, an adaptive neural non-model decoupling method control is put forward in this paper. Single neural controller (SNC) is adopted as the decentralized controller. With a simple parameter tuning algorithm, the SNC in each loop is able to learn to control a variable process by merely observing the process output error in the same loop. The learning-type method to training neuron and its decoupling mechanism in multi-variable are analyzed by taking a two-input and two-output system for example. The result of the simulation shows that multi-input and multi-output (MIMO) system is difficult to achieve high performance and the system is instable sometimes, because of loops interactions and modeling errors. The decentralized nonlinear control strategy presented in this paper is simple and promising for interacting multivariable process control