Simple method for neural network decoupling of multi-variable system based on model reference adaptive control

Abstract

The invention discloses a shortcut method of nerve network decoupling based on the adaptive control multiple-variable system in the complex system intelligent modeling and decoupling control technique domain, which is characterized by the following: adapting on-line identification and decoupling of nerve network; guiding the other two-phase control signal into the input end of nerve network identifier as two signal reference and identification when certain one phase of the three-phase electrode system is on-line identifying; decoupling and controlling the three-phase electrode.

Description

technical field [0001] The invention belongs to the technical field of complex system intelligent modeling and decoupling control, and in particular relates to a simple method for decoupling a neural network of a multivariable system based on model reference adaptive control. Background technique [0002] The control of multi-variable complex process control systems faces two major difficulties: one is that it is difficult to use traditional theories and methods to model the system in real time and online because of the complexity and serious uncertainty of the system; Deterministic, it is difficult to decouple it with decoupling theory based on analytical models. Although the research papers of many scholars have conducted in-depth and systematic research and discussion on the decoupling control of multivariable process control systems, these theories are all based on a premise: that is, it is necessary to write an accurate analytical model of the multivariable process syst...

AI Technical Summary

Problems solved by technology

[0002] The control of multi-variable complex process control systems faces two major difficulties: one is that it is difficult to use traditional theories and methods to model the system in real time and online because of the complexity and serious uncertainty of the system; Deterministic, it is difficult to decouple it with analytical model-based decoupling theory

Although the research papers of many scholars have conducted in-depth and systematic research and discussion on the decoupling control of multivariable process control systems, these theories are all based on a premise: that is, it is necessary to write an accurate analytical model of the multivariable process system, and this It is almost impossible to do this in reality, especially for the process control system that changes from time to time with the operating conditions, the system structure and parameters are seriously uncertain, nonlinear, hysteresis, many disturbances and other factors, it is impossible to write its analytical model

Images