A double variable decoupled controller based on a BP network PID and the research on its application to an air-cushioned headbox

Abstract

The coupling relationship existing between total pressure and liquid level of an air-cushioned headbox in the paper making industry is both complex and nonlinear. This paper analyzes and summarizes the advantages of a traditional BP neural network and PID control methods. A new controller based on BP neural network PID is designed to reduce the influence between the two variables. The controller modifies the parameters: proportion (k p ), integral (k i ), and differential coefficient (k d ) of a PID itself on line using the neural network's capability for non-linear description, in order to find the best PID parameters, which can reduce the influence between total pressure and liquid level of an aircushioned headbox. The results of simulations using linear and nonlinear models indicate that the controller has the characteristics of simple realization, quick dynamic behavior, high control and practical value.