Distributed PID type dynamic matrix control method for furnace pressure of coke furnace

Abstract

The invention discloses a distributed PID type dynamic matrix control method for a furnace pressure of a coke furnace. Real-time step response data are collected to establish a dynamic matrix model vector of a controlled object; an on-line optimization realization problem of a large-scale system is converted into optimization solution problems of all small-scale sub systems; and each sub system in a network environment is used as an intelligent agent and all intelligent agents complete information exchange through network communication. And then an improved performance index is established by introducing a PID operator; PID type dynamic matrix controllers of all intelligent agents are designed based on the nash optimality idea; a real-time control rule at a current time is applied to each intelligent agent, and a time domain rolls to a next time; and the optimization process is repeated, thereby completing the optimization task of the whole large-scale system. Using the method, the control quality of the system is guaranteed; defects of the traditional DDMC method is overcome effectively; and the freedom degree of the control parameter design is improved.

Description

technical field [0001] The invention belongs to the technical field of automation and relates to a distributed PID type dynamic matrix control method for coke oven furnace pressure. Background technique [0002] In the actual process, there are often a large number of complex and high-dimensional large-scale systems, and the number of variables and constraints is usually dozens or even hundreds. Therefore, the online implementation of such a large-scale predictive control problem is particularly important. . Distributed Dynamic Matrix Control (DDMC), as a main branch of Distributed Predictive Control (DMPC), comprehensively utilizes network communication technology and control theory to disperse the online solution problem of a complex large-scale system to various subsystems for distributed implementation, effectively The scale and complexity of the problem are reduced, and the controlled objects with multi-variable, strong coupling and uncertainty can be well controlled, ...

AI Technical Summary

Problems solved by technology

However, the traditional DDMC method usually has a large amount of online calculation, and the sampling period should not be too small, so that the disturbance cannot be quickly suppressed

Therefore, for control systems with large uncertainties in terms of model order, environmental disturbance, nonlinearity, etc., the conventional DDMC method is usually difficult to achieve the expected control effect

Images