Multivariable system two-stage online identification method and system

Abstract

The invention discloses a two-stage online identification method and system for a multivariable system. The method comprises the following steps: acquiring continuous historical data of each controlled variable and each controlled variable in the multivariable system; constructing a data sample by adopting a CARMA model; obtaining the optimal pure lag time from each controlled variable to each controlled variable in the multivariable system, and establishing an optimal pure lag CARMA model; obtaining a step response curve of each controlled variable when each controlled variable is subjected to step change on the basis of an optimal pure lag CARMA model; acquiring continuous change data of the step response curve as an identification sample, acquiring the optimal pure lag time from each controlled variable to any controlled variable in the single-variable system based on a second-order pure lag CARMA model, and establishing a second-order optimal pure lag CARMA model; and performing matrix combination on the second-order optimal pure lag CARMA model to obtain a multivariable system identification model. According to the method, the multivariable model is decoupled and simplified while the model precision is ensured, and prediction control can be realized.

Description

technical field [0001] The invention belongs to the technical field of advanced process control, and more particularly, relates to a two-stage online identification method and system for a multivariable system. Background technique [0002] In the field of advanced process control, system identification is one of the most active and rapidly developing topics. Like optimal control and adaptive control, system identification is one of the main pillars of modern control theory. The purpose of system identification is to fully grasp the motion law of the research object, and to describe the causal relationship of the motion of things in mathematical language. System identification is modeling, that is to establish an optimal model of a system according to the input-output data of the actual system. This model can be used for adaptive control, model predictive control, etc., and is the basis and foundation of control simulation. [0003] In the prior art, the least-squares algor...

AI Technical Summary

Problems solved by technology

[0004] However, in the existing actual control system, the controlled object is usually a multivariable system with coupling characteristics. In order to improve the identification accuracy during model identification, it is necessary to increase the order, but when the model order is increased, it is easy to lead to overfitting And the higher the order is, the more roots the function has. When it is displayed on the curve, the high-order model is easy to fluctuate with the change of time domain. When it is discretized, the curve is discontinuous and not smooth, which affects the control accuracy.

In addition, the high-order model increases the amount of calculation. For predictive control, the increase of the model order will greatly increase the calculation workload and affect the control speed.

Images