Abstract:Closed-loop identification might be necessary because the system is unstable in open loop or the system contains inherent feedback mechanisms, or the system will drift from the-nominal operating point. How to evaluate identification model has been a challenging problem of multivariable closed-loop identification. Firstly, a simulator was developed in the paper, which could be used for closed-loop simulation and identification for paper-making multivariable control. Then, the paper proposed a novel criterion to evaluate the identified model, which was called foot mean square error (FMSE), and defined as: FMSE = norm(y-(y) over cap)/length((y) over cap) *100. Here, norm function-gives the norm of the vector; length function gets the length of the vector. Identification simulation tests were carried out by simulations using the simulator. The duration of sampling was 50,000 seconds and the sampling interval was 5 seconds. Outputs were disturbed by uncorrelated white noises. To examine the influence of the disturbance, three disturbance levels were tested: the signal-to-noise ratios (N2S) of 10/3 (high disturbance), 10/1 (medium disturbance) and 100/3 (low disturbance) were used in simulations. Those tests suggest that the evaluation criterion would be an effective method for closed-loop identification of paper-making multivarlable control.

Technical Communique: Optimal closed-loop identification test design for internal model control

Some Study on Identification of Ill-Conditioned Processes for Control

Simple Control-Relevant Identification Test Methods for A Class of Ill-Conditioned Processes

Closed-loop Identification Based PID Tuning Without External Excitation

MPC Closed-Loop Identification Without Excitation

Mimo Process Identification for Controller-Design - Test Signals, Nominal Model and Error-Bounds

Robust PID Tuning Using Closed-Loop Identification

Multi-loop Adaptive Internal Model Control Based on a Dynamic Partial Least Squares Model

Dynamics and Control Designs for Internal Thermally Coupled Distillation Columns with Different Purities, Part 2: Close Loop Analysis and Control Designs

Control-Relevant Identification Test Design for Open-Loop Experiment

Case Studies on Closed-Loop Identification for MPC

Design of Identification Experiment Signal for Minimum Variance Control

Model Optimization of Servo System Based on Internal Model Control

Internal Model Control Design for Transition Control

Single-Loop Controller Design Via Imc Principles

New Insight into Internal Model Control Filter Design for Load Disturbance Rejection

IMC-PID design based on model matching approach and closed-loop shaping

Excitation Signal Design for Closed-Loop System Identification

A Class of Internal Model Controller Design for Time Delay Systems

A Novel Evaluation Method Of Closed-Loop Identification For Paper-Making Multivariable Control

Method of Closed-Loop Identification Based on Frequency Domain for Two-Input and Two-Output (TITO) Systems