Theory Researches And Software Design Of PID-Parameter Self-tuning Based On DCS Platform

PID controller is the most conventional controller in process control, and being usedin about 90 percent of the industrial process control loops, As the development of theadvanced control techniques and computer techniques, PID self-tuning is the mostapplicable control algorithm. If the PID tuning procedure merely depends on themanual operation, it would be troublesome and time-consuming. Thus the controlresults would mainly refer to the experience and knowledge of the operator self,which leads to many problems in industrial field application. Therefore, theresearches on self-tuning of PID parameters would be meaningful for both scientificresearcher and field engineer.Researches both in theory and application about PID self-tuning technique have beenconducted, and the control algorithm of first-coarse-tuning and second-fine-tuning isproposed. During the coarse-tuning, two methods based onamplitude-and-phase-margin specifying, and relay-modeling are introduced. Thesetwo methods are different in control effect, so engineers could choose based onspecific need. Moreover, PID-parameter adaptive algorithm is added in fine-tuningstage for time-varying processes. Real-time excitation is analyzed, making ajudgment of whether the excitation is large enough and the second-fine-tuningshould start. During the adaptation, recursive least-squares estimator is used, andthen the PID-parameter is obtained and updated based on pole placement technique.Based the theory researches, software is built on ECS-100 DCS system platform.Field tests and comparison with similar commercial products show this software isflexible, and suitable for industry field application.