skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: A comparison of three self-tuning control algorithms developed for the Bristol-Babcock controller

Technical Report ·
DOI:https://doi.org/10.2172/5287267· OSTI ID:5287267

A brief overview of adaptive control methods relating to the design of self-tuning proportional-integral-derivative (PID) controllers is given. The methods discussed include gain scheduling, self-tuning, auto-tuning, and model-reference adaptive control systems. Several process identification and parameter adjustment methods are discussed. Characteristics of the two most common types of self-tuning controllers implemented by industry (i.e., pattern recognition and process identification) are summarized. The substance of the work is a comparison of three self-tuning proportional-plus-integral (STPI) control algorithms developed to work in conjunction with the Bristol-Babcock PID control module. The STPI control algorithms are based on closed-loop cycling theory, pattern recognition theory, and model-based theory. A brief theory of operation of these three STPI control algorithms is given. Details of the process simulations developed to test the STPI algorithms are given, including an integrating process, a first-order system, a second-order system, a system with initial inverse response, and a system with variable time constant and delay. The STPI algorithms' performance with regard to both setpoint changes and load disturbances is evaluated, and their robustness is compared. The dynamic effects of process deadtime and noise are also considered. Finally, the limitations of each of the STPI algorithms is discussed, some conclusions are drawn from the performance comparisons, and a few recommendations are made. 6 refs.

Research Organization:
Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
Sponsoring Organization:
USDOE; USDOE, Washington, DC (United States)
DOE Contract Number:
AC05-84OR21400
OSTI ID:
5287267
Report Number(s):
ORNL/TM-12040; ON: DE92014472
Country of Publication:
United States
Language:
English