Fuzzy logic supervisory control for steam turbine prewarming automation

Abstract

The process of warming a large steam turbine after an extended period of maintenance or inactivity is called "prewarming." Currently, many large steam turbines are prewarmed manually in an otherwise completely automated startup procedure. An approach is outlined for automating the prewarming of large steam turbines using fuzzy logic. In this approach, heuristics employed by a human operator during prewarming are captured in a fuzzy rule base and used by a supervisory controller.

Steam Turbine Rotor Stress Control through Nonlinear Model Predictive Control

“…In more recent years, these operations have been automated, thanks to a wider use of sensors, which have led to in-depth studies of their behavior through increasingly reliable models. An example of automated pre-warm is described by Badami et al [5], in which a fuzzy logic-based supervisor has been designed to speed up the rotor warming phase, while taking into account the rotor stress conditions before the cold start of the machine.…”

Fuzzy supervisory controller tuning and analysis for optimal steam distribution in industry

“…For realization a Fuzzy supervisory control scheme was suggested, that would enable reduced number of operator's control actions, less interventions and stabilization of steam distribution. [5], [6] The development and set up of the fuzzy control algorithm were realized offline hence the production cycle wasn't disturbed. The task demanded a process study, identification of key process variables and their correlations, so the behaviour (model) of steam production was developed.…”