Design and Analysis of Boiler-Turbine-Generator Controls Using Optimal Linear Regulator Theory

The demand for improved dynamic response of fossilfired power plants has motivated a comprehensive program of control system design and analysis. Previous papers have reported the development of a nonlinear mathematical model of a drum-type, twin furnace, reheat boiler-turbine-generator (RBTG) system which is suitable for control system analysis and has been extensively verified by field test. On the basis of this model, local stability, observability, and controllability have been examined over the load range, using linearization and modal analysis. An approach to control system design has been developed based on optimal linear regulator theory and which recognizes the limitation of an imperfect model. This approach produces “integral-type” action which guarantees zero steady-state errors. The controller does not require complete state feedback. Improved performance has been demonstrated by comparison with the existing control structure through simulation using the nonlinear process model. T I. IKTRODUCTION HE CONTINUISG increase in demand for electric power, una.nticipa.t.ed delays in new generat,ing capacity addit,ions, and the trend toward larger generahg stations and larger interconnections are among the many factors which have magnified the importance of individual unit response capability to the pon-er system operating objective of providing reliable and efficient. electric service. During normal operation, .good unit response capability is essent.ia1 for stable implementation of the megawatt dispatch system load control concept [1 3. In emergency sit uatiom, responsive generation can be coordinated for load pickup or rejection in order t o avoid or minimize casmding of system disturbances [2]. It. is essent,ial that generating units have a sufficiently high degree of stability to be able to stick with the system through an emergency situation nithout unreasonable risk. Should isolation become necessary, the unit must be capable of controlled rejection of generat,ion without complete shutdown in order to service its local loads and to beavailable for system restoration [3]. These system operating requirements conflict with the obvious desire t o maximize the life and t o avoid damage of enormously expensive and complex primary equipment. This is a particularlyimportant concernat the present time when replacement generation is frequentlnot available or at best. involves extremely high operating costs. In recent years? new information concerning turbine metal fatigue due to cyclic thermal stress [4], [5] has made this a major Paper recommended by J. Peschon, Past Chairman of the IEEE lianuscript received September 1, 19i2; revised January 26, 19i3. S-CS Utility S-stems Committee. Electric Company, 2301 JIarket Street, Philadelphia, Pa. 19101. J. P. 1IcDonald is with the Research Division, Philadelphia H. G. Kaatny is n7it.h the College of Engineering, Ihesel University, Philadeiphia, Pa. 19104. consideration in operating generating stat,ions. Nevertheless, numerous reports [6]-[SI suggest, on t.he basis of t,est experience, that, the primary equipment itself does not. impose a serious inherent limitation on load change capabilhy and that, with suitably designed automatic controls, the objectives of system operation can be met-consistent. ni th 4 unit safety a.nd life requirements. I n [SI, Durrant and Vollnler suggest a variety of alternative operating and control strategies for boiler-t.urbinegenerator systems t o meet different, system operating objectives. Among these are nonstandard automatic control procedures such as using attemperating sprays to generate steam in the superheater for assisting load pickup, mariipulating gas flow for control of temperatures, incorporating variable steam pressure operation t o regulate t,urbine rotor temperature variations, and relaxing throttle temperature tolerances, also to obtain better control of rotor t,emperature. They note that the operating objectives are frequently conflicting with respect t o a given procedure and suggest further investigation t o clarify the implications of these alt ernatives for specific applications. Optimization and simulation prolqde a framework particularly n-ell suited to the identification and evaluation of alternative control strategies. There have been some previous attempts to apply optimal control t.heory to the control of a power boiler. Notable among these are the works of Sicholson [9][ 111 and Anderson [12]. Xcholson’s use of an oversimplified boiler model has made his positive results essentially meaningless for large power boiler applications. Xnderson’s work! on the other hand, follon-ed an extensive effort of model development. [13]. In E121 -4nderson concludes that integrated optimal con* trol schemes do not significantly improve the performance of the unit considered. -4nderson’s conclusions are c0ntra.rt o the optimism generated for coordinated control schemes by test experience and are also subject, to question on the basis that. the model used is still not an adequate characterization of a typical power boiler. In the work reported herein, every effort has been made to avoid such criticism. The model used in these studies has been used to simulate the Philadelphia Electric Company’s Cromby Xumber 2 unit, and has been subjected to extensive comparisons ni th closed-loop steady-state and open-loop transient, field tests [14], [16]. The model is nonlinear, and all manipulated variables normally considered for automated operation are included. Several rather subtle details which have been previously overlooked but which are critical to wide-range unit operation have been represented, such as multiple MC DONALD AND KWATNY: BOILER-TLTRBINE-GENERATQR CONTROL 203