Simulation of power plant transients with artificial neural networks: Application to an existing combined cycle

Abstract

To maintain the high performance of gas-turbine-based combined cycles, transients must be properly taken into account in the design phase and efficiently monitored in the operational phase, because they are not negligible time intervals. The use of artificial intelligence techniques such as expert systems, fuzzy sets and neural networks (NNs), coupled with advanced measurement and monitoring devices, can provide a reliable and efficient monitoring system.

An existing two-pressure-level combined cycle has been simulated by dividing its simplified model into blocks representative of the main elements. An NN is associated with each of these blocks. Once the training and testing of the NN are complete, using data from a simulator, the blocks are put either in a cascade arrangement or in a parallel arrangement, providing reliable systems that can predict the load-change transient behaviour of the entire plant. The parallel approach was then tested on data from the real plant. The excessive simplification introduced with the simulator required the addition of selected real cases to the training set that are able to fit the NN response to reality. The results obtained are encouraging for use in an on-line monitoring system which evaluates the difference between the measured data and the predicted data.

Keywords

simulation of transients power plants neural networks combined cycle

Get full access to this article

View all access options for this article.

References

Desideri

Fibbi

A simplified approach to off-design performance evaluation of combined cycle powerplants with single pressure steam cycles. In Proceedings of the 1993 ASME Cogen Turbo, Bournemouth, 1993 (American Society of Mechanical Engineers, New York).

Desideri

Facchini

Analisi dei sistemi di regolazione nei cicli combinati gas-vapore per la produzione di energia elettrica (in Italian). In Proceedings of the Convegno Nazionale on Gruppi Combinati—Prospettive Tecniche ed Economiche, Milan, Italy, 1993.

Chefneux

Mathieu

Ph.

Part load performance of fired combined cycle power plants. In Proceedings of the VIII Convegno Nazionale on Tecnologie e Sistemi Energetici Complessi, Bologna, Italy, 1995, pp. 125–136.

Dechamps

P. J.

Pirard

Mathieu

Ph.

Part load operation of combined cycle plants with and without supplementary firing. In Proceedings of the 1994 ASME Cogen Turbo, Portland, Oregon, 1994, Vol. 9, pp. 371–380 (American Society of Mechanical Engineers, New York).

Bettagli

Facchini

Comportamento di un gruppo combinato di media potenza al variare del carico elettrico e termico (in Italian). In Proceedings of the VIII Convegno Nazionale on Tecnologie e Sistemi Energetici Complessi, Bologna, Italy, 1995, pp. 243–254.

Mathieu

Ph.

Dechamps

P. J.

Modeling of transients in fired one pressure level CC plants. In Proceedings of the ECOS1996, Stockholm, Sweden, 1996.

Desideri

Fantozzi

Bidini

Mathieu

Ph.

Use of neural networks for the simulation of combined cycle transients. ASME paper 97-GT-442, 1997.

Bidini

Grimaldi

C. N.

Mariani

Applicazione delle reti neurali artificialiai problemi energetici (in Italian). In Proceedings of the 49 Ã, ° Congresso Nazionale dell' Associazione Termotecnica Italiana, Perugia, Italy, 1994.

Grimaldi

C. N.

Mariani

Artificial neural network for the evaluation of the energy requirements in manufacturing plants. In Proceedings of the ASME Winter Annual Meeting, Chicago, Illinois, 1994.

10.

Bacha

Meyer

Automated load forecasting using neural networks. In Proceedings of the American Power Conference, Chicago, Illinois, 1992, Vol. 2, pp. 1144–1152.

11.

Song

Wilkie

Grimble

M. J.

Integrated robust/ neural controller with gas turbine applications. In Proceedings of the 1994 IEEE Conference on Control Applications, Glasgow, UK, 1994, pp. 411–415 (IEEE, New York).

12.

Pomfret

C. J.

Engine life measurement and diagnostics

Aerospace Engng, 1994, 14(6), 11–14.

13.

Brownell

T. A.

Neural networks for sensor management and diagnostics. In Proceedings of the National Aerospace Electronics Conference, Dayton, Ohio, 1992, pp. 923–929.

14.

Chbat

N. W.

Rajamani

Ashley

T. A.

Estimating gas turbine internal cycle parameters using a neural network. ASME paper 96-GT-316, 1996.

15.

Sakawa

Inuiguchi

Matsubara

Uemura

Shiromaru

Suginohara

Application of neuro-like fuzzy method to gas turbine NO_x concentration prediction problem. In Proceedings of the 1993 International Joint Conference on Neural Networks, Nagoya, Japan, 1993, pp. 2913–2916.

16.

Kanelopoulos

Stamatis

Mathioudakis

Incorporating neural networks into gas turbine performance diagnostics. ASME paper 97-GT-35, 1997.

17.

Torella

Lombardi

Neural networks for the diagnostics of gas turbine engines. ASME paper 96-TA-39, 1996.

18.

Roemer

Atkinson

Real-time health monitoring and diagnostics for gas turbine engines. ASME paper 97-GT-30, 1997.

19.

Grimaldi

C. N.

Mariani

On board diagnosis of internal combustion engines: A new model definition and experimental validation. SAE paper 970211, 1997.

20.

Bidini

Grimaldi

C. N.

Mariani

Innovative control and diagnostic techniques for cogeneration plant steam generators. ASME paper 95-CTP-74, 1995.

21.

Ikonomopoulos

Uhrig

R. E.

Tsoukalas

L. H.

Use of the neural networks to monitor power plant components. In Proceedings of the American Power Conference, Chicago, Illinois, 1992, Vol. 2, pp. 1132–1137.

22.

Kraft

Hybrid neural net and rule based systems for boiler monitoring and diagnosis. In Proceedings of the American Power Conference, Chicago, Illinois, 1991, Vol. 2, pp. 952–957.

23.

Rumelhart

D. E.

McClelland

J. L.

Parallel Distributed Processing, 1990 (MIT Press, Cambridge, Massachusetts).

24.

Vogl

T. P.

Accelerating the convergence of the back-propagation method

Biol. Cybernetics, 1988, 59, 257–263.

25.

Dayoff

Neural Network Architectures. An Introduction, 1990 (Van Nostrand Reinhold, New York).

26.

Nguyen

Widrow

Improving the learning speed of a 2-layer neural network by choosing initial values of the adaptive weights. In Proceedings of the International Joint Conference on Neural Networks, 1990, Vol. 3, pp. 21–26.

27.

Dechamps

P. J.

Modelling the transient behaviour of combined cycle plants. ASME paper 94-GT-238, 1993.

28.

Lombardi

Previsione del funzionamento in transitorio di sistemi energetici con tecniche di intelligenza artificiale (in Italian). MSc thesis, University of Perugia, Perugia, Italy.