A Neural Network Simulator of a Gas Turbine With a Waste Heat Recovery Section

Abstract

The objective of the paper is to assess the feasibility of the neural network (NN) approach in power plant process evaluations. A “feed-forward” technique with a back propagation algorithm was applied to a gas turbine equipped with waste heat boiler and water heater. Data from physical or empirical simulators of plant components were used to train such a NN model. Results obtained using a conventional computing technique are compared with those of the direct method based on a NN approach. The NN simulator was able to perform calculations in a really short computing time with a high degree of accuracy, predicting various steady-state operating conditions on the basis of inputs that can be easily obtained with existing plant instrumentation. The optimization of NN parameters like number of hidden neurons, training sample size, and learning rate is discussed in the paper.

Prediction of power output of a coal-fired power plant by artificial neural network

“…Similar works are also reported for the mathematical modeling of turbines [11][12][13][14][15][16][17][18][19][20]. Different aspects of the development of ANN models for gas turbines have been reported also by several authors [21][22][23]. However, ANN models for steam turbines using real plant data are not available.…”

Thermodynamic optimization of solid oxide fuel cell-based combined cycle cogeneration plant

“…A. Carretero and B. V. Reddy (10)(11)(25)(26). This superheated steam is used to drive the steam turbine (26, 27, 28).…”

“…The gas turbine compressor provides the compressed air required throughout the various components in the plant, making use of multistage compression (1-2,3-6) and inter-cooling (2-3). A mixture of air, coal, and steam is used to produce syngas in the gasifier (9,10). The high temperature syngas is cooled in a counter flow heat exchanger where superheated steam is produced Figure 1.…”

“…The performances of the HRSG in the combined cycle cogeneration unit using neural networks also have been investigated [10]. A genetic algorithm routine to obtain optimal operating conditions of a combined cycle cogeneration has been investigated by Rovira et al [11] to resolve the uncertainties caused by the Newton-Raphson algorithm.…”

A new adaptive Mamdani-type fuzzy modeling strategy for industrial gas turbines

“…[6] has also presented the use of ANNs in condition motoring of engine gas generators to improve data quality and for performance trend change detection. Gas turbine health monitoring has been carried on in [7] using ANNs for a high bypass ratio military turbofan engine, where double-component faults have been presented to demonstrate the application of ANNs as a diagnostic tools, and [8] applied a feed forward ANN with a back propagation algorithm to a gas turbine equipped with a waste heat recovery section to simulate and predict various steady-state operating conditions. ANNs have therefore been demonstrably successful for modeling gas turbines -however, the black-box model provides little or no insight of the physical underlying attributes of such complex systems.…”