Improving the accuracy of electricity and heat production forecasting in a supervision computer system of a selected gas-fired CHP plant operation. (15th January 2022)

• Record Type: Journal Article
• Authors: Szega, Marcin; Żymełka, Piotr; Janda, Tomasz
• Is Part Of: Energy. Volume 239:Part E(2022)
• Publication Date: 2022-01-15
• Subjects: Combined heat and power unit -- Gas turbine -- Mathematical modelling -- Advanced data validation and reconciliation -- Cross-validation -- Stepwise regression; Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Energy consumption -- Periodicals; 333.7905
• DOI: 10.1016/j.energy.2021.122464
• Languages: English
• ISSNs: 0360-5442

Abstract

Abstract: The process of optimal planning of heat and electricity production in CHP plants is one of the most important aspects influencing economic efficiency. The effectiveness of the optimal planning results directly from the quality of generated forecasts and available calculation tools. To improve the accuracy of forecasting the production of energy carriers in the CHP plant, it has been proposed to extend the existing computer system for supervising operation with an additional calculation module. A procedure for calibrating simulation models of gas turbines based on manufacturer's correction curves used to calculate electricity production has been presented. Due to the occurring measurement redundancy, advanced data validation and reconciliation of measurements was used. In the preparation of measurement data for the development of calibration functions, the generalized likelihood ratio method was used to separate the steady-state periods of gas turbine operation. In the procedure of determining the mathematical form and coefficients of empirical functions, the cross-validation method was applied. Empirical functions for the prediction of the exhaust gases temperature at the outlet of heat recovery hot water boilers for forecasting the production of heat have been developed. Very high values of quality-of-fit measures for regression were obtained. The developed calibration models and empirical functions characterized by high quality can significantly improve the accuracy of the optimal planning for the production of energy carriers in the investigated CHP plant. Highlights: Calibration procedure for a gas turbine model using correction curves is presented. Generalized likelihood ratio algorithm was used to detect the process changes. Application of the generalized advanced data reconciliation method. Method of k-fold cross-validation was applied in the calibration function development. Achieving an increase in the accuracy of energy carriers production planning in a CHP.