Using data reconciliation to improve the reliability of the energy evaluation of a gas-and-steam CHP unit

Abstract

The development of distributed control systems and automatic storage of measuring data permits a better supervision of processes of energy conversion. In most cases the running energy evaluation of these processes based on the raw measuring data. The paper deals with the application of the data reconciliation method in order to improve the reliability of thermal measurements of a gas-and-steam CHP unit. Identification of the thermal system of an example gas-and-steam CHP unit and its measurements system have been carried out. A mathematical model of investigated CHP unit equipped with a gas turbine, heat recovery steam generator, extraction-condensing steam turbine and heat exchangers has been presented. In the case of energy evaluation mainly balance equations of mass and energy are taken into account.From the data reconciliation method point of view a surplus of measurements information in the distributed control system of analyzed CHP unit has been found. As a conditional equations in the data reconciliation algorithm, mass and energy balances formulated so far within the model of gas-and-steam CHP unit have been applied. Data reconciliation calculations for analyzed gas-and-steam CHP unit have been carried out. The statistical test of control of the assumed uncertainty of raw measurements using variance-covariance matrix of measurements corrections has been applied.For global assessment of results of data reconciliation calculations a relative information entropy has been applied. The relative entropy – Kullback-Leibler divergence, which is a non-symmetric measure of the difference between two probability distributions, has been used. The necessity of data reconciliation method to increase reliability and uncertainty of measurements data in calculations of parameters characterizing the process of energy conversion in gas-and-steam CHP unit has been demonstrated. Usefulness of application of an information entropy for global assessment of reliability improvement of measurements after data reconciliation calculations has been presented.