Abstract: Due to the difficulties to measure the coal consumption and calorific value, using positive balance method is not an accurate way to determine the net coal consumption rate for the coal fired units. In this work, the diagnostic analysis method for coal consumption rate by taking advantage of positive balance procedures in power plants is proposed. Integrated with the designed performance of boiler and steam turbine, the thermodynamic test data through the indirect balance method and the accuracy class of instruments, this method is employed on the basis of the maximum likelihood principle to reconcile the daily measurement data in coal consumption rate test. The practice on a 350 MW supercritical unit shows that through the application of the reconciliation model the gross measurement errors can be detected from the test data and then reconciled to reasonable values. After the reconciliation, coal consumption rate values are highly accurate and stable, which decrease as the unit load increases. Besides, even though as much as 52% of the measurement data have been reconciled, the average measurement value of the coal consumption rate will not be altered by this model within the entire period, because of the positive and negative deviations of the reconciled data cancelling out each other over a longer period.

Key words: coal-fired unit, net coal consumption rate, measurement error, data reconciliation