Inequality constrained nonlinear data reconciliation of a steam turbine power plant for enhanced parameter estimation

Abstract

Measurement errors inevitably exist amongst on-line measured data in steam turbine power plants. Problematic steam turbine isentropic efficiencies and turbine expansion curves are therefore obtained in the existence of measurement errors. Data reconciliation is widely used for uncertainty reduction of measurements and parameter estimation. Inequality constraints or bounds are rather necessary in some cases to adjust parameter estimates to be physically meaningful. In this work, we apply an inequality constrained nonlinear data reconciliation approach to the thermal system of a power plant, and compare its effect with equality constrained approaches. Case studies using performance test data and operational measurement data of a real-life 1000 MW steam turbine power plant are provided. The necessity and difference brought by inequality constraints are also discussed. Corrected expansion curve with reasonable enthalpy–entropy relationships and better estimates of isentropic efficiencies are obtained after implementation of inequality constraints. Results show that uncertainties of most measured parameters are reduced by 30–80 percent, and uncertainty of the calculated exhaust steam enthalpy is reduced by 22 percent.