A component map adaptation method for compressor modeling and diagnosis

The accuracy of model-based gas turbine performance prediction, diagnosis, and prognosis is mainly determined by the precision of compressor thermodynamic model. This article proposes a novel component map adaptation method for gas turbine compressor modeling and gas-path diagnosis. First, compressor characteristic maps are generated through stage-stacking calculation. And the generated compressor characteristic maps are expressed in the form of polynomial functions by partial least squares regression. After integrating the fitting polynomial functions into the gas turbine performance model, the adaptation of characteristic maps in the performance model by gas-path measurements from the target engine can be made easily, through tuning the fitting polynomial coefficients. Application and analysis have illustrated that the proposed approach can effectively improve the accuracy of the gas turbine thermodynamic model during off-design performance calculation. The proposed approach exhibits a great potential to tune compressor characteristic maps for both compressor performance modeling diagnosis, with full consideration of their highly nonlinearity.