Real-time Linear Modeling of Single-shaft Gas Turbine by Kalman Filter

In this study, dynamic modeling of a single shaftgas turbine examined. A simple multiple linear regressionwas chosen because of its simplicity for implementation,while its unknown parameters estimated according toKalman filter technique. A robust practical method ofKalman filter used due to maximum accuracy. The suggestedmodel in this paper is the simplest parametric model, whichsimulates and predicts the behavior of gas turbine with asatisfactory performance. Proposed model can be applied toall types of gas turbines that has a similar arrangement forinput signals. Easy implementation and generalization ofoffered model is exclusive among most available models.Actual data of V94.2 v5 heavy duty gas turbine used forproof of claims.

Keywords:

Single-shaft gas turbine , start-up modeling , multiple linear regression , kalman filter