Multiple-shaft Turbine Model Reduction by Fractional-order Modeling

Traditionally, turbines are usually represented by a single mass to simplify the analysis of its dynamics. However, without accurately modelling its multiple shafts, some dynamic issues such as sub-synchronous resonance may be overlooked. This paper proposes a fractional-order model optimization based model reduction approach to significantly reduce a large number of parameters in the detailed multiple-shaft turbine model while still keeping its accuracy. In this approach, an optimization problem is defined to minimize the frequency response difference between the reduced fractional-order model and the original complex integer-order model. The proposed approach is validated on an 11-shaft steam turbine and a 243shaft gas turbine within New England power system.