Solving characteristic parameters of heavy-duty gas turbines using parameter estimation method.

Area of Science:

• Thermodynamics
• Mechanical Engineering
• Computational Fluid Dynamics

Background:

• Accurate gas turbine simulation requires precise component parameterization.
• Manufacturers often provide limited operational data, hindering reliable performance prediction.

Purpose of the Study:

• To develop a modeling approach for gas turbine simulation using limited operational parameters.
• To apply this framework to a 9FA heavy-duty gas turbine for performance analysis.

Main Methods:

• Utilized maximum likelihood parameter estimation within an inverse problem formulation.
• Employed a modular methodology to reconstruct compressor characteristic curves.
• Established a full-condition mathematical model for the gas turbine.

Main Results:

• Maximum relative error for compressor discharge flow rate and temperature was 0.2%.
• Relative errors for compressor isentropic efficiency (0.34%) and combustion efficiency (0.1%) were low.
• Parameter prediction errors were below 0.5%, with combustion pressure loss coefficient error at 2.86%.

Conclusions:

• The proposed method enables accurate full-condition gas turbine modeling with limited data.
• Cross-validation confirmed the approach's accuracy under data-scarce conditions.
• This provides a valuable tool for gas turbine performance analysis and simulation.