Equation-oriented methods for design optimization and performance analysis of radial inflow turbines

This paper presents methods for design optimization and performance analysis of radial inflow turbines. Both methods are formulated in an equation-oriented manner and involve a single mathematical problem that is solved by an efficient, gradient-based optimization algorithm. In addition, the comparison of the model output with experimental data showed that the underlying mean-line flow model accurately predicts the variation of mass flow rate and isentropic efficiency as a function of the pressure ratio, rotational speed, and nozzle throat area. Moreover, the capabilities of the proposed methods were demonstrated by carrying out the preliminary design and performance prediction of the radial inflow turbine of an organic Rankine cycle. The results indicate that the design optimization method converges to the global optimum solution, regardless of the start values for the independent variables. In addition, the performance maps generated by the performance analysis method are physically consistent and agree with general findings from experimental data reported in the open literature. Considering the accuracy, robustness and low computational cost of the proposed methods, they can be regarded as a powerful tool for the preliminary design and performance prediction of radial inflow turbines, either as a standalone component or as part of a larger system. •New methods for design optimization and performance analysis of radial turbines.•The design optimization method reliably converges to the global optimum.•The analysis method predicts physically consistent performance maps.•Validation of the underlying mean-line model against experimental data.•Novel treatment of choked flow in the case of supersonic flow velocities.