A Quasi-One Dimensiona Model for Axial Turbines

in Proceedings of the 18th ISABE Conference (2007, September)

An axial turbine model is presented that is intended to predict its aerodynamic performance based only on the turbine geometry and thermodynamic environment. The model is an extension of an existing quasi ... [more ▼]

An axial turbine model is presented that is intended to predict its aerodynamic performance based only on the turbine geometry and thermodynamic environment. The model is an extension of an existing quasi-1D compressor representation. The simulation tool is able to compute the flow through a whole multistage turbine, with detail at the blade row level. It relies upon a quasi-one dimensional Euler system of equations, expressed here in curvilinear coordinates, and resulting from the application of mass, momentum and energy conservation principles in finite-volume formalism. The source terms expressing the interactions between the flow, the blades and the flowpath are determined using the velocity triangles for each blade row, at mid-span. The solver performs an elaborate implicit time-marching resolution of the equations. The enthalpy loss coefficients as well as the blade outlet flow angles are evaluated through open literature correlations. An efficient representation of the Craig-Cox loss coefficients and the Ainley-Mathieson outlet flow angle correlation brings the necessary empirical information for the velocity triangle computations. The computer code was validated against a high pressure turbine test case featuring multiple cooling flows. The results show the good capabilities of the turbine model using only standard correlations. The computed efficiency also shows the need to model the cooling losses. The low speed, low expansion rate results may finally indicate that the code accuracy would benefit from a correlation parametric identification such as the one led in the compressor case. [less ▲]

Explicit Thermodynamic Properties Using Radial Basis Functions Neural Networks

in Proceedings of the second SIAM international Conference on Data Mining (2002)

Gas turbine design, development, monitoring and maintenance are widely based on numerical simulations of the steady and transient engine performance. Most of the equations that are solved in the ... [more ▼]

Gas turbine design, development, monitoring and maintenance are widely based on numerical simulations of the steady and transient engine performance. Most of the equations that are solved in the simulation programs involve the thermodynamic properties of the fluid flowing through the engine. These properties depend on temperature, pressure, humidity and fuel dosage. As the solution of chemical equilibrium is not compatible with real-time computations, a chemical solver is used off-line to generate a large database which neural networks are trained on. These networks are built on radial basis functions such as multiquadrics. A forward selection approach is used to select data points from the training set as the centers of the transfer functions. The selection stops when the prediction error starts growing. The resulting networks for specific heat and enthalpy of the gas mixture are 3 orders of magnitude faster than the chemical solver. In order to further increase the efficiency and the generalization capabilities of the model, an external optimization solver has been used to tune the shape of the transfer functions. Several solutions are proposed and preliminary results are presented. [less ▲]

Développement d’une méthodologie d’optimisation aérodynamique et mécanique d’aubes de compresseurs

in Proceedings of the 4th Belgian National Congress on Theoretical and Applied Mechanics (1997)