An axial compressor loss and deviation model that was developed and presented during previous research has now been used to develop a computer program for multistage compressor design. A set of input data including overall parameters such as pressure ratio and mass flow rate and the first-stage parameters such as inlet flow coefficient and rotor tip Mach number are used to determine a number of stages and their geometry, speed and relevant flow properties. Then a subsequent redistribution of parameters for separate stages can be carried out in order to increase the stability indicators and efficiency over a desired operating range. A selected stage vortex law determines velocity triangles and blading geometry for hub and tip sections, which allows the generation of a realistic flow path shape. The developed program is considered to be a flexible and stable tool useful for tasks of manual or automated optimization when combined with an external optimization algorithm. This paper presents a basic mathematical model and the level of accuracy achieved. This is demonstrated through examples of manual design and redesign cases, while automated optimization will be included in future research.

The aim of this work was to develop a new system for optimization of parameters for combined cycle power plants (CCGTs) with triple-pressure heat recovery steam generator (HRSG). Thermodynamic and thermoeconomic optimizations were carried out. The objective of the thermodynamic optimization is to enhance the efficiency of the CCGTs and to maximize the power production in the steam cycle (steam turbine gross power). Improvement of the efficiency of the CCGT plants is achieved through optimization of the operating parameters: temperature difference between the gas and steam (pinch point P.P.) and the steam pressure in the HRSG. The objective of the thermoeconomic optimization is to minimize the production costs per unit of the generated electricity. Defining the optimal P.P. was the first step in the optimization procedure. Then, through the developed optimization process, other optimal operating parameters (steam pressure and condenser pressure) were identified. The developed system was demonstrated for the case of a 282 MW CCGT power plant with a typical design for commercial combined cycle power plants. The optimized combined cycle was compared with the regular CCGT plant

Part I of this paper presents a method and a computer program for the mean design of multistage axial compressors. This second part describes a method and an additional computer routine that use the basic mean line design to create a fully two-dimensional flow solution and a compressor design. The two-dimensional solution according to a selected swirl vortex function is calculated using streamline curvature throughflow equations and spanwise distribution of losses. An iterative calculation procedure slightly reshapes the initial flow path in order to retain the desired input flow coefficients. Other variables such as stage loading parameters are changed in order to obtain the desired overall pressure ratio. A spanwise distribution of certain stage parameters can then be adjusted to achieve desired radial flow field variations. The basic one-dimensional input data can be varied at any moment to obtain a new one-dimensional result and the corresponding two-dimensional solution. A new output is created instantaneously and can be used for further CFD analysis, external throughflow, blade-to-blade flow computations or mechanical and vibration analysis.

No abstract

By matching a well established fast through-flow analysis code and an efficient optimization algorithm, a new design system has been developed which optimizes hub and shroud geometry and inlet and exit flow-field parameters for each blade row of a multistage axial flow turbine. The compressible steady state inviscid through-flow code with high fidelity loss and mixing models, based on stream function method and finite element solution procedure, is suitable for fast and accurate flow calculation and performance prediction of multistage axial flow turbines at design and significant off-design conditions. A general-purpose hybrid constrained optimization package has been developed that includes the following modules: genetic algorithm, simulated annealing, modified Nelder-Mead method, sequential quadratic programming, andDavidon-Fletcher-Powell gradient search algorithm. The optimizer performs automatic switching among the modules each time when the local minimum is detected thus offering a robust and versatile tool for constrained multidisciplinaryoptimization. An analysis of the loss correlations was made to find parameters that have influence on the turbine performance. By varying seventeen variables per each turbine stage it is possible to find an optimal radial distribution of flow parameters at the inlet and outlet of every blade row. Simultaneously, an optimized meridional flow path is found that is defined by the optimized shape of the hub and shroud. The design system has been demonstrated using an example of a single stage transonic axial gas turbine, although the method is directly applicable to multistage turbine optimization. The comparison of computed performance of initial and optimized design shows significant improvement in the multistage efficiency at design and off-design conditions. The entire design optimization process is feasible on a typical single-processor workstation.