Investigation Of Multistge Axial-compressor Aerodynamic Optimization Design

Gas turbine is a power equipment which is of great significance to both national defence and economy. As one of the three main subsystems of gas turbine, the design technology of multistage axial flow compressor is a core technology. However, it's very difficult to obtain high performance design due to the complicated flowfield within the multistage axial-flow compressor, which becomes one of the bottle-necks in the development of gas turbine technique.The general idea of domain decomposition and targeted optimization was proposed based on the consideration of the flow feature and loss mechanism in multistage axial-flow compressor. Several different optimization methods was introduced for controlling corresponding loss including profile loss, tip leakage loss, corner separation loss and seal leakage loss. The Complex Optimization Problem was then decomposed and simplified. The aerodynamic optimization design could be carried out with more directivity and controllability.An aerodynamic optimization design system for multistage axial-flow compressor with the combination of inverse design method and true design method was established, in which parameterized modeling, flowfeild evaluation and genetic algorithms were integrated. The 2-D blade profile was modeled by a camberline described by cubic Bezier curve and given thickness distribution. The 3-D blade was then modeled by stacking the 2-D profile along the stacking line, which is a combination of three curves. The performance of 3-D blade was evaluated by CFD, and the genetic algorithms were improved for seeking of the optimal solution.A solution of multistage optimization is proposed as objective function decomposition and optimize concurrently. By applying this method a multistage optimization problem could be solved more efficiently. Samples of single row and multistage optimization were carried out for the purpose of verification. The optimization result of high pressure dual-stage in a multistage axial-flow compressor and transonic 2.5 stages in a multistage axial-flow compressor confirmed the reliability of the aerodynamic optimization design system. The time consume of optimization was cut down by the solution of objective function decomposition and optimize concurrently, which was of high engineering application value.In addition, an optimization method of hub ring cutting is proposed to reduce the seal leakage loss. Results of both steady and unsteady CFD indicated that the method was able to control the mixing of leakage flow and main flow, reduce the mixing loss and improve compressor performance.In conclusion, the idea of domain decomposition and targeted optimization is reasonable and effective. The aerodynamic optimization design system for multistage axial-flow compressor solved the problem of multistage optimization and was able to optimize multistage axial-flow compressor design efficiently.