Multidisciplinary Optimization of Complex Technical Systems

Modern computer technologies now allow us to conduct rather complex mathematical calculations in a relatively short period of time. Thus, it has become possible to employ optimization methods in the design of complex technical systems, even when calculations require large computational resources (structural, thermal, and gasdynamics calculations).

The designer may have to vary more than a hundred design variables and constraints during the optimization process. Therefore the procedure of preparing the initial data for optimization may take a long time. That is why we developed the optimization software for designing technical systems. This software system includes the IOSO optimization procedure and modules of automatic data preparation and handling. The data is represented in the format that is convenient and understandable for a designer.

The optimization procedure is based on the response surface methodology, when response surfaces are constructed for objective functions and constraints and then optimized at each iteration in a current search region. The objective function and constraints are then evaluated at the optimal point using the mathematical model of the system under consideration.

The paper presents the results of optimizing a three-stage axial compressor. The optimization goal was to improve the compressor efficiency at two flight conditions by optimizing geometry of the 5 compressor rows (62 design parameters). As the analysis tools the well-known commercial software package (FINE/Design3D) is used.

The conducted investigations showed that it is possible to perform optimization studies using 3-D method of calculating the flow parameters in the axial compressor. The most important elements in organizing the optimization process, that influence the efficiency of such studies are the following: the capability to automatically generate the mesh for a wide range of blade geometry configurations and an efficient optimization algorithm, that allows for a maximum improvement in objective function with minimum number of flow model evaluations. In our particular case as a result of optimizing compressor blade geometry parameters we got a 3% improvement in maximum efficiency while satisfying all the constraints.