Compressor design methods today have evolved significantly due to the growing capabilities of computer technology. It became possible to optimize them using 3D CFD models. The application of such approaches allows in an automated mode to form the design options of the compressor that best satisfy all the design requirements and limitations.

The main problems in optimizing multistage compressors are the large size of the computational model and consequently the time for obtaining one solution and the increased requirements for computing power. Also, the number of variables often exceeds 100. As a result, to obtain the optimum, many iterations with the computational model are required (over 1000), which makes the time to obtain a solution unacceptable. An important aspect of the successful optimization of a multistage compressor is the availability of a high-quality parametric model of the blades that works stably over the entire range of variation of geometric parameters. It is also necessary that the entire optimization cycle be carried out automatically with minimal user involvement.

In this paper, an approach to solving the above problems of optimizing multistage axial compressors is described.

At the first stage, a study was conducted to select the optimal settings for 3D numerical models for optimizing multistage compressors, which allow obtaining results close to experimental data in a minimum time. Recommendations were made on choosing the density of grids of finite volumes in the radial direction, along streamlines, and in the circumferential direction. They were obtained based on the physical idea of the flow structure in the compressor blade. Validation of the conclusions was made by comparing the calculated and experimental results of the NASA Rotor 37 compressor stage.

During the study, an algorithm and a program based on it were developed that provides automatic change of the shape of the blade flat sections of the axial compressor and the stacking laws for the sections along the blade height. The algorithm has various laws for variation of parameters along the blade height. As a result of the program execution, a file is generated that describes the shape of the blade in the format perceived by the CFD program. The developed program was used during optimization as a parameterizer, which converts the compressor geometric parameters into the “new” blade geometry.

Based on the results, an optimization algorithm for multi-stage axial compressors using 3D CFD modeling was developed. It can improve the basic parameters of the compressor (efficiency, pressure ratio, flow rate, etc.) by changing the shape of the blade profiles and their relative position. When searching for the optimum, it considers the possible parametric and geometric constraints.

The optimization process is based on the optimizer program (IOSO), which creates a vector of variable parameters. It is automatically transferred to the program for the parametric construction of the blade rows, where files with the geometry of the blades are formed, which are then transferred to the mesh-generator. After generating the computational grid, CFD calculation of the compressor is performed. Integral results are written to a text file, which is returned to the optimizer. There, based on the algorithms, a new combination of initial data is formed and the cycle is repeated until the desired optimum is reached. The developed algorithm allows performing multi-criteria optimization and getting the Pareto front, and then the user must choose the final option.

Using the developed optimization method, several optimization problems for multi-stage gas turbine axial compressors for various purposes (the number of stages from 3 to 15) were solved. During optimization, the objectives were to increase the margins of gas-dynamic stability, the pressure ratio and efficiency.


Optimization of the Workflow of Multistage Axial Compressors Using Modern Gas Dynamics Computing Systems

Division / Sponsor:
International Gas Turbine Institute (IGTI)

Paper Proceedings: https://asmedigitalcollection.asme.org/GT/GT2020/volume/84058

Turbo Expo - Virtual Conference

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.