Thermo-fluid dynamic framework for probabilistic design of gas turbines (OptiSysKom AP 3.2b)

Project director:	Prof. Dr.-Ing. habil. Ronald Mailach
Research Associate:	Dipl.-Ing. Marius Stricker
Scientific cooperation:	AG Turbo
Financing:	BMWK
Duration:	08/2020 - 07/2023

Description:

The work package 3.2b "Thermo-fluid dynamic framework for probabilistic design of gas turbines" of the joint project "Optimization of processes and systems as well as the service life of the entire system and its components (OptiSysKom)" is being processed at the chair.

Stationary gas turbines are operated much more frequently in transient alternating loads and in partial loads. In addition to the effects on the existing gas turbines, this also has an impact on the development of new gas turbines and their maintenance schedules. Manufacturers are focusing their attention on robust operation with high efficiency over a wide load range.

The focus of this work package is on the probabilistic investigation of the secondary air system, which makes an important contribution to efficiency and reliable operation through cooling, sealing and control of the axial loads.

As can be guessed from Figure 1, this is a complex subsystem in which the influence on the scatter of the result variables is often not fully apparent.

Probabilistics can make a valuable contribution here. Probabilistic methods make it possible to include uncertainties directly in the calculation of lifetime consumption. In this way, the previously used safety factors can be checked or even eliminated. Any existing conservatism can thus be identified and abandoned in favor of a more flexible operating mode. This ensures optimal material utilization and machine utilization. Likewise, the data resulting from probabilistic analyzes can be used to efficiently determine important influencing variables or for design recommendations. This also allows the design process to be made more efficient and the available information to be used more productively. Due to the necessary flexibilization of turbomachinery, one focus point for the proposed project is to quantify probabilistic or stochastic methods for describing uncertainties in the secondary air system of gas turbines and to make these methods applicable and suitable for industry. The resulting time-dependent findings will contribute significantly to a better understanding and to the robustness assessment of the machine in the case of heavily fluctuating loads. In addition to the development and application of suitable methods from the field of probabilistic, the adaptation and application of the methods to an engineering problem is also in the foreground. Specifically, the robust sealing of the cavities against hot gas and the sensitivity of the cooling air supply to the rotor blades and vanes of the turbine as well as their effect on the component temperatures and stresses are of interest. The objective of sub-project 3.2b is therefore the development of a workflow for the execution of probabilistic investigations based on the tool chain provided from an in-house secondary air system solver and Ansys Mechanical. In addition, the focus is on the development of methods investigate the effects of a large number of input parameters in a limited number of evaluations and to use them for further applications.

Industrial partner:  MAN Energy Solutions SE