As the computational power of computers increases the application of high-end numerical methods continues to allow ever greater insight into the complex flow physics of modern turbomachines and therefore the design of ever more efficient and powerful turbomachinery components.
Since the early 1990’s the Numerical Methods department at DLR’s Institute of Propulsion Technology has been working on the CFD code TRACE (Turbomachinery Research Aerodynamic Computational Environment) in order to calculate and investigate the complex flows in turbomachinery. Within DLR, TRACE is the standard method for the simulation of internal flows. Outside of DLR, at universities and other research institutes, TRACE is used for the scientific analysis of turbomachinery flows. In addition, at MTU Aero Engines and Siemens Energy, TRACE is employed in industrial design environments for the design and optimization of turbomachinery components.
The future development of the program system TRACE is based on three pillars: software engineering, mathematical models and physical modeling. The aim is to develop a simulation and research tool specifically for the challenges particular to turbomachines and identify the modules needed to ensure high quality, trustworthy results. To achieve these goals special focus is being placed on the areas of aeroelasticity, aeroacoustics, aero-thermodynamics, turbulence and the usage of next generation computer architectures.
The core tasks of the department are:
The simulation system TRACE consists of the modules PREP, POST and TRACE. PREP is a preprocessor for mapping blade eigenmodes of a FEM calculation on a CFD grid for subsequent simulation of flutter or forced response. TRACE is the hybrid (structured and unstructured) flow solver with a nonlinear solver in time domain - as well as in the frequency domain, a linearized module in the frequency domain and an adjoint flow solver. POST is a comprehensive software tool for the global analysis of stationary and non-stationary multi-stage turbomachinery simulations. Furthermore, the template and parameter-based mesh generator G3DHexa is developed for structured grid generation.
The simulation system TRACE is subject to a rigorous quality management, which includes a test and validation suite, supported by an automated nightly build process. The Python-based processes are controlled by the use of state of the art software development tools (Eclipse, SVN, Jenkins, etc.)
The program system TRACE thus offers a quality assured basis for the simulation of turbomachinery flows and is used by the industry for the design of aero engine components (MTU) and stationary gas turbines (Siemens Energy). In addition, another 15 universities within Germany, alongside the KTH in Sweden, use the software package to support their turbomachinery research. Three of the universities within Germany are also actively involved in the further development of TRACE.