AM Software

Broccoli

Broccoli is an open source C++ header-only library for general robotics applications. The library includes tools and algorithms for the control of robots under hard real-time requirements. Broccoli is largely platform-independent - at the chair it is used on QNX Neutrino and LinuxRT.

The project has evolved from refactored code of the humanoid robot project and serves as a development basis for robotics research at the chair. Broccoli has a stable API and is licensed under the GPL v3 license.

GitLab Page: https://gitlab.lrz.de/AM/broccoli

AMfe

AMfe is a finite element code written in Python that focuses on

  • modularity
  • extensibility
  • easy prototyping of new methods for research

It originates from the demand to make the assembly of the finite element routine easy accessible to the researcher to be able rapidly test new developed hyper-reduction methods. The code has grown over the past years and has gained a very flexible modular structure that enables researchers to rapidly test new methods for simulation of structures. Visit our repository at GitHub: https://github.com/AppliedMechanics/AMfe

Features

Basic features

  • Completely written in Python
  • Input: Gmsh and GiD
  • Output: Paraview HDF5, XDMF
  • Geometric Nonlinear total Lagrangian formulation
  • Nonlinear Constraints
  • Constraint formulations: Elimination (only fixed dirichlet), Lagrange Multiplier, Nullspace Elimination
  • Fixed direction Neumann conditions and forces that keep their orientation relative to the body
  • Hyperelastic and viscoelastic materials
  • HTML Documentation with Tutorials, Fundamentals, Examples and Reference Guide
  • Fortran routines to accelerate critical parts of the code
  • Parametric meshes (experimental)
  • Substructuring (experimental)

Model Reduction Features

  • Modal truncation
  • Modal and static modal derivatives
  • Krylov reduction
  • Energy Conserving Sampling and Weighting Hyperreduction (ECSW)
  • Polynomial Expansion (Hyperreduction)

pyFBS

pyFBS is an open source Python package for Frequency Based Substructuring and Transfer Path Analysis. It enables the user to use state-of-the-art dynamic substructuring methodologies in an intuitive manner. With the package also basic and application examples are provided, together with real datasets so you can directly try out the capabilities of the pyFBS. The package was developed as a part of collaboration between the Laboratory for Dynamics of Machines and Structures (LADISK), University of Ljubljana, Faculty of Mechanical Engineering (UL FME) and the Chair of Applied Mechanics (AM), Technical University of Munich (TUM).

Features

Basic features

  • 3D display
  • FRF synthetization
  • Virtual Point Transformation (VPT)
  • System Equivalent Model Mixing (SEMM)
  • Singular Vector Transformation (SVT)

Application features

  • Operational Deflection Shapes
  • VPT Coupling
  • VPT Decoupling
  • Experimental Modal Analysis (EMA)
  • Transmission Simulator in FBS
  • In-situ Transfer Path Analysis (iTPA)
  • SVT Decoupling

For more information on features, basic and application examples check out the documentation or visit our repository at GitLab: https://gitlab.com/pyFBS/pyFBS

AMrotor - Rotor Simulation Code

During the last years, a Matlab toolbox for simulation of rotating machines has been developed at the Chair of Applied Mechanics of the Technical University of Munich.
For the geometry of a rotor, a 2d silhouette can be given by a simple point description. Then, a mesh of beam elements is created which are assembled into a MCK-model. In the next step, different components like bearings, external forces and loads (e.g. unbalance) are added to the system. It is also possible to add time variant loads and nonlinear or even active components, e.g. magnetic bearings. Different types of analysis can then be performed, like modal analysis, campbell diagrams or time integration.
The whole toolbox is programmed in an object oriented way. The code is meant to be a research code which focuses more on easy architecture than on execution performance. The goal is to enable easy implementation of new components with own methods and testing them.

The source code is published on github.com/AppliedMechanics/AMrotor/.