Decoding Drone FEM: A Journey with Code_Aster and CloudHPC
Finite Element Analysis (FEA) is a powerful tool for understanding the structural behavior of complex designs. In this post, we’ll delve into an FEA analysis performed on a drone frame using the open-source software Code_Aster. Our goal was to analyze its static and dynamic characteristics, and we leveraged the power of parallel processing on remote hardware via CloudHPC to expedite the computations.
Laying the Foundation: Mesh Generation with SALOME
The first crucial step in any FEA is creating a discrete representation of the physical geometry – the mesh. For this drone frame, we utilized SALOME, an open-source platform for pre- and post-processing. The bottom left quadrant of the image showcases a detailed view of the generated mesh, specifically employing the netgen algorithm. This algorithm is known for its ability to create high-quality tetrahedral elements, which are well-suited for complex geometries like the intricate arms of our drone frame. A finer mesh in critical areas ensures higher accuracy in the subsequent analysis.
Analyzing Static Loads with STAT_NON_LINE
Once the mesh was ready, we moved on to the static analysis. This aimed to understand how the drone frame would deform and where stress concentrations would occur under applied loads. In Code_Aster, the function block responsible for performing static, non-linear analysis is STAT_NON_LINE. While our analysis here was linear for simplicity, STAT_NON_LINE can handle complex material behaviors and large deformations. The top left image visually represents the results of this static analysis, with the color gradient indicating the magnitude of displacement under the applied load.
Unveiling Vibrational Modes with CALC_MODES
Understanding a structure’s natural frequencies and mode shapes is vital for predicting its dynamic behavior and preventing resonance. To achieve this, we performed a modal analysis using the CALC_MODES function in Code_Aster. This function extracts the eigenvalues (natural frequencies) and eigenvectors (mode shapes) of the structure. The top right and bottom right quadrants of the image illustrate two distinct mode shapes of the drone frame, with different colors representing varying amplitudes of vibration. Identifying these modes helps engineers design the control system and avoid operational frequencies that could lead to structural failure.
Harnessing Parallel Power for Faster Results
FEA, especially for complex models and analyses, can be computationally intensive. Code_Aster is capable of running in parallel, utilizing multiple processor cores to significantly reduce computation time. To enable parallel processing, you typically need to modify the command-line arguments or the execution script used to launch Code_Aster. This often involves specifying the number of processors to be used. For instance, you might use a command like as_run --ncpus <number_of_cores> ... to instruct Code_Aster to utilize the specified number of cores.
CloudHPC: Your Remote High-Performance Computing Solution
For projects demanding significant computational resources, local workstations may not always suffice. This is where platforms like CloudHPC prove invaluable. CloudHPC allowed us to rent powerful remote hardware resources equipped with a high number of cores. This drastically reduced the turnaround time for our FEA simulations. Instead of waiting hours or even days for the analyses to complete on a local machine, CloudHPC provided the necessary horsepower to obtain results much faster. This agility is crucial in iterative design processes, allowing engineers to explore more design options and optimize their products efficiently.
Conclusion
This FEA journey on a drone frame highlights the capabilities of open-source tools like SALOME for meshing and Code_Aster for static and modal analyses. By strategically utilizing functions like STAT_NON_LINE and CALC_MODES, we gained valuable insights into the structural behavior of our design. Furthermore, leveraging the parallel processing capabilities of Code_Aster on remote hardware provided by CloudHPC significantly accelerated our simulations, demonstrating the power of combining advanced simulation software with accessible high-performance computing resources. This approach empowers engineers to tackle complex structural challenges with greater efficiency and speed.
CloudHPC is a HPC provider to run engineering simulations on the cloud. CloudHPC provides from 1 to 224 vCPUs for each process in several configuration of HPC infrastructure - both multi-thread and multi-core. Current software ranges includes several CAE, CFD, FEA, FEM software among which OpenFOAM, FDS, Blender and several others.
New users benefit of a FREE trial of 300 vCPU/Hours to be used on the platform in order to test the platform, all each features and verify if it is suitable for their needs
