Design and development of a measuring method to determine the strain state during hot crack initiation

• Applicant: Dr.-Ing. Andrey Gumenyuk
• Subject Area: Joining and Separation Technology
• Term: since 2021
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 465316565

Over the last few decades the laser beam welding technology developed significantly and was established as an efficient and economic tool in industry. Solidification cracking as well as the weldability of materials is still for many years a highly contentious issue, particularly regarding the causes of the hot crack formation. The aim of this proposal is development of a 2-dimentional optical technique to determine the local critical strain and local critical strain rate required for solidification crack formation during laser beam welding. To realize this aim, firstly, the video quality of the weld pool and its vicinity, particularly for a solidified in a hot crack susceptible zone during the welding process material must be optimised with choice of the right lighting and camera sensor. After that, the most relevant measurement algorithms should be implemented in the Open Source Computer Vision Library (OpenCV) to determine the 2D transient strain at the surface of tensile loaded metallic specimens. The results will be compared to each other as well as to the reference measurements from the ARAMIS system (Digital Image Correlation system). The measurement principle and evaluation algorithm should perform with sufficient accuracy, low calculation time and the capability of evaluation without any special surface preparation. The most suitable algorithm corresponding to these requirements will be selected for further development. The next step is the development of the selected algorithm to obtain the displacements and the strains in the mushy-zone i.e. in the critical temperature range behind the weld pool. The challenge is: How can the strain evaluation perform in defined region in moving images sequences. Two concepts will be tested to overcome that challenge. Afterwards, the measurement algorithm must be developed to be able to analysis the strain in realtime. Several factors as the image resolution, the minimum required frame rate, the used filters functions and the requirements for the hardware platform should be optimised to reduce the computational cost of the algorithm. Mainly due to the fact that there is no measuring technique that allows direct determination of displacements or strains next to the weld, the result obtained by the developed algorithm will be validated using the FEM model. Finally, after the definition of hardware requirements for the measuring system as well as validation of the measurements, the algorithm must be modified in order to enable automatic evaluation of the critical strain conditions, which could potentially lead to hot cracking in the material.

• DFG Programme: Research Grants
• International Connection: Russia
• Co-Investigator: Professor Dr.-Ing. Michael Rethmeier
• Cooperation Partner: Privatdozent Sergey Volvenko, until 3/2022