Now showing 1 - 2 of 2
No Thumbnail Available
Publication

Relative density prognosis for directed energy deposition with the help of artificial neural networks

2021 , Marko, A. , Schafner, A. , Raute, J. , Rethmeier, M.

Additive manufacturing, and therefore directed energy deposition, is gaining more and more interest from industrial users. However, quality assurance for the components produced is still a challenge. Machine learning, especially using artificial neuronal networks, is a potential method for ensuring a high-quality standard. Based on process parameters and monitoring data, part quality can be predicted. A further advantage is the ability to constantly learn and adopt to slight process changes. First tests using artificial neural networks focus on the prediction of track geometry. The results show that even a small data set is enough to provide high accuracy in the predictions. In this work, an artificial neural network for the predictive analysis of relative density in laser powder cladding has been developed. A central composite experimental design is used to generate 19 data sets. Input variables are laser power, feed rate and powder mass flow. Cubes are built up where density is considered as a target value. Several neural networks are trained and evaluated with these data sets. Different topologies and initial weights are considered. The best network reaches a confidence level of around 90 % for the prediction of relative density based on the process parameters. Finally, the optimization of the generalization performance is investigated. To this purpose, methods of variation in error limit as well as cross-validation are applied. In this way, density is predictable by an artificial neural network with an accuracy of about 95 %.

No Thumbnail Available
Publication

Mechanical properties characterization of resistance spot welded DP1000 steel under uniaxial tensile tests

2019 , Javaheri, E. , Pittner, A. , Graf, B. , Rethmeier, M.

Resistance spot welding (RSW) is widely used in the automotive industry as the main joining method. Generally, an automotive body contains around 2000 to 5000 spot welds. Therefore, it is of decisive importance to characterize the mechanical properties of these areas for the further optimization and improvement of an automotive body structure. The present paper aims to introduce a novel method to investigate the mechanical properties and microstructure of the resistance spot weldment of DP1000 sheet steel. In this method, the microstructure of RSW of two sheets was reproduced on one sheet and on a bigger area by changing of the welding parameters, e. g. welding current, welding time, electrode force and type. Then, tensile tests in combination with digital image correlation (DIC) measurement were performed on the notched tensile specimens to determine the mechanical properties of the weld metal. The notch must be made on the welded tensile specimen to force the fracture and elongation on the weld metal, enabling the characterization of its properties. Additionally, the parameters of a nonlinear isotropic material model can be obtained and verified by the simulation of the tensile specimens. The parameters obtained show that the strength of DP1000 steel and the velocity of dislocations for reaching the maximum value of strain hardening, are significantly increased after RSW. The effect of sample geometry and microstructural inhomogeneity of the welded joint on the constitutive property of the weld metal are presented and discussed.