No abstract

This paper proposes a computationally efficient methodology to predict the damage progression in solder contacts of electronic components using temperature-time curves. For this purpose, two machine learning algorithms, a Multilayer Perceptron and a Long Short-Term Memory network, are trained and compared with respect to their prediction accuracy and the required amount of training data. The training is performed using synthetic, normally distributed data that is realistic for automotive applications. A finite element model of a simple bipolar chip resistor in surface mount technology configuration is used to numerically compute the synthetic data. As a result, both machine learning algorithms show a relevant accuracy for the prediction of accumulated creep strains. With a training data length of 350 hours (12.5 % of the available training data), both models show a constantly good fitting performance of 𝑅² of 0.72 for the Multilayer Perceptron and 𝑅² of 0.87 for the Long Short-Term Memory network. The prediction errors of the accumulated creep strains are less than 10 % with an amount of 350 hours training data and decreases to less than 5 % when using further data. Therefore, both approaches are promising for the lifetime prediction directly on the electronic device.

In this contribution, we apply adaptive isogeometric analysis to a phase-field model for topology optimization. To increase the efficiency of the computation, we perform local mesh refinement and coarsening between the time increments of the simulation. To provide a trial solution for the iterative solver of the next increment, state variables have to be transferred from the old to the new mesh. We therefore compare the application of a discrete least squares fit with an L 2-projection in well known two-and three-dimensional benchmarks and analyze the influence of these methods on the quality of the results.

In this contribution, we apply adaptive isogeometric analysis to a diffuse interface model for topology optimization. First, the influence of refinement and coarsening parameters on the optimization procedure are evaluated and discussed on a two-dimensional problem and a possible workflow to convert smooth isogeometric solutions into 3D printed products is described. Second, to assess the required numerical accuracy of the proposed simulation framework, numerical results obtained adopting different stopping criteria are experimentally evaluated for a three-dimensional benchmark problem.