Fast-to-run Approximate Thermal Models for Predicting and Optimizing Melt Strategies

Numerical modelling methods (e.g. finite element, computational fluid dynamics) can provide approximate descriptions of temperature fields in laser or electron-beam melting processes useful for build strategy planning. The high computational costs at part-scale, however, makes these methods unsuitable for predictions at (or near) part scale. Alternative methods such as semi-analytical solutions based on a moving heat source reduce the computational expense but at the cost of unrealistic assumptions. In such approaches, non-linear factors such as radiation, temperature-dependent physical properties and latent heat can’t be included in a formal treatment though these have a significant impact on the predicted thermal history. Here, we present an attempt at overcoming these challenges with approximate methods that can provide improved predictions relative to closed form solutions and at much less computational cost (while retaining resolution) compared to full FE or CFD methods. The method that is developed is illustrated against a number of test cases and experimental trials.