Development of efficient numerical method for distortion prediction of laser additive manufactured parts

Part distortion is a technical bottleneck in the field of laser solid forming additive manufacturing. Finite element modelling has shown great power in analysis and predicting thermal distortion during laser additive manufacturing (AM) process. However, as the global size of the manufactured component increases, the conventional numerical method appears limited due to the long computation time. In this paper, the temperature distribution and evolution of a Ti-6Al-4V thin wall during AM process were investigated firstly via transient heat transfer analysis. “Quasi-steady state” characteristic of the temperature distribution was observed after depositing several layers. Base on this, an efficient equivalent temperature field (ETF) method was developed to predict thermal distortion by extracting the quasi-steady temperature field and applying it as thermal boundary during mechanical analysis. The developed ETF method was validated by the good agreement in the predicted distortion distribution pattern and magnitude compared with that predicted by the conventional move heat source numerical method. The developed ETF method in this paper significantly saved computation time by above 90% during mechanical analysis. Furthermore, the distortion of laser additive manufactured thin wall with 266 layers was successfully predicted by the ETF method within several hours. The maximum deviation is 29.3% compared with the experimental results. The proposed method provides the possibility to predict distortion for large scale AM parts, which may have the potential application in engineering.