No abstract

For effective bead deposition based additive manufacturing (AM) processes such as directed energy deposition, the final mechanical and physical properties should be predicted in tandem with the bead geometry characteristics. Experimental approaches to investigate the final geometry and the mechanical properties are costly, and simulation solutions are timeconsuming. Alternative artificial intelligent (AI) systems are explored as they are a powerful approach to predict such properties. In the present study, the geometrical properties, as well as the mechanical properties (residual stress and hardness) for single bead clads are investigated.Experimental data is used to calibrate multi-physics finite element models, and both data sets are used to seed the AI models. The adaptive neuro-fuzzy inference system (ANFIS) and a feed-forward back-propagation artificial neural network (ANN) system are utilized to explore their effectiveness in the 1D (discrete values), 2D (cross sections), and 3D (complete bead) domains. The prediction results are evaluated using the mean relative error measure. The ANFIS predictions are more precise than those from the ANN for the 1D and 2D domains, but the ANN had less error for the 3D scenario. These models are capable of predicting the geometrical and the mechanical properties values very well, including capturing the mechanical properties in transient regions; however, this research should be extended for multibead scenarios before a conclusive 'best approach' strategy can be determined.

For effective bead deposition based additive manufacturing (AM) processes such as directed energy deposition, the final mechanical and physical properties should be predicted in tandem with the bead geometry characteristics. Experimental approaches to investigate the final geometry and the mechanical properties are costly, and simulation solutions are time-consuming. Alternative artificial intelligent (AI) systems are explored as they are a powerful approach to predict such properties. In the present study, the geometrical properties, as well as the mechanical properties (residual stress and hardness) for single bead clads are investigated. Experimental data is used to calibrate multi-physics finite element models, and both data sets are used to seed the AI models. The adaptive neuro-fuzzy inference system (ANFIS) and a feed-forward back-propagation artificial neural network (ANN) system are utilized to explore their effectiveness in the 1D (discrete values), 2D (cross sections), and 3D (complete bead) domains. The prediction results are evaluated using the mean relative error measure. The ANFIS predictions are more precise than those from the ANN for the 1D and 2D domains, but the ANN had less error for the 3D scenario. These models are capable of predicting the geometrical and the mechanical properties values very well, including capturing the mechanical properties in transient regions; however, this research should be extended for multi- bead scenarios before a conclusive ‘best approach’ strategy can be determined.

Salt gradient solar ponds are the ponds in which due to existence of saline and salt gradient layers, lower layers are denser and avoid the natural convection phenomenon to occur so that solar radiation energy can be stored in the lowest zone. In this study, one-dimensional (1D) and two-dimensional (2D) numerical approaches have been implemented to simulate unsteady buoyancy-driven flow of solar ponds. In 1D method, the pond has been investigated in terms of the layers thicknesses so that the variation of temperature is calculated by energy conservation equation. The formulized radiation term was used as energy source term in energy equation. The results of 1D approach were validated with an experimental study and then optimization was carried out to determine the maximum thermal efficiency for an interval of layers height. Since the stability of the solar pond cannot be determined by 1D simulation, a 2D approach was considered to show the stability for different nonconvective zone (NCZ) heights and different salt gradients. In 2D study, in order to investigate hydrodynamic and thermal behavior of saltwater fluid, a numerical approach was used to simulate temperature gradients throughout the pond. The results of 2D numerical method are validated with an experimental data. The effect of linear and nonlinear salt gradient was considered.

No abstract