Syamak Pazireh

This review explores the evolution and current state of machine learning (ML) and artificial intelligence (AI) applications in direct energy deposition (DED) and wire arc additive manufacturing (WAAM) processes. A Python-based automated search script was developed to systematically retrieve relevant literature using the Crossref API, yielding around 370 papers published between 2010 and July 2025. The study identifies significant growth in ML-related DED research starting in 2020, with increasing adoption of advanced techniques such as deep learning, fuzzy logic, and hybrid physics-informed models. A year-by-year trend analysis is presented, and a comprehensive categorization of the literature is provided to highlight dominant application areas, including process optimization, real-time monitoring, defect detection, and melt pool prediction. Key challenges, such as limited closed-loop control, lack of generalization across systems, and insufficient modeling of deposition-location effects, are discussed. Finally, future research directions are outlined, emphasizing the need for integrated thermo-mechanical models, uncertainty quantification, and adaptive control strategies. This review serves as a resource for researchers aiming to advance intelligent control and predictive modeling in DED-based additive manufacturing.

Purpose: This paper aims to investigate the use of the element-birth-and-death technique within the finite element method (FEM) to analyze direct energy deposition (DED) processes, an additive manufacturing (AM) technique. The focus is on understanding the interaction between different geometries and toolpaths in multi-layer depositions and establishing a foundation for generating optimal tool paths. The long-term goal of this research is optimizing thermal and structural performance, minimizing residual stress and improving overall part quality. Design/methodology/approach: This study integrates DED toolpath vectors from process planning software into ANSYS to conduct thermal and structural analyses. This novel integration of design and process simulation enables a more accurate prediction of part quality. Four geometries and three toolpath strategies were analyzed, requiring the computational mesh domain to adapt to the toolpath characteristics. Experimental measurements validated the model, and the influence of dwell time intercooling on residual stress formation was explored. The research also compared the effects of various toolpath strategies and deposition patterns on the mechanical properties of the parts, highlighting the importance of toolpath generation for subsequent process certification and qualification activities. Findings: The results show that shorter beads with longer dwell times lead to better part quality compared to longer beads with shorter laser-off intervals. A transverse deposition with identical dwell times produced the best part quality, despite requiring longer build times. In addition, frequent laser-off cycles were less effective at reducing high residual stress areas but resulted in lower average tensile stresses at mid-layers, though these cycles pose challenges for powder-based systems. This analysis helps in benchmarking and enhancing DED as an AM process. Originality/value: This study offers a novel method of integrating DED toolpath characteristics into FEM-based simulations, providing valuable insights into the thermal and mechanical behavior of parts under various toolpath and geometry configurations. The research contributes to advancing the design-for-AM paradigm by demonstrating the critical role of thermal management and toolpath directionality in minimizing residual stress. This foundational work will support further research into more complex geometries and hybrid manufacturing techniques, aiding in the industrialization and optimization of DED processes.

This study examines the coupled effects of geometry type and toolpath patterns on residual stresses and strains in parts manufactured using laser-based Directed Energy Deposition (DED). Finite Element Method (FEM) simulations were conducted for four distinct geometries, each subjected to three different tool-path strategies: one-way longitudinal, longitudinal zigzag, and one-way transverse. The plastic strain is highly correlated with the maximum temperature within the deposition. The longer walls also gain higher strains in the deposition direction. The highest tensile strains were observed in the upper layers of the structures, while the mid-layers exhibited the maximum principal residual stresses. Additionally, thin walls are more sensitive to deposition patterns than thick walls due to limited heat dissipation. These insights provide a foundation for optimizing toolpath strategies in laser-based DED processes, aiming at improved mechanical performance and reduced distortion in complex geometries.

The purpose of this study is to investigate the effects of toolpath patterns, geometry types, and layering effects on the mechanical properties of parts manufactured by direct energy deposition (DED) additive manufacturing using data analysis and machine learning methods. A total of twelve case studies were conducted, involving four distinct geometries, each paired with three different toolpath patterns based on finite element method (FEM) simulations. These simulations focused on residual stresses, strains, and maximum principal stresses at various nodes. A comprehensive analysis was performed using a linear mixed-effects (LME) model, principal component analysis (PCA), and self-organizing map (SOM) clustering. The LME model quantified the contributions of geometry, toolpath, and layer number to mechanical properties, while PCA identified key variables with high variance. SOM clustering was used to classify the data, revealing patterns related to stress and strain distributions across different geometries and toolpaths. In conclusion, LME, PCA, and SOM offer valuable insights into the final mechanical properties of DED-fabricated parts.

Despite advances in computational power, the cost of time-accurate flows in axial compressor and fan stages with spatially non-uniform inflow is still too high for design-stage use in industry. Body force modeling reduces the computation time to practical levels, mainly by reducing the problem to a steady one. These computations are important to determine efficiency penalties associated with non-uniform inflows. Previous studies of body force methods have, in most cases, relied on computations with the presence of the blades to calibrate loss models. In some recent studies, uncalibrated models have been used, but such models can drop off in accuracy at conditions where separation would occur on the blade surfaces. In this paper, a neural-network-based loss model introduced in a recent paper by the authors is implemented for NASA rotor 67 for both uniform and non-uniform inflow conditions. For uniform inflow, the spanwise trend of entropy variation is generally captured with the new body force model. Although there are discrepancies at some span fractions, the present model generally predicts the compressor’s isentropic efficiency to within 3% compared to bladed Reynolds-averaged Navier–Stokes simulations. For non-uniform inflow, we consider a stagnation pressure profile representative of boundary layer ingestion. The results show that the region of maximum entropy generation is captured by the present model and the prediction of isentropic efficiency penalty due to the non-uniform inflow is only 0.2 points less than that determined from bladed computations.