Hossein Afshari

Mr. Hossein Afshari

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
29 October 2024 in Applied Sciences
Verified
Journal Article
Open access
PDF file
Abolfazl FoorginejadAbolfazl Foorginejad
Abolfazl Foorginejad
SKSiamak Khatibi
Siamak Khatibi
HTHojjat Torshizi
Hojjat Torshizi
SESayyed Mohammad Emam
Sayyed Mohammad Emam
HAHossein Afshari
Hossein Afshari
Additive manufacturing encompasses technologies that produce three-dimensional computer-aided design (CAD) models through a layer-by-layer production process. Compared to traditional manufacturing methods, additive manufacturing technologies offer significant advantages in producing intricate components with minimal energy consumption, reduced raw material waste, and shortened production timelines. AM methods based on shielded gas welding have recently piqued the interest of researchers due to their high efficiency and cost-effectiveness in manufacturing critical components. However, one of the most formidable challenges in additive manufacturing methods based on shielded gas welding lies in the irregularity of weld bead height at different points, compromising the precision of components produced using these techniques. In this current research, we aimed to achieve uniform weld heights along the welding path by considering the most influential parameters on weld bead geometry and conducting experimental tests. Input parameters of the process, including nozzle angle, welding speed, wire speed, and voltage, were considered. Simultaneously, image processing and wavelet transform were employed to assess the uniformity of weld bead height. These parameters were applied to produce intricate parts after identifying optimal parameters that yielded the smoothest weld lines. According to the results, the appropriate bead for manufacturing the part was extracted. The results show that the smoothest bead line is achieved in 27 V as the highest level of voltage, at a 90° nozzle position and the maximum wire feed rate. Parts manufactured using this method across different layers exhibited no distortions, and the repeatability of production substantiated the high reliability of this approach for component manufacturing.

1 January 2026 in International Journal of Engineering
Verified
Journal Article
Open access
PDF file
Abolfazl FoorginejadAbolfazl Foorginejad
Abolfazl Foorginejad
KKkhalil khalili
khalil khalili
VAV. Abdollahi
V. Abdollahi
HAHossein Afshari
Hossein Afshari
Scaffold geometry plays a crucial role in determining its mechanical strength, as changes in shape can significantly impact its properties. Additionally, porosity, which varies with geometry, weakens the scaffold's mechanical performance. This study investigates the influence of scaffold geometry on mechanical properties and porosity in biomedical applications. Seven distinct geometries were designed using identical materials and fabricated through 3D printing. The scaffolds underwent compressive strength testing and finite element simulations to evaluate their load-bearing capacity and porosity. Among the designs, hexagonal and circular geometries demonstrated superior mechanical performance and controlled porosity. A total of 81 hexagonal and 27 circular scaffold samples were analyzed using Abaqus software. Initially, Response Surface Methodology (RSM) was employed to model the relationship between pressure and porosity, identifying the optimal design space with high predictive accuracy (R² > 96%). Then, a multi-objective optimization process using the Multi-Objective Particle Swarm Optimization (MOPSO) algorithm was implemented. The results revealed a Pareto front for each geometry, enabling the selection of scaffolds with specific load-bearing capacities and maximum porosity levels. Validation tests showed a mean error of 3.4% for circular geometries and 3.53% for hexagonal geometries, demonstrating the reliability of the simulation and optimization methods. This comprehensive approach integrates experimental, simulation, and optimization techniques, offering a robust framework for designing high-performance scaffolds tailored to biomedical needs.

27 October 2026 in Research Square
Verified
Preprint
Open access
PDF file
JAJavad Afshari
Javad Afshari
MAMahmoud Afshari
Mahmoud Afshari
HAHossein Afshari
Hossein Afshari
Data-driven models are increasingly used in petroleum engineering for tasks such as reservoir characterization, well-log interpretation, and production prediction. However, predictive performance may change when a model is applied to data from a different well, field, or operating condition. Differences between source and target datasets can therefore limit model transferability, yet this issue is commonly assessed only after model training. This study proposes a pre-training framework for quantifying source–target distribution shift and examining its relationship with subsequent zero-shot model transfer performance. The proposed approach uses predefined engineering features and Wasserstein-based distributional distances to construct a model-independent shift index that can be calculated before predictive model development. A prediction model is then trained exclusively on the source data and evaluated on the target data without retraining or adaptation. Pearson and Spearman correlation analyses are used to examine the relationship between the pre-training shift index and relative performance degradation. A controlled synthetic experiment is first used to verify the computational pipeline and methodological design. Across 24 synthetic well-transfer cases, the proposed indices captured different levels of source–target distribution shift, while the zero-shot model performance varied across target wells. None of the tested correlations reached the conventional 0.05 significance level, indicating that the present experiment provides preliminary rather than conclusive evidence of a systematic relationship between distribution shift and transfer degradation. Application to the real FORCE 2020 dataset is proposed as the next stage of validation. The framework provides a model-independent diagnostic that can be used to identify potentially challenging transfer settings before investing in model development or deployment.

27 October 2026 in Research Square
Verified
Preprint
Open access
PDF file
JAJavad Afshari
Javad Afshari
MAMahmoud Afshari
Mahmoud Afshari
HAHossein Afshari
Hossein Afshari
Federated learning (FL) has emerged as a promising approach for fault diagnosis in rotating machinery, particularly where raw data cannot be shared across machines or factories. However, differences in machine structure, operating conditions, and sensing systems introduce inter-machine domain shifts that degrade a global model on previously unseen machines. Existing studies mainly mitigate these shifts after model training, and few quantify their severity beforehand. This study investigates whether such severity can be measured before training. A Domain Shift Index (DSI) is defined as the data-volume-weighted sliced Wasserstein distance between model-independent vibration-feature distributions; it is computed from quantile summaries of shared random projections, so no raw samples are exchanged, and it is linked to a Lipschitz-type error bound whose constants are calibrated by quantile regression. The index is evaluated against six baseline discrepancy measures on 40 controlled synthetic machines (960 source–target cases) and on 16 domains of the public CWRU bearing benchmark (432 cases), using machine-level cluster bootstrap, Holm-corrected paired tests, and leave-one-machine-out calibration. The DSI is numerically stable (median coefficient of variation of about 2–3% with 200 projections) and moderately rank-correlated with zero-shot transfer error (Spearman ρ between 0.35 and 0.49 on synthetic data and between 0.30 and 0.49 on CWRU), but it is not superior to simple baselines such as MMD, and the calibrated bound, although it attains its nominal coverage, is loose and improves only marginally on a no-shift predictor. The index is therefore best regarded as a privacy-friendly screening dissimilarity rather than a quantitative predictor of accuracy loss; validation on industrial field data remains future work.

11 November 2026 in Journal of Advanced Manufacturing Systems
Verified
Journal Article
Closed access
PDF file
IAIshraga Galal-Eldin Abdalla Awad
Ishraga Galal-Eldin Abdalla Awad
MAMahmoud Afshari
Mahmoud Afshari
SASeyed Amirhossein Alavi
Seyed Amirhossein Alavi
HAHossein Afshari
Hossein Afshari
MSMohammad Reza Samadi
Mohammad Reza Samadi
Walid mahmoud ShewakhWalid mahmoud Shewakh
Walid mahmoud Shewakh