Welding, mechanical properties and microstructure of different grades of austenitic stainless steels: A review

Singh, Sahadevuda Sai Rakesh; Praneeth, Rayala Venkata; Sankalp, Valluri Sai; Sashank, S. Sravan; Karthikeyan, R.

doi:10.1016/j.matpr.2022.04.425

Cited by 3 publications

(3 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Comparison with other possible defense strategy. We have conducted additional comparison experiments with more recent baselines, including [18,31,39] under a label attack scenario (Scenario 1). FL-WBC is a client-side defense, while the others are server-side defenses.…”

Section: A2 Further Results On Defense Performancementioning

confidence: 99%

See 1 more Smart Citation

FedDefender: Client-Side Attack-Tolerant Federated Learning

Park

Han

et al. 2023

Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

View full text Add to dashboard Cite

Federated learning enables learning from decentralized data sources without compromising privacy, which makes it a crucial technique. However, it is vulnerable to model poisoning attacks, where malicious clients interfere with the training process. Previous defense mechanisms have focused on the server-side by using careful model aggregation, but this may not be effective when the data is not identically distributed or when attackers can access the information of benign clients. In this paper, we propose a new defense mechanism that focuses on the client-side, called FedDefender, to help benign clients train robust local models and avoid the adverse impact of malicious model updates from attackers, even when a server-side defense cannot identify or remove adversaries. Our method consists of two main components: (1) attack-tolerant local meta update and (2) attack-tolerant global knowledge distillation. These components are used to find noise-resilient model parameters while accurately extracting knowledge from a potentially corrupted global model. Our client-side defense strategy has a flexible structure and can work in conjunction with any existing server-side strategies. Evaluations of real-world scenarios across multiple datasets show that the proposed method enhances the robustness of federated learning against model poisoning attacks. CCS CONCEPTS• Security and privacy → Software and application security; Intrusion/anomaly detection and malware mitigation; • Computing methodologies → Artificial intelligence.

show abstract

Section: A2 Further Results On Defense Performancementioning

confidence: 99%

“…This global knowledge distillation loss is optimized in conjunction with original cross-entropy loss to train the FedDefender (Eq. 18).…”

Section: Attack-tolerant Global Knowledge Distillationmentioning

confidence: 99%

FedDefender: Client-Side Attack-Tolerant Federated Learning

Park

Han

et al. 2023

Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

View full text Add to dashboard Cite

show abstract

“…6 Despite facing challenges due to differing properties, fusion welding of these alloys yields weld deposits with intermediate chemical compositions, impacting microstructure, corrosion resistance, and mechanical properties. 7 In both candidate materials subjected to the thermal cycle of welding, the introduced alloying content makes the microstructure more prone to significant segregation. This segregation significantly compromises both mechanical and corrosion properties, leading to the formation of a fully austenitic matrix.…”

Section: Introductionmentioning

confidence: 99%

Analyzing microstructural, residual stress, and mechanical characteristics in dissimilar welds of Inconel 59 and AISI 904L through double-pulsed gas metal arc welding

Muthukumaran,

Arulmurugan,

Manikandan

2024

Proceedings of the Institution of Mechanical Engineers, Part L:

View full text Add to dashboard Cite

In this research, dissimilar welding of the nickel-based superalloy Inconel 59 and austenitic stainless steel AISI 904L steel was successfully done with double-pulsed gas metal arc welding (DP-GMAW) using the filler ERNiCrMo-13. Macroscopic analysis showed a defect-free weld with an aspect ratio of 2.4, attributed to the efficient heat delivery (0.642 kJ/mm) in DP-GMAW. The microstructure of the weld joint was examined through both optical microscopy and scanning electron microscopy. Examination of the microstructure of the weld interface exposed grain coarsening near the heat-affected zone on the AISI 904L side. In contrast, the weld fusion zone contained fine equiaxed dendrites. Microsegregation and alloying element migration across the entire weld area were assessed through energy dispersive X-ray spectroscopy point analysis and line mapping, revealing controlled microsegregation across the region. X-ray diffraction (XRD) analysis was used to ascertain the average grain size (38.41 nm) and secondary phase composition. The XRD-cosα method was used to assess the residual stress of the weld joint, revealing tensile residual stress in the weld zone and compressive residual stress in other regions. Tensile, Charpy impact, and Vickers hardness tests were conducted in the study to evaluate the strength, toughness, and hardness of a dissimilar weld. The impact of two cryogenic treatments, namely shallow and deep, on tensile and toughness characteristics was investigated in this study. The cryogenic treatments significantly enhanced the mechanical properties of weld samples compared to their as-welded condition, leading to an 11.5% increase in tensile strength and a 10% improvement in toughness.

show abstract