Browsing by Author "Upadhyay, Ram Krishna"

Now showing 1 - 3 of 3
  • No Thumbnail Available
    Item
    Deep learning for object recognition and defect analysis in additive manufacturing
    (Discover, 2025-10-15) Thakur, Chandrbhan; Mishra, Santosh Kr.; Jose, S. Silva Sajin.; Singh, Purushottam Kumar; Upadhyay, Ram Krishna
    Additive manufacturing (AM) offers design flexibility and shorter lead times in manufacturing newer products. The layer by layer manufacturing approach often introduces defects that can deteroriate part performance, which require reliable inspection to detect quality. Manual inspection often followed in industries are slow and error prone, which thrusts for the need to automate inspection. This work proposes a modular deep learning framework for objection detection and defect analysis components manufactured by fused deposition modelling. A real-world dataset of 950 images from additively manufactured elevator maintainance components, is taken to test the model and evaluate performance.YOLOv12 and Faster R-CNN are used for object detection, CNN for binary defect classification and MobileNet to classify both defect severity(Levels 0–3) and defect type. The model successfully predicted, whether an object was defective, determined its defect severity level, and correctly labeled the defect type with greater accuracy. YOLOv12 yielded slightly better results in comparison to Faster R-CNN for object detection with a precision of around 93.6% and recall of 95.6%. The model accurately identified whether an object was defective, determined its defect severity level, and correctly labeled the defect type with high precision.
  • No Thumbnail Available
    Item
    Effect of Scanning Strategy on the Microstructure and Load-Bearing Characteristics of Additive Manufactured Parts
    (Multidisciplinary Digital Publishing Institute (MDPI), 2024-07-05) Sajin Jose, S. Silva; Mishra, Santosh Kr.; Upadhyay, Ram Krishna
    Additive manufacturing has witnessed significant growth in recent years, revolutionizing the automotive and aerospace industries amongst others. Despite the use of additive manufacturing for creating complex geometries and reducing material consumption, there is a critical need to enhance the mechanical properties of manufactured parts to broaden their industrial applications. In this work, AISI 316L stainless steel is used to fabricate parts using three different strategies of the additively manufactured Laser Powder Bed Fusion (LPBF) technique, i.e., continuous, alternate, and island. This study aims to identify methods to optimize grain orientation and compaction support provided to the material under load, which influence the frictional and wear properties of the manufactured parts. The load-bearing capacity is evaluated by measuring the frictional and wear properties. The wear patch track is also examined to establish the physical mechanisms at the surface interface that lead to the smooth transition in response to the load. Grain orientation is compared across different strategies using Electron Backscatter Diffraction (EBSD) maps, and the influence of surface roughness on sliding behavior is also evaluated. The results demonstrate that the island scanning strategy yields the best performance for load-bearing applications, exhibiting superior grain orientation and hardness in the additively manufactured parts.
  • No Thumbnail Available
    Item
    Prediction of irregularities and study of mechanical properties in laser shock peened nickel alloy
    (Discover, 2025-02-12) Jose, S. Silva Sajin; Mishra, Santosh Kr.; Upadhyay, Ram Krishna; Muthukannan, Duraiselvam
    Nimonic 263, a high-performance Ni-based super alloy is known for its excellent resistance to creep, corrosion, and oxidation, in addition to surface stability at higher temperatures. The present study investigates the impact of Laser Shock Peening (LSP) on the microstructure and mechanical properties of Nimonic 263. Although LSP is a promising technique for extending the fatigue life of turbine components and other high-temperature applications, the choice of crucial parameters is important in order to avoid the negative implications including surface irregularities and reduced stress benefits. The research aims to analyze how different laser parameters, such as laser beam diameter and energy, affect surface compressive residual stresses, microhardness, surface roughness, and microstructure. Defect detection is also performed on the collected scanning electron microscopy (SEM) images using machine learning (ML) models. With an increase in energy density, there is an improvement in the microhardness and surface roughness. XGBoost is found to be the best-performing algorithm for defect prediction in the SEM images, among the ML models with an accuracy value of 0.95.