Yazar "Sharma, Vishal S." seçeneğine göre listele

Listeleniyor 1 - 3 / 3
  • Küçük Resim Yok
    Öğe
    Measuring Surface Characteristics in Sustainable Machining of Titanium Alloys Using Deep Learning-Based Image Processing
    (Ieee-Inst Electrical Electronics Engineers Inc, 2023) Ross, Nimel Sworna; Shibi, C. Sherin; Mustafa, Sithara Mohamed; Gupta, Munish Kumar; Korkmaz, Mehmet Erdi; Sharma, Vishal S.; Li, Z.
    A crucial method of maintenance in the manufacturing industry is machine vision-based fault diagnostics and condition monitoring of machine tools. The friction that occurs between the tool and the workpiece has a greater influence on the surface properties of the material. Effective problem diagnosis is necessary for machine systems to continue operations safely. Data-driven approaches have recently exhibited great promise for intelligent fault diagnosis. Unfortunately, the data collected under real-world conditions may be imbalanced, making diagnosis difficult. In dry, minimum quantity lubrication (MQL), and cryogenic circumstances, the method of failure detection of the proposed design is novel. The purpose of this interrogation is to evaluate the roughness profiles obtained from the machined surfaces and class separation. Markov transition field (MTF) is adopted to encode the surface profiles. In addition to this, conditional generative adversarial network (CGAN) for augmentation and bidirectional long-short term memory (BLSTM), multilayer perceptron (MLP), and 2-D-convolutional neural network (CNN) models are used for surface profile classification and correlation with process parameters. According to the study's finding, the 2-D-CNN was significantly more accurate than the models in predicting surface profiles, with an average accuracy of above 99.6% in both training and testing. In the limelight, the suggested approach can demonstrate to be quite useful for categorizing and proposing appropriate machining circumstances, specifically in situations with minimal data.
  • Küçük Resim Yok
    Öğe
    A new intelligent approach of surface roughness measurement in sustainable machining of AM-316L stainless steel with deep learning models
    (Elsevier Sci Ltd, 2024) Ross, Nimel Sworna; Mashinini, Peter Madindwa; Shibi, C. Sherin; Gupta, Munish Kumar; Korkmaz, Mehmet Erdi; Krolczyk, Grzegorz M.; Sharma, Vishal S.
    Due to the manufacturing sector ' s digitalization and ability to combine quality measurement and production data, machine learning and deep learning for quality assurance hold enormous potential. In this situation, industries may process data to inform data-driven estimates of product quality, thanks to predictive excellence. This research investigates the machinability of Laser Powder Bed Fusion (LPBF) - 316L stainless steel specimens, focusing on the impact of cutting parameters and cooling conditions (Dry, MQL, CO 2 and CO 2 + MQL) on surface roughness. The research employs advanced data augmentation techniques, incorporating TransGAN and multihead attention (MHA) based Alexnet model for surface imperfection classification. The results highlight the effectiveness of the proposed methodology in accurately classifying surface conditions and underscore the superior performance of the MHA-Alexnet algorithm compared to alternative models (Alexnet and AE-Alexnet). Overall, the study contributes valuable insights into optimizing machining parameters and cooling strategies for enhanced surface finish in additively manufactured alloys.
  • Küçük Resim Yok
    Öğe
    A novel approach of tool condition monitoring in sustainable machining of Ni alloy with transfer learning models
    (Springer, 2024) Ross, Nimel Sworna; Sheeba, Paul T. T.; Shibi, C. Sherin; Gupta, Munish Kumar; Korkmaz, Mehmet Erdi; Sharma, Vishal S.
    Cutting tool condition is crucial in metal cutting. In-process tool failures significantly influences the surface roughness, power consumption, and process endurance. Industries are interested in supervisory systems that anticipate the health of the tool. A methodology that utilizes the information to predict problems and to avoid failures must be embraced. In recent years, several machine learning-based predictive modelling strategies for estimating tool wear have been emerged. However, due to intricate tool wear mechanisms, doing so with limited datasets confronts difficulties under varying operating conditions. This article proposes the use of transfer learning technology to detect tool wear, especially flank wear under distinct cutting environments (dry, flood, MQL and cryogenic). In this study, the state of the cutting tool was determined using the pre-trained networks like AlexNet, VGG-16, ResNet, MobileNet, and Inception-V3. The best-performing network was recommended for tool condition monitoring, considering the effects of hyperparameters such as batch size, learning rate, solver, and train-test split ratio. In light of this, the recommended methodology may prove to be highly helpful for classifying and suggesting the suitable cutting conditions, especially under limited data situation. The transfer learning model with Inception-V3 is extremely useful for intelligent machining applications.