Browsing by Author "Sharma, M."

Now showing 1 - 9 of 9

Assessment of distributed generation source impact on electrical distribution system performance
(2010) Sharma, M.; Vittal, K.P.
The recent trends in electrical power distribution system operation and management are aimed at improving system conditions in order to render good service to the customer. The reforms in distribution sector have given major scope for employment of distributed generation (DG) resources which will boost the system performance. This paper proposes a heuristic technique for allocation of distribution generation source in a distribution system. The allocation is determined based on overall improvement in network performance parameters like reduction in system losses, improvement in voltage stability, improvement in voltage profile. The proposed Network Performance Enhancement Index (NPEI) along with the heuristic rules facilitate determination of feasible location and corresponding capacity of DG source. A Priority list is prepared with decreasing values of NPEI so that the designer can select most feasible location. The developed approach is tested with different test systems to ascertain its effectiveness. © 2010 AECE.
Deep Learning for Odor Prediction on Aroma-Chemical Blends
(American Chemical Society, 2025) Sisson, L.; Barsainyan, A.A.; Sharma, M.; Kumar, R.
The application of deep-learning techniques to aroma chemicals has resulted in models that surpass those of human experts in predicting olfactory qualities. However, public research in this field has been limited to predicting the qualities of individual molecules, whereas in industry, perfumers and food scientists are often more concerned with blends of multiple molecules. In this paper, we apply both established and novel approaches to a data set we compiled, which consists of labeled pairs of molecules. We present graph neural network models that accurately predict the olfactory qualities emerging from blends of aroma chemicals along with an analysis of how variations in model architecture can significantly impact predictive performance. © 2025 The Authors. Published by American Chemical Society.
Dense Sense: a novel approach utilizing electron density augmented machine learning paradigm to understand the complex odour landscape
(Royal Society of Chemistry, 2025) Saha, P.; Sharma, M.; Balaji, S.; Barsainyan, A.A.; Kumar, R.; Steuber, V.; Schmuker, M.
Olfaction is a complex process where multiple nasal receptors interact to detect specific odorant molecules. Elucidating structure–activity-relationships for odorants and their receptors remains difficult since crystallization of the odor receptors is an extremely difficult process. Therefore, ligand-based approaches that leverage machine learning remain the state of the art for predicting odorant properties for molecules, such as the graph neural network approach used by Lee et al. In this paper we explored how information from quantum mechanics (QM) could synergistically improve the results obtained with the graph neural network. Our findings underscore the possibility of this methodology in predicting odor perception directly from QM data, offering a novel approach in the machine learning space to understand olfaction. This journal is © The Royal Society of Chemistry, 2025
Deployment of Computer Vision Application on Edge Platform
(Institute of Electrical and Electronics Engineers Inc., 2021) Geetha, V.; Kiran, C.; Sharma, M.; Rakshith Kumar, J.
In our work, we propose a low cost device which will aid visually impaired people to understand what is in their surroundings without the requirement of internet. Current technology makes use of Cloud Architecture and would require internet to achieve this purpose. But these systems will not work in areas with poor internet connectivity. Edge platform built on Raspberry Pi powered with Intel Neural Compute Stick is used by us for this purpose. Multi Label Image Classification Deep Learning Model is trained in the cloud. It is later optimised and deployed on Edge Device which is Raspberry Pi. Setup also consists of PiCamera which will record the video and give it as input to deployed model. Model will describe the items present in video, basically describing the surroundings. The output is in the form of audio which is played through speakers, thus enabling visually impaired people to understand their surroundings without the requirement of internet. Deployment of popular Machine Learning and Deep Learning Models is also examined in the edge device and a comprehensive performance evaluation is performed. Â© 2021 IEEE.
Effect of electrolyte temperature on the formation of highly ordered nanoporous alumina template
(Elsevier B.V., 2016) Boominathasellarajan, B.; Sharma, M.; Ghosh, S.K.; Nagaraja, H.S.; Barshilia, H.C.; Chowdhury, P.
In this work, we present a systematic influence of electrolyte temperature along with anodizing potential on the pore parameters during two-step anodization of Al in H2SO4 electrolyte. Top surface morphology of the nanoporous templates was examined with the help of field emission scanning electron microscope and atomic force microscope. Three-dimensional (3D) Fast Fourier Transform (FFT) image analysis was then employed to quantify pore regularity and pore periodicity as a function of both the bath temperature (1-15 °C) and the anodic potential (15-25 V). A highest pore regularity ratio of 5 × 108 was obtained at 3°C and 25 V with a pore diameter of 32 ± 3 nm and inter-pore distance of 65 nm. With further increase in temperature, the pore regularity ratio was found to decrease drastically. It was found that higher temperature favored the dissolution of compact aluminum oxide layer isotropically along the pore length. This process in effect enhanced the pore size, growth rate, and template top surface roughness without affecting much inter-pore distance. Self-ordering of the pores was found to improve with increasing anodizing potential with a critical influence of the current density along with inter-pore distance. The mechanism of pore growth was discussed in terms of temperature-dependent activation energy controlled dissolution of aluminum. The typical activation energy evaluated at 25 V was 72.8 kJ/mol at 3°C. © 2015 Elsevier Inc. All rights reserved.
Green intelligence for cloud data centers
(2016) Karthik, C.; Sharma, M.; Maurya, K.; Chandrasekaran, K.
In this paper the problem of energy consumption by large data centers has been tackled. Power consumption is major problem from both economic and environmental point of view. One of the main components of data centers is virtualization. We have addressed the problem of Virtual Machine (VM) consolidation in the data center servers using the technique of Bin Completion. Bin Completion is basically an artificial intelligence based algorithm used for bin packing problem. We have scaled up and modified the algorithm to fit our problem statement of VM consolidation and analysed the results obtained against Best Fit algorithm. After that we did an extensive study of the application of machine learning algorithms for the purpose of CPU utilisation prediction and analysed its effects on the overall energy consumption of a data center as well as the SLA violations. � 2016 IEEE.
Green intelligence for cloud data centers
(Institute of Electrical and Electronics Engineers Inc., 2016) Karthik, C.; Sharma, M.; Maurya, K.; Chandrasekaran, K.
In this paper the problem of energy consumption by large data centers has been tackled. Power consumption is major problem from both economic and environmental point of view. One of the main components of data centers is virtualization. We have addressed the problem of Virtual Machine (VM) consolidation in the data center servers using the technique of Bin Completion. Bin Completion is basically an artificial intelligence based algorithm used for bin packing problem. We have scaled up and modified the algorithm to fit our problem statement of VM consolidation and analysed the results obtained against Best Fit algorithm. After that we did an extensive study of the application of machine learning algorithms for the purpose of CPU utilisation prediction and analysed its effects on the overall energy consumption of a data center as well as the SLA violations. Â© 2016 IEEE.
Hardware Accelerator for Object Detection using Tiny YOLO-v3
(Institute of Electrical and Electronics Engineers Inc., 2021) Sharma, M.; Rahul, R.; Madhusudan, S.; Deepu, S.P.; Sumam David, S.
For applications that require object detection to be performed in real-time, this paper presents a custom hardware accelerator, implementing state of the art Tiny YOLO-v3 algorithm. The proposed architecture achieves a reasonable tradeoff between the speed of computation (measured in frames per second or FPS) and the hardware resources required. Each CNN layer is pipelined and parameterized to make the complete design re-configurable. The proposed hardware accelerator was synthesized using the SCL(Semi-Conductor Laboratory, India) 180 nm CMOS process and also using Vivado Xilinx software with Virtex Ultrascale+ FPGA as the target device. The pipelined architecture, along with other architectural novelties, provided a higher frame-rate of 32.1 FPS and a performance of 166.4 GOPS at 200 MHz clock frequency. Â© 2021 IEEE.
Hyperlocal referent systems in television advertising
(2011) Sharma, M.; Philip, P.J.
We identify and analyze the creation of a hyperlocal space and referent system in contemporary Indian television advertising to mediate the transmission of commodity messages by advertisers. The hyperlocal is defined as an imagined heterotopic space for cognition and interpretation. A similarity between techniques used in construction of hyperlocal space and Indian folk theater form of Nautanki is identified. A discussion of its linguistic, narrative facets, construction of identity and meaning of consumption within the hyperlocal is presented. Exemplar advertisements for Chlormint, a consumer product are analysed semiologically with a discussion of various dialectics that present a semiotic and qualititative study of hyperlocal space in Chlormint advertisements. © Common Ground, Manjula Sharma, P. J. Philip, All Rights Reserved.