Search

The main aim of the work is to increase the energy efficiency analysis in photovoltaic systems through solar PV plants by comparing two PV systems Non-Shadow effect and Shadow effect using Photovoltaic systems to minimize the panel array loss. Each set contains 12 samples and the total samples obtained is 24 it taken from Gpower value is 80%. By using Non-Shadow effect, the performance ratio is 82% per year, and the same for Shadow effect the performance ratio is 56%. The significance is 0.039 (p<0.05) by analysis the two group there are statistically significant. The efficiency of the proposed Non-Shadow effect has performed better than the Shadow effect. [ABSTRACT FROM AUTHOR]

The pupose of this paper is to analyze and maximize energy yield of the novel Rooftop Photovoltaic system compared with the Ground-mounted Photovoltaic system by limiting the ohmic wiring loss by using the PVsyst software. For the innovative Rooftop PV system, 12 samples—that is, 12 months of a year—were acquired by running simulations in the programme. The final outcome is the total amount of energy generated annually. To collect 12 more samples for a ground-mounted photovoltaic system, the same procedures are performed. Gpower is taken as 0.8. A new rooftop photovoltaic system generates 3832.0 kWh of electricity, while a ground-mounted photovoltaic system generates 3551.4 kWh. A statistically significant value of 0.026 (p<0.05) has been observed. It has been shown that when compared to ground-mounted PV, the novel rooftop PV technology offers a higher energy output, which improves efficiency. [ABSTRACT FROM AUTHOR]

The main objective of this research is to improve the accuracy of performing the novel session time analysis for various Ecommerce customer orders using supply chain management in ordering Ecommerce products using Linear Regression in comparison with Lasso Regression. Materials and Methods: Linear Regression algorithm with sample size=154 and Lasso Regression algorithm with sample size=154 with G power (value=0.8) were evaluated many times to forecast the accuracy percentage. Results: Linear Regression has better accuracy (85.62%) when compared to Lasso Regression (62.47%). The statistical significance difference (two-tailed) is 0.047 (p<0.05). Conclusion: Linear Regression algorithm performed significantly better than the Lasso Regression algorithm. [ABSTRACT FROM AUTHOR]

In this manuscript, a model is proposed for the Queuing issue with things which changes straightly concerning time. The idea of standby server and set up time is thought of and consolidated in this model. In addition, worker interference is an unavoidable wonder in Queuing framework, which must be sifted through promptly by making the worker to get into the repair procedure of three stages. The lining model characterized here is illuminated by notable supplementary variable strategy a Queuing approach of operational examination. The solution procedure of proposed Queuing model is delineated by notable genuine correspondence WhatsApp server model. A convexity check of the execution proportions of the lining model is performed by plotting a two-dimensional chart. The sensitivity trial of the proposed model is performed to examine the impact of changing the least just as the most sensitive parameters in the proposed Queuing model. Some graphical portrayals are developed to talk about the consequences and results so acquired for a decision of different parameters. [ABSTRACT FROM AUTHOR]

Improved quality of service is provided by the Internet of Things (IoT) devices deployed over IoT networks that form a large part of the smart city environment. At the network's edges, IoT devices are provided with computing abilities through a suitable and powerful paradigm termed as edge computing, which makes significant contributions to the distributed and large scale IoT network features. For the purpose of execution, in edge computing, edge computing units (ECUs) host the IoT services lessening the burden of bandwidth while providing low latency. Improving the overall execution performance of an ECU in terms of power consumption, load balance level and resource utilization is challenging. However, these parameters are essential in the service placement of IoT devices to ensure leakage of privacy information. A trust-oriented IoT service placement (TSP) scheme is presented in this paper to tackle the aforementioned challenges during edge computing in the IoT networks of the smart city environments. The privacy preservation and execution performance metrics tradeoffs are balanced by the placement strategies by enhancing the strength Pareto evolutionary algorithm. Out of the several placement strategies obtained, the optimal strategy is identified using multicriteria decision making and technique for order preference by similarity to ideal solution schemes. Further, the TSP is evaluated in terms of its reliability and efficiency through systematic experiments. [ABSTRACT FROM AUTHOR]

In agriculture field, recent research works mainly focus on the recognition and classification of plant leaf diseases from images. Image processing techniques can be used to recognize the diseases in agricultural plants which reduce the reliance on farmers for agricultural yields protection. The diseases in the rice plants can be reduced through developing a greenhouse environment. In this work, Paddy Leaf Disease Recognition in greenhouse environment using Optimized Deep Neural Network (DNN) is proposed. A greenhouse environment is developed for growing the rice plants. In image acquisition, the normal leaf images and the diseased leaf images are collected from the developed greenhouse environment. Background removal is processed by converting RGB images into HSV images also binary images which are obtained from hue and saturation. The diseased, normal and background portion is segmented via clustering approach. The classification process is performed using DNN optimized by Jaya Optimization Algorithm (DNN_JOA). The stability of this method is defined by the generation of feedback loop. The experimental outcomes are evaluated and the performance is compared with DNN, ANN and DAE. The proposed approach for blast affected image attained high accuracy of 98.34% and normal leaf image with 97.52%. [ABSTRACT FROM AUTHOR]