GIET University

Fault detection in microgrids is a critical element of system stability and uninterrupted power delivery. Herein, a comparative study using LSTM and bidirectional LSTM networks is performed based on three-phase current data for multi-class fault classification. Five major fault types, namely LG, LL, LLG, LLL, and LLLG, were simulated using a Real-Time Digital Simulator (RTDS) under grid-connected and islanded modes. Collected current signals were preprocessed, normalized, and segmented for sequence learning. Later, both models were trained using the best hyperparameter setting to enhance their capabilities and classify faults. To measure how well they identified faults, evaluation metrics, like accuracy, precision, recall, F1-score, and ROC-AUC, were calculated. The results revealed that the Bi-LSTM outperformed the LSTM and classical machine learning models consistently, with more than 99% accuracy for most fault types. More importantly, the proposed framework also checked classification performance for LLLG faults, with the Bi-LSTM model having a test accuracy of 98.8%. These results confirm that the Bi-LSTM model can robustly and precisely classify and detect faults in real time within specific phases of microgrids; therefore, it provides a scalable foundation for the development of intelligent protection in smart power systems.

ABSTRACT The vast amount of electronic health records (EHRs) is a norm to specify the health data in a digital world. As the amount of data starts to grow, handling the tasks on the basis of attributes such as historical performance, computational capacity, and workload results in a serious problem. The high processing loads can consume more resources and result in inefficiencies or delays in using and evaluating sensitive health data. With the emergence of the internet of things (IoT), artificial intelligence (AI), machine learning, and deep learning strategies, data‐driven applications become a promising solution for designing robust diagnostic approaches utilizing healthcare data. These mechanisms obtained much attention from industries and the educational sectors and resulted in better advancements in healthcare models. The AI‐driven approaches associated with healthcare applications still encounter some problems including security, privacy, and the quality‐of‐service (QoS). Machine learning with privacy‐preserving approaches safeguards the data; yet still, it is complex to formulate the infrastructural facilities. Certain problems are efficiently resolved by federated learning (FL) since it enables personal communication between distinct technologies and lightens the system's computational overload. Moreover, it efficiently manages the EHR data. Therefore, this work designs an efficient AI‐driven load distribution in the federated network by considering the limitations of the existing mechanisms. In the developed framework, at first, the essential EHRs are collected from the benchmark resources and provided to the FL network to resolve the overloading issues. Further, the EHR data is used to detect the disease using an Adaptive Temporal Convolution Network with Attention Mechanism (ATCN‐AM) in the FL network. Moreover, the parameters in the ATCN‐AM model are tuned using Improved Dung Beetle Optimizer (IDBO). Finally, different experimental validations are performed in the developed framework over the conventional mechanisms.

The human genome is folded into chromatin loops by the cohesin complex, forming functional chromatin domains that underlie transcription and DNA replication/repair. However, how cohesin organizes these domains in living cells, especially in active euchromatin, remains elusive. To address this question, we combined single-nucleosome imaging/tracking and super-resolution 3D-structured illumination microscopy (3D-SIM) with euchromatin-specific labeling of histone H3.3. Using this nanoscopic approach, we revealed that euchromatin forms condensed domains that are constrained by cohesin-mediated loops. This organization refines the classical textbook view of euchromatin as largely open, in line with emerging evidence. Transcription machinery appears to be located near the condensed domain surfaces/borders. Cohesin loss increased nucleosome fluidity within these domains without altering their overall compaction, leading to local mixing of domains and compromising transcriptional insulation. These findings uncover an unexpected physical role of cohesin in maintaining the integrity of condensed euchromatic domains and ensuring proper higher-order regulation of gene expression.

An assessment of the soil fertility status of the RRTTS and KVK farms at Bhawanipatna revealed that the soils were neutral to slightly acidic in reaction. Electrical conductivity values were within permissible limits, indicating suitability for crop production. The soil organic carbon content varied from medium to high across the study area. Available potassium levels were found to be high, whereas available nitrogen and Bray’s phosphorus ranged from low to medium. Analysis of micronutrients showed that iron, manganese, and copper were present in sufficient quantities. However, about 49 per cent of the soils were deficient in available zinc, while the remaining plots exhibited adequate zinc levels. The application of fertilizers and organic manures based on soil test recommendations would help enhance crop productivity while sustaining soil health.