College of Engineering Roorkee (COER) was established in 1998 by Seth Roshan Lal Jain Trust. It is in Roorkee on National Highway (NH-58), approximately 180 km (110 mi) north of Delhi, the capital of India. It is 7 km (4.3 mi) from Roorkee towards Haridwar. The campus is 30 ha (74 acres). It is affiliated to Uttarakhand Technical University.
Electrocardiogram (ECG) analysis is essential for diagnosing various cardiac conditions, particularly arrhythmias, which can result in severe health complications if not addressed promptly. This review paper provides an overarching analysis of extant literature on the grounds of four prominent neural network architectures-Recurrent Neural Networks (RNN), Gated Recurrent Units (GRU), Convolutional Neural Networks (CNN), and Long Short-Term Memory (LSTM) networks-in the detection of arrhythmias. We examine the performance of these models using two well-established ECG data dumps: the Holter Data dump for Sudden Cardiac Death and the MIT-BIH Arrhythmia Data dump. The review synthesizes findings from multiple studies, highlighting the accuracy of each model. Notably, CNNs have achieved accuracies of 99.57% and 99.69% on the Holter and MIT-BIH data dumps, respectively. LSTM networks also demonstrated high accuracy rates of 99.53% and 98.50%. GRU models showed competitive performance with accuracies of 99.29% and 97.90%, while RNNs, despite slightly lower accuracy, still yielded promising results. This review aims to unveil into the effectiveness of these neural network architectures for ECG-based arrhythmia detection, guiding future research directions in this critical area of healthcare.
The liability of Distribution Control Centre (DCC) becomes more important to handle irregular load demand of distribution network and supply good quality of electrical power to the consumers. In this work, wind power has been optimally placed with the Demand Response (DR) program under uncertain load demand for the purpose of improvement of power loss reduction, voltage profile, total operating cost reduction and emission reduction. Two DR model are considered named as Emergency Demand Response Program (EDRP) and Time of Use (TOU) program for demand side management. Due to irregular behaviour of wind power therefore, it is normalized by Rayleigh probability distribution function. Moreover, in order to find optimal position and number of wind turbine units are identified via Gbest-guided Artificial Bee Colony (GABC)-based two stage algorithm. The proposed methodology is divided into two stages. In first stage, identify the weak buses which are having large voltage deviation using voltage deviation index. In second, the optimization algorithm identifies the optimal number of wind generator units. As a result, it improves the reduction of system power loss, voltage profile and reduction of emission generation level from thermal power plant. This approach is implemented on standard 33-bus and 85-bus distribution systems and their results are presented. Finally, it is analysed that, the collective impact of wind power integration and DR models make the system efficient, cost-effective for operation and environmental friendly.
Since many real-world performance concerns are becoming bigger, more complex, and more dynamic, natural-inspired algorithms are becoming more and more popular. Due to the scale and complexity of the problems, it is necessary to develop an optimization strategy whose effectiveness is measured by finding the nearly optimal solution in a reasonable amount of time. The bio-inspired metaheuristic techniques are based on swarm intelligence, biological, physical, and chemical features, depending on the source of inspiration. Cancerous tumors form and propagate largely as a result of genetic factors. A high-dimensional data processing difficulty exists with the identification of the pertinent genes. The Pyramid-GSA hybrid framework was developed to tackle the issue of dimensionality reduction through a cyclical gene-reduction method. This framework integrates two advanced modules: a high-throughput filtering component and an iterative wrapper technique inspired by the gravitational search algorithm (GSA). Using a recursive selection approach, top-performing genes are retained in each reduction cycle, facilitating efficient dimensionality compression while preserving essential genetic features. This methodology aims to optimize the gene subset to enhance classification accuracy and minimize interference from non-informative genes. Testing was conducted on a multi-class breast cancer microarray gene expression dataset, with comparisons to established feature selection methods to ensure a comprehensive analysis. The various approaches are assessed and contrasted using datasets from the UCI repository using a set of assessment metrics. Findings demonstrate that in spite of initialization and the use of stochastic operators, for the most favorable feature combination, the suggested version of GSA aims to search the feature space for the optimal feature combination.
Wildlife and animal behaviour tracking and recording is a discipline that has received a lot of attention mainly because of the growing environmental changes and human activities that are influential in many ways to wildlife and their habitats. Former methods of data collection by direct observation, and handwriting, are now either supplemented or substituted by complex technological tools. This review paper also discusses different ones like deep learning models such as YOLOv8 for the constant monitoring of animal behaviour, hybrid models that combine VGG-19 and Bi-LSTM for real-time animal detection and alertness, as well as the multitask learning framework for efficient animal activity recognition using low-cost embedded systems. Such technologies provide potential strategies for tracking, identifying, improving the quality of life, and reducing the risk of potential incidences between humans and wildlife. Furthermore, the paper also enlightens the possibilities of applying CNN for sheep breed recognition, which plays a part in how machine learning helps enhance livestock use. The various accomplishments discussed in this paper reflect clear progress in the area of smart farming and wildlife protection and the technologies presented can be described as viable, practical, and feasible for implementation on a large scale.
Oceans and seas are wobble and the life under it is also fragile. However there is a visibility issue due to main factors like color distortion and other low light conditions. But while enhancing the quality of the image underneath, it is possible that the original meaning of the image is disrupted. This paper tells about the Stable Diffusion model using DreamBooth algorithm for enhancing the image and retaining its original meaning. This is evaluated using the SSIM index which tells the difference between the original and the augmented image. This paper uses this approach and attains a value of $\mathbf{9 8 \%}$ suggesting that the image is similar to the original one but is enhanced in terms of brightness, variance, and structure.