Electrical power networks depend on power system stability to provide dependable and continuous operation. It speaks to a power system's capacity to continue operating in a balanced and steady manner after minor and major disruptions. The stability of the electricity system is necessary to avoid large-scale blackouts and guarantee that customers will always have access to energy. The most momentous obstacle to generating a steady flow of electricity at a low cost is the fluctuating pattern of generation, load, and demand. To address this problem, a day-ahead energy management system (EMS) based on deep learning has been proposed; it offers the necessary functionality to guarantee that the energy production systems balance stability levels at the lowest possible operational costs. A Convolutional Neural Network (CNN) and CNN-Multilayer Perceptron (CNN-MLP) approach has been created for forecasting the accuracy level of electricity balance stability with the use of demand-side management (DSM). To improve outcomes, the dataset has been subjected to data analysis and feature scaling. The customer response time regarding the electricity production dataset, established statistical parameters, and minimum price elasticity coefficients (0.05–1) were examined in addition to the time-domain simulation findings to verify the stability of the power network. A score of 93.75% accuracy is offered by the proposed CNN-MLP model, whereas CNN delivers 93.55% accuracy. The outcome of this suggested optimized deep learning CNN-MLP predictive model, which uses the Adam optimizer, indicates that it is much more beneficial than the CNN predictive model in terms of properly monitoring the balance of power and providing it to clients at a reasonable price.
The growth of drone and Unmanned Aerial Vehicle (UAV) use in cities brings a new set of challenges to safely manage air traffic in crowded urban skies. This research proposes an innovative IoT-enabled and AI-driven air traffic control system designed specifically to meet the unique demands of urban drone applications such as deliveries, surveillance and disaster response. The proposed system integrates advanced AI techniques, namely Particle Swarm Optimization (PSO) for decentralized control, Ant Colony Optimization (ACO) for adaptive routing and Non-dominated Sorting Genetic Algorithm II (NSGA-II) for multi-objective optimization. These methodologies enable real-time route optimization, collision avoidance and energy efficiency. A combination of three-dimensional (3D) urban navigation and Voronoi-based geofencing helps in avoiding obstacles and restricted areas. This proposed system also incorporates IoT-based battery monitoring for real-time energy management and utilizes hybrid communication networks to maintain seamless operations in high-density urban airspaces. Cellular, satellite and local access make for effective and dependable delivery of the communication network. Weather conditions are used in optimization to achieve adaptive routing. Case studies illustrate the practical implementation of this scalable framework, highlighting its ability to enhance safety, operational efficiency and sustainability in urban UAV operations.
Smart grid stability prediction (SGSP) is essential for ensuring reliable energy delivery and protecting critical infrastructure. Fluctuations in generation, load, and demand create significant challenges in maintaining stable and affordable electricity. The research presents a hybrid approach to SGSP by integrating Logistic Regression, Decision Tree, and Random Forest models using a hard voting method. This hybrid model achieves 99% accuracy, significantly outperforming standalone methods. Among the individual algorithms, Random Forest demonstrates strong performance with 98.7% accuracy, showcasing its reliability for complex data handling. The proposed approach enhances stability predictions, providing robust support for smarter and safer grid operations. By ensuring stable energy delivery, the system facilitates the seamless integration of renewable energy sources, contributing to sustainable energy management. Extensive validation confirms the effectiveness of the proposed model and its potential to address critical challenges in modern energy systems. These contributions aid utility providers, policymakers, and consumers by creating a more reliable and efficient energy infrastructure, paving the way for a resilient and future-ready grid.
Healthcare faces a tough challenge with misdiagnosis, which contributes to $\text{1 0 - 1 5}$ percent of adverse medical events globally, according to the World Health Organization. Accurately identifying diseases based on symptoms is challenging because many conditions share similar signs, and traditional methods often struggle to keep up. In this study, we explored the Symptom-Disease Prediction Dataset, which contains 4961 patient records, to develop a machine learning solution for better diagnosis. We tested various models, including Random Forests, Support Vector Machines and Neural Networks, when combined using ensemble techniques. Our K-Nearest Neighbors model proved to be the standout, achieving a solid accuracy of 90.74 percent, surpassing other approaches, with a precision of 90.37 percent and an F1-score of $\text{8 9. 3 2 \%}$. This study offers a useful solution for improving diagnosis, minimizing mistakes, and building trust in AI-powered healthcare tools. We suggest conducting experiments in real-world healthcare settings to enable the use of this method on a daily basis and improve patient care.
This paper presents a design of hybrid energy system for three different locations around the world. A residential load supporting 80 households were taken into the study and the optimal cost-effective solution was determined using Homer Pro software. The locations were selected based on the annual solar radiation and wind speed data. Bali, Phoenix and Wellington were selected based on the data from National Solar Radiation Database and NASA Surface Meteorology and Solar Energy (SSE) database. The optimization shows full scale renewable sources deployment in Phoenix and Wellington where both sun and wind plays a vital role on the climate respectively. Partial (almost 50%) usages of solar PV arrays are required for electric production in tropical areas like Bali. For sunny Phoenix, the solar portion is 79% to support the load whereas for windy Wellington, the portion is 26.4% for solar PV arrays.
1 October 2026 in IEEE Transactions on Systems, Man, and Cybernetics: Systems