The increasing integration of large solar PV and wind farms into the power grid has fueled, over the past two decades, growing demands for high-power, high-voltage, utility-scale inverters. Multilevel inverters have emerged as the industry’s choice for megawatt-range inverters because of their reduced voltage stress, capability for generating an almost-sinusoidal voltage, built-in redundancy and other benefits. This paper presents a novel switched-source multilevel inverter (SS MLI) architecture. This new inverter shows superior capabilities when compared to existing topologies. It has reduced voltage stress on the semiconductor, uses fewer switches (i.e., reduced size/weight/cost) and exhibits increased efficiency. The proposed SS MLI is comprised of two voltage sources ( ) and six switches. It is capable of generating five-level output voltage in symmetric mode (i.e., ) and seven-level output voltage in asymmetric mode (i.e., ). We present simulations results (using MATLAB®/Simulink®) for five- and seven-level output voltages, and they strongly support the validity of the proposed inverter. These positive results are further supported experimentally using a laboratory prototype.
There has been an effort for a few decades to keep energy consumption at a minimum or at least within a low-level range. This effort is more meaningful and complex by including a customer’s satisfaction variable to ensure that customers can achieve the best quality of life that could be derived from how energy is used by different devices. We use the concept of Shapley Value from cooperative game theory to solve the multi-objective optimization problem (MOO) to responsibly fulfill user’s satisfaction by maximizing satisfaction while minimizing the power consumption, with energy constrains since highly limited resources scenarios are studied. The novel method uses the concept of a quantifiable user satisfaction, along the concepts of power satisfaction (PS) and energy satisfaction (ES). The model is being validated by representing a single house (with a small PV system) that is connected to the utility grid. The main objectives are to (i) present the innovative energy satisfaction model based on responsible wellbeing, (ii) demonstrate its implementation using a Shapley-value-based algorithm, and (iii) include the impact of a solar photovoltaic (PV) system in the energy satisfaction model. The proposed technique determines in which hours the energy should be allocated to maximize the ES for each scenario, and then it is compared to cases in which devices are usually operated. Through the proposed technique, the energy consumption was reduced 75% and the ES increased 40% under the energy constraints.
The past two decades has seen a growing demand for high-power, high-voltage utility scale inverters mostly fueled by the integration of large solar PV and wind farms. Multilevel inverters have emerged as the industry choice for these megawatt range inverters because their reduced voltage stress, capable of generating an almost sinusoidal voltage, in-built redundancy, among others. This paper present a new Switched-Source Multilevel Inverter (SS MLI) architecture. The new inverter show superior over existing topologies. It has reduced voltage stress on the semiconductor, uses less number of switches –reduced size/weight/cost and increased efficiency. The new SSMLI is comprised of two voltage sources (V1, V2) and 6 switches. It is capable of generating 5-level output voltage in symmetric modes (i.e., V1 = V2), and 7-level output voltage in asymmetric modes (i.e., V1 ≠ V2). To demonstrate the validity of the proposed inverter, simulations results using MATLAB® /Simulink® for 5- and 7-level output voltages are presented . The simulations are also verified experimentally using a laboratory prototype.
Synthetic Aperture Radar (SAR) Automatic Target Recognition (ATR) systems face significant challenges due to limited labeled data availability and persistent domain gaps between synthetic and measured imagery. This paper presents a comprehensive self-supervised learning (SSL) framework that eliminates dependency on synthetic data while achieving state-of-the-art performance through multi-task pretext training and extensive downstream classifier evaluation. We systematically evaluate our SSL framework across diverse downstream classifiers spanning different computational paradigms and architectural families. Our study encompasses traditional machine learning approaches (SVM, Random Forest, XGBoost, Gradient Boosting), deep convolutional neural networks (ResNet, U-Net, MobileNet, EfficientNet), and a generative adversarial network. We conduct extensive experiments using the SAMPLE dataset with rigorous evaluation protocols. Results demonstrate that SSL significantly improves SAR ATR performance, with SVM achieving 99.63% accuracy, ResNet18 reaching 97.40% accuracy, and Random Forest demonstrating 99.26% accuracy. Our multi-task SSL framework employs nine carefully designed pretext tasks, including geometric invariance, signal robustness, and multi-scale analysis. Cross-validation experiments validate the generalizability and robustness of our findings. Rigorous comparison with SimCLR baseline validates that task-based SSL outperforms contrastive learning for SAR ATR. This work establishes a new paradigm for SAR ATR that leverages inherent radar data structure without synthetic augmentation, providing practical guidelines for deploying SSL-based SAR ATR systems and a foundation for future domain-specific self-supervised learning research in remote sensing applications.
Modern data centers are consuming more energy than ever before due to the rapid growth of cloud services, artificial intelligence (AI), and large-scale digital applications. As energy demand continues to rise, accurate load forecasting has become an important tool for improving energy management and operational planning. However, predicting data-center power demand remains challenging because computing workloads, cooling systems, and facility operations are closely connected and constantly changing. This review examines the use of artificial intelligence (AI) and machine learning (ML) techniques for data-center load forecasting. The surveyed literature covers traditional statistical methods, supervised learning algorithms, deep learning models, probabilistic forecasting techniques, and physics-informed hybrid approaches. Important topics such as forecasting horizons, feature selection, performance evaluation, and practical deployment challenges are also discussed. The reviewed studies indicate that advanced approaches, including long short-term memory (LSTM), gated recurrent unit (GRU), transformer-based models, and digital twin (DT) frameworks, can improve forecasting performance and support more energy-efficient operation. These methods can also assist carbon-aware and grid-interactive data-center management. Several challenges remain. These include limited data availability, poor generalization, interpretability issues, and real-time implementation constraints. In this paper, the reviewed studies are classified and analyzed to provide a clear reference for researchers working in AI-based data-center energy forecasting.