Event-Triggered Decentralized Intelligence with Energy-Aware Federated Learning for Real-World Systems

Gradient-Complementary Asynchronous Federated Learning for Multi-Task Anomaly Detection in Smart Grid

The increasing deployment of heterogeneous IoT devices has transformed smart grids into large-scale distributed cyber–physical systems, where anomaly detection becomes a critical yet challenging computational intelligence problem. In such environments, anomaly knowledge is sparse, fragmented, and highly non-independent across users, while device participation is asynchronous and communication-constrained. This paper proposes a gradient-complementary asynchronous federated learning (GC-AFL) framework, which explicitly models gradient complementarity as a distributed intelligence fusion mechanism for multi-task anomaly detection. Unlike conventional federated aggregation that suppresses heterogeneity, GC-AFL exploits dissimilar gradient information to preserve task-specific anomaly knowledge. The framework further integrates a communication-aware collaboration strategy and a staleness-compensated aggregation scheme to ensure efficiency and long-term fairness under asynchronous updates. Extensive experiments demonstrate that GC-AFL consistently outperforms state-of-the-art synchronous and asynchronous federated learning methods in terms of detection accuracy, robustness to Non-IID data, anomaly recall, and communication efficiency. The results validate the effectiveness of gradient complementarity as a general computational intelligence principle for distributed anomaly detection.

Communication-Aware Federated Learning for Energy Management in Edge-Cloud Autonomous Systems

A Federated Learning Driven Sustainable Computing Framework for Global Development Applications

The rapid expansion of data-driven technologies across global development sectors such as healthcare, agriculture, education, smart governance, and climate monitoring has introduced significant challenges related to data privacy, energy consumption, and equitable access to computational resources. Centralized machine learning paradigms often exacerbate these issues by requiring large-scale data aggregation, high communication overhead, and energy-intensive cloud infrastructures. To address these limitations, this paper proposes a Federated Learning–Driven Sustainable Computing Framework tailored for global development applications. The proposed framework enables decentralized model training across geographically distributed edge and fog nodes, allowing sensitive data to remain locally stored while collaboratively learning a global model. The architecture integrates energy-aware client selection, adaptive communication scheduling, and lightweight model aggregation to minimize computational overhead and carbon footprint. Sustainability is further enhanced through dynamic workload balancing and resource-adaptive learning, enabling the framework to operate efficiently in low-power and resource-constrained environments common in developing regions. The framework is designed to support heterogeneous devices, intermittent connectivity, and non-IID data distributions, ensuring robustness and scalability across diverse socio-economic contexts. Experimental evaluation across representative global development use cases demonstrates that the proposed framework reduces communication costs by up to 45%, lowers energy consumption by approximately 38%, and maintains competitive model accuracy compared to centralized learning approaches. Moreover, privacy preservation and data sovereignty are inherently supported, aligning with ethical AI and regulatory requirements. The proposed federated learning-driven sustainable computing framework offers a practical and scalable pathway toward environmentally responsible, privacy-preserving, and inclusive artificial intelligence, contributing directly to the United Nations Sustainable Development Goals (SDGs) through responsible digital innovation.

Distributed Artificial Intelligence for Smart City Infrastructure Management

Rapid urbanization has increased the need for intelligent and sustainable smart city infrastructure management. Smart cities generate massive amounts of data through IoT devices, sensors, cloud platforms, and communication networks, making traditional centralized AI systems less effective due to scalability, latency, privacy, and reliability challenges. Distributed Artificial Intelligence (DAI) addresses these issues by distributing intelligence across multiple interconnected nodes, enabling decentralized learning and decision-making. This study examines the role of DAI technologies such as multi-agent systems, edge computing, federated learning, IoT networks, and cloud-edge collaboration in managing urban services. A review of recent applications demonstrates DAI’s effectiveness in traffic management, energy distribution, water systems, predictive maintenance, public safety, and environmental monitoring. The proposed framework enhances real-time processing, resource optimization, fault tolerance, and data privacy. Results indicate that DAI outperforms centralized approaches in response time, scalability, accuracy, and reliability. The study concludes that DAI is a key enabler of future smart cities, with emerging technologies such as Explainable AI (XAI), blockchain, digital twins, and autonomous urban management expected to further improve smart city operations and citizen services.

DISTRIBUTED INTELLIGENCE AT THE EDGE: A MATHEMATICAL FRAMEWORK FOR DECENTRALIZED LEARNING IN IOT NETWORKS

A novel Federated Edge Learning (FEL) architecture that integrates software-defined networking principles with gossip-based communication protocols to facilitate collaborative model training while preserving data locality is proposed, offering a scalable, privacy-preserving solution for deploying artificial intelligence at the network edge.

Green Federated Learning with Adaptive Model Pruning for Sustainable Edge Intelligence in Smart Industries

Due to the rapid proliferation of Industrial Internet of Things (IIoT) systems within smart industries, scalable, energy-efficient, and sustainable edge intelligence solutions are in demand. Despite the fact that Federated Learning (FL) allows collaborative training without raw data sharing, traditional FL methods have high communication overhead, huge model sizes, and require a lot of energy, which restricts their implementation in resource-constrained industrial settings. To address these issues, the present paper introduces a Green Federated Learning framework, GFed-AMP, a sustainable IIoT edge intelligent framework that includes Adaptive Model Pruning.The suggested framework incorporates the dynamic magnitude-based pruning of local updates to eliminate redundant parameters and lessen the computational complexity. A strategy of energy-conscious client selection assigns priority to the devices with larger residual energy and with lower carbon intensity, to make sure that participation is environmentally conscious. It has real-time energy and carbon monitoring modules that can be used to gauge power consumption and environmental effects during federated training. Moreover, a sparse-awares aggregation mechanism is an effective method which optimally manages pruned model updates and ensures stable convergence. Experimental assessment of industrial anomaly detection data sets shows that GFed-AMP can achieve up to 40% model reduction, 35% communication overhead reduction, and 30% overall energy reduction in comparison with traditional FedAvg. Notably, the decline in the accuracy of predictions is less than 1.5%, which proves robustness and stability of learning. Better scalability, bandwidth, and lower carbon emission are proven by statistical comparisons. All in all, the overall experience with GFed-AMP’s trade-off between accuracy, communication efficiency, and environmental sustainability is a viable green AI solution to Industry 4.0 deployments.