Federated Learning (FL) is a transformative decentralized approach in machine learning and deep learning, offering enhanced privacy, scalability, and data security. This review paper explores the foundational concepts, and architectural variations of FL, prominent aggregation algorithms like FedAvg, FedProx, and FedMA, and diverse innovative applications in thermal comfort optimization, energy prediction, healthcare, and anomaly detection within smart buildings. By enabling collaborative model training without centralizing sensitive data, FL ensures privacy and robust performance across heterogeneous environments. We further discuss the integration of FL with advanced technologies, including digital twins and 5G/6G networks, and demonstrate its potential to revolutionize real-time monitoring, and optimize resources. Despite these advances, FL still faces challenges, such as communication overhead, security issues, and non-IID data handling. Future research directions highlight the development of adaptive learning methods, robust privacy measures, and hybrid architectures to fully leverage FL’s potential in driving innovative, secure, and efficient intelligence for the next generation of smart buildings.

Detecting plant diseases is vital for maintaining agricultural productivity and ensuring food security. Advances in computer vision, particularly with Vision Transformers (ViTs), have shown significant potential in improving the accuracy and efficiency of plant disease identification. This study provides a comprehensive evaluation of various ViT parameters to determine the most effective configuration for this purpose. Using the extensive PlantVillage dataset, we systematically analyzed the effects of patch sizes, image resolutions, embedding dimensions, the number of transformer blocks (depth), the number of heads in the multi-head attention layer, and the dimension of the MLP (FeedForward) layer on model performance. We introduced saliency map visualizations to enhance interpretability and evaluate the critical regions contributing to classification decisions, ensuring the approach’s transparency and robustness. Our experiments identified the optimal ViT configuration as follows: image size = 224×224, patch size = 16, embedding dimension = 512, depth = 6, number of heads = 8, and MLP dimension = 1024. This configuration achieved an impressive accuracy of 99.77% on the PlantVillage dataset. In addition, we incorporated a novel cross-dataset transferability evaluation to validate the generalizability of the proposed model. Comparative analysis with traditional convolutional neural network architectures, such as VGG19 and AlexNet, revealed that our optimized ViT model not only surpasses these models in accuracy but also requires significantly fewer trainable parameters and storage space. The incorporation of a lightweight, domain-specific fine-tuning process ensures the model’s adaptability to new datasets with minimal computational overhead. Our findings highlight the scalability and adaptability of ViTs, emphasizing their ability to effectively handle varying image sizes and resolutions. Moreover, our approach outperforms recent state-of-the-art methods across multiple databases, underscoring the efficacy of the chosen ViT parameters.

Early fault detection in photovoltaic grid-connected systems is crucial for optimizing energy output and ensuring system reliability. However, challenges such as noise interference, limited computational resources, and the need for high diagnostic accuracy make fault detection complex and resource-intensive. This study introduces a novel fault diagnosis approach aimed at enhancing diagnostic accuracy and efficiency, even in noisy environments, while being efficient enough for real-time deployment on resource-limited devices. Raw one-dimensional time series data are transformed into 2D pixel-wise images to meet the requirements of a lightweight MobileNet architecture, specifically chosen for its ability to efficiently extract discriminative features in this context. Alongside, an ensemble stacked classifier incorporating Support Vector Machines, Random Forest, and Decision Trees is employed, enabling the use of MobilNet features for improving the classification. This combination provides insights into the relative strengths of each model type, showcasing the benefits of contrasting traditional machine learning with more complex deep learning methods for fault detection in PV systems. The model demonstrates high test accuracy on both noiseless (97.36%) and noisy datasets (91.67%), with precision, recall, and F1 scores above 91% across all conditions, while maintaining low processing times. This approach not only advances fault diagnosis in grid-connected PV systems but also offers a practical framework for selecting optimal models tailored to specific application requirements.