Load forecasting plays a pivotal role in the efficient energy management of smart grid. However, the complex, intermittent, and nonlinear smart grids and the complexity of large dataset handling pose difficulty in accurately forecasting loads. The important issue is considering the cyclic features, which have not yet been adequately addressed through the trigonometric transformations. Furthermore, using long short‐term memory (LSTM) or 1D convolution neural network (1D CNN) and existing hybrid models involve stacked CNN‐LSTM architectures, employing 1D convolutions as a preprocessing step to downsample sequences and extract high‐ and low‐level spatial features. However, these models often overlook temporal features, emphasizing higher‐level features processed by the subsequent recurrent neural network layer. Therefore, this study considers a novel approach to independently process features for spatial and temporal characteristics using a parallel multichannel network comprising 1D CNN and bidirectional‐LSTM (Bi‐LSTM) models. The proposed model evaluated the National Transmission and Dispatch Company (NTDC) load dataset, with additional assessment on two datasets, American Electric Power and Commonwealth Edison, to ensure its generalizability. Performance evaluation on the NTDC dataset yields results of 3.4% mean absolute percentage error (MAPE), 513.95 mean absolute error (MAE), and 623.78 root mean square error (RMSE) for day‐ahead forecasting, and 0.56% MAPE, 94.84 MAE, and 115.67 RMSE for hour‐ahead load forecast. The experimental results demonstrate that the proposed model outperforms stacked CNN‐LSTM models, particularly in forecasting hour‐ and day‐ahead loads. Moreover, a comparative analysis with previous studies reveals superior performance in reducing the error gap between predicted and actual values.

The Internet of Things (IoT) is emerging as an innovative phenomenon concerned with the development of numerous vital applications. With the development of IoT devices, huge amounts of information, including users’ private data, are generated. IoT systems face major security and data privacy challenges owing to their integral features such as scalability, resource constraints, and heterogeneity. These challenges are intensified by the fact that IoT technology frequently gathers and conveys complex data, creating an attractive opportunity for cyberattacks. To address these challenges, artificial intelligence (AI) techniques, such as machine learning (ML) and deep learning (DL), are utilized to build an intrusion detection system (IDS) that helps to secure IoT systems. Federated learning (FL) is a decentralized technique that can help to improve information privacy and performance by training the IDS on discrete linked devices. FL delivers an effectual tool to defend user confidentiality, mainly in the field of IoT, where IoT devices often obtain privacy-sensitive personal data. This study develops a Privacy-Enhanced Federated Learning for Intrusion Detection using the Chameleon Swarm Algorithm and Artificial Intelligence (PEFLID-CSAAI) technique. The main aim of the PEFLID-CSAAI method is to recognize the existence of attack behavior in IoT networks. First, the PEFLID-CSAAI technique involves data preprocessing using Z-score normalization to transform the input data into a beneficial format. Then, the PEFLID-CSAAI method uses the Osprey Optimization Algorithm (OOA) for the feature selection (FS) model. For the classification of intrusion detection attacks, the Self-Attentive Variational Autoencoder (SA-VAE) technique can be exploited. Finally, the Chameleon Swarm Algorithm (CSA) is applied for the hyperparameter fine-tuning process that is involved in the SA-VAE model. A wide range of experiments were conducted to validate the execution of the PEFLID-CSAAI model. The simulated outcomes demonstrated that the PEFLID-CSAAI technique outperformed other recent models, highlighting its potential as a valuable tool for future applications in healthcare devices and small engineering systems.