Key points are not available for this paper at this time.
Smart cities are having the ability to monitor and manage their environments in real time due to the emergence of Internet of Things technology. In the context of energy management, energy prediction can be carried out by monitoring and evaluating dynamic environmental data from the energy user side. The decision-making process related to energy production can then be aided by this information in order to achieve flexible production and avoid an excess or insufficient supply of energy. The quantity and variety of IoT data makes it difficult to create an efficient energy forecast system that effectively captures the changing conditions of the IoT environment. This research aims to forecast energy in order to guarantee efficient energy management in networks of smart cities. Here, unique framework, called as, Contiguous Temporal Chebyshev Convolutional Optimized Network (CoC-TemNet) is developed for the energy management and load forecasting of IoT-enabled smart city applications. For choosing the list of crucial properties for computing the energy function, the Chebyshev Non-Spiritual Network Model (CN2M) is used in this instance. Then, the Contiguous Temporal Convolution Network (CTCN) model is used to forecast energy with accuracy using the chosen features. The Hybrid Leaping Lizard Immune Optimization (HLIO) technique is used to calculate the objective function for improving the prediction process. The proposed method was validated on multiple datasets: Southern China, IHEPC, AEP, and ISO-NE. Outperforms baseline models with low RMSE, MSE, MAE values, and 28.1% MAPE. Significantly lower execution times: 0.98ms for IHEPC and 0.11ms for the AEP dataset.
Priyadarsini et al. (Wed,) studied this question.