Treffer: IoT enabled sustainable energy management system with renewable power forecasting and adaptive load strategy using walrus optimization algorithm and continual spatio-temporal graph convolutional network.

The IoT-enabled Smart Energy Management System (SEMS) plays a crucial role in managing renewable energy generation (REG) and optimizing load distribution. However, its performance is largely dependent on accurate forecasting models; data inconsistency creates inefficiencies in energy distribution. To address this challenge, this study proposes a hybrid approach combining the Walrus Optimization Algorithm (WOA) and the Continual Spatio-Temporal Graph Convolutional Network (CSTGCN) for accurate solar energy prediction. Solar radiance data is attained from the National Solar Radiation Database (NSRDB) and preprocessed using Cauchy Robust Correction-Sage Husa Extended Kalman Filtering (CRCSHEKF) to normalize values and handle missing data. The Waterwheel Plant Algorithm (WPA) is then employed for feature selection, identifying key variables such as Global Horizontal Irradiance (GHI), temperature, pressure, and wind speed (WS). Selected features are fed into CSTGCN, which predicts solar energy generation on both a day-ahead and a monthly basis. To further enhance forecasting accuracy, WOA is utilized to optimize the hyperparameters of CSTGCN, reducing prediction errors. The suggested WOA-CSTGCN model is implemented in MATLAB and evaluated against existing techniques, including Convolutional Neural Network (CNN), Support Vector Machine-Particle Swarm Optimization (SVM-PSO), Recurrent Neural Network (RNN), Sail Fish Optimization-Adaptive Neuro-Fuzzy Inference System (SFO-ANFIS), and Deep Dilated Multi-Kernel Convolutional Neural Network-Modified Elephant Herd Optimization Algorithm (DDMKCNN-MEHOA). Results demonstrate that WOA-CSTGCN achieves 98% accuracy, outperforming other methods, which range from 90 to 95%. Furthermore, the suggested method yields lower error metrics, with MAE, MAPE, and RMSE values of 0.41, 0.23, and 0.34, respectively, establishing its effectiveness in accurately forecasting renewable energy (RE) in smart grid (SG) applications. [ABSTRACT FROM AUTHOR]

Copyright of Environment, Development & Sustainability is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)