Comparative Analysis of Peak Current Prediction based on Random Forest and MLP Neural Network Algorithms

• Main Authors: Bhoyar P., Rahman M.S.A., Irfan S.A., Amirulddin U.A.U.
• Other Authors: 58522765200
• Format: Conference Paper
• Published: Institute of Electrical and Electronics Engineers Inc. 2024
• Subjects: artificial intelligence; deep learning; hybrid optimization; Lightning nowcasting; Computational efficiency; Deep learning; Forecasting; Learning systems; Comparative analyzes; Fault analysis; Hybrid optimization; Learning models; Nowcasting; Peak currents; Prediction-based; Random forests; Lightning

Lightning events have significant impacts on power systems, infrastructure, and the environment. Accurate and timely nowcasting of lightning occurrences is crucial for effective fault analysis and mitigation. This paper presents the development of a hybrid optimization-based deep learning model for lightning nowcasting, aiming to improve the accuracy and efficiency of lightning prediction. The objectives include the development of a deep learning model utilizing lightning data, spatial prediction of lightning events within a 1 km diameter, investigating the model's capability for predicting specific time intervals and optimizing the computational cost and prediction accuracy. The proposed model demonstrates enhanced predictive capabilities and optimized computational efficiency, highlighting the potential of AI-driven techniques in lightning nowcasting and fault analysis applications. � 2023 IEEE.