Rainfall infilling model with novel artificial neural network

The purpose of this thesis is to enhance the imputation of missing rainfall observations through the implementation of novel artificial neural networks that coupled with metaheuristic algorithms, the Grey Wolf Optimiser, the Multiverse Optimiser, and the Moth-Flame Optimisation algorithm. The proposed rainfall infilling models were compared against several existing Artificial Intelligence-based imputation models. The findings showed that the implementation of the proposed rainfall infilling models could boost the imputation accuracy, which can help in enhancing the accuracy and effectiveness of hydrological studies in Malaysia for better flood mitigation and water resources management.