Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{197411,
        author = {Dr.Kamel Alikhan Siddiqui and Mr. Mohammed Muzammil Uddin Ahmed and Ms. Syed Amatul Asma and Ms. Syeda Mah Zehra Zaidi},
        title = {Federated Learning for Flood Forecasting: Secure and Scalable Prediction Models Using FFNN and CNN2D},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {6047-6056},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=197411},
        abstract = {Floods are one of the most common natural disasters that occur frequently, causing massive loss of life and economic damage across the globe. This paper proposes a Flood Forecasting Model (FFM) based on Federated Learning (FL) integrated with a Feed-Forward Neural Network (FFNN) and a 2D Convolutional Neural Network (CNN2D) extension. The model trains locally at eighteen regional client stations using historical water-level and climate data spanning 2010–2021, transmits only model parameters never raw data to a centralized aggregation server, and issues five-day-ahead flood alerts. The proposed architecture achieves an overall prediction accuracy of 84% on the FFNN model, while the CNN2D extension further reduces Mean Squared Error (MSE) and Root Mean Squared Error (RMSE) and improves accuracy. The system simultaneously preserves data privacy, reduces communication bandwidth, and enables scalable deployment across additional river basins. Evaluation on five major rivers and barrages in Pakistan confirms the model's effectiveness for early flood warning, with potential applications in any region facing recurrent flood risk.},
        keywords = {Federated Learning, Flood Forecasting, Feed Forward Neural Network (FFNN), CNN2D, Data Privacy, Machine Learning, Water Level Prediction, Natural Disaster Mitigation, Decentralised Training, Flood Alert System.},
        month = {April},
        }