An Intelligent Multi-agent Approach for Flood Disaster Forecasting Utilizing Case Based Reasoning

Flood disaster refers to a short-term or continuous heavy rainfall, and forecasting flood disaster accurately is of great importance in human life. Therefore, in this paper, we present a novel intelligent multi-agent approach for flood disaster forecasting based on case based reasoning. Firstly, the proposed framework is constructed by three main modules, which are "Front end user computer", "Back end server" and "Flood disaster forecasting servers". Particularly, the proposed flood disaster forecasting system is made of several agents, in which each agent is designed to implement a particular functional unit. Secondly, the flood disaster forecasting algorithm is illustrated. In our algorithm, each agent has its own case base and can not visit the case base of other agents directly, and each case is made up of a problem part and a solution part. Finally, experiments are conduct to make performance evaluation based on the "Active Archive of Large Floods, 1985-Present" dataset. From the experimental results, the conclusions can be drawn that the proposed algorithm can predict the water level accurately, and the forecasting error rate of our algorithm is lower than exisiting method.