Design and implementation of a visual modeling tool to support interactive runoff forecasting

To tackle the problems of low modeling efficiencies involved in implementing runoff forecasting using conventional modeling technologies, a visual modeling tool is established by integrating a visual modeling editor with the artificial neural network (ANN) modular to support interactive and fast modeling of the complex and dynamic runoff process. The workflow of visual modeling includes the prediction schema definition, ANN architecture design, data processing, ANN training and validation, runoff forecasting. In particular, to facilitate the user's activities for runoff simulation, an operational data exchange and model linking mechanism is proposed to allow for interoperability between models and multi-data sources. A case study for interactive runoff forecasting is given for Qingjiang River basin located in the middle part of China. A serial simulation experiments were carried out to verify the feasibility of the visual tool, and the results show that the tool can greatly enhance the easy-to-use capabilities of visual modeling and appeal in offering a positive prospect of improving the efficiency and robustness of current practice in hydrological modeling.