LSTM Based Hybrid Method for Basin Water Level Prediction by Using Precipitation Data
S. Liu, Lei Puwen, Koji Koyamada
Abstract
Water level prediction is becoming increasingly important. However, physical models tend to become difficult to apply when it comes to some small rivers which have insufficient hydrological data. To address it, nowadays, deep learning methods are increasingly being applied to climate prediction analysis as an alternative to computationally expensive physical models for its features of flexible data-driven learning and universality. In our paper, we focus on the precipitation-only water level forecasting problem by using long-short-term memory (LSTM) based hybrid model, and try predicting the future water level of all the rivers in Japan by using simulated precipitation data from the database for Policy Decision making for Future climate change (d4PDF).