System predicting tsunami based on artificial neural network

摘要

The present invention relates to a tsunami prediction system, which predicts the height of a wave and the flooding rate of a target area by using information measured at an initial stage of an earthquake outbreak. In particular, the present invention relates to a tsunami prediction system based on an artificial neural network, which can stochastically provide the height of a wave of the target area by acquiring the probability distribution of the height of a wave by using a prediction model of an artificial neural network to which statistical estimation is applied, and can provide an evacuation range, an evacuation shelter and the like in a more detailed manner by classifying a safe region and a dangerous region of the target area which is flooded in a more reliable manner. To this end, the tsunami prediction system based on an artificial neural network according to the present invention comprises: a database in which data regarding the topography and the water level of the target area, and data regarding tsunami of the target area, are registered; a prediction model generation unit which generates a wave height prediction model and a flooding rate prediction model by using the data registered in the database; and a tsunami prediction unit which calculates the height of a wave of the target area by applying the epicenter and the scale of the earthquake to the wave height prediction model, and calculates the flooding rate of the target area by applying the calculated height of a wave to the flooding rate prediction model. Here, the wave height prediction model is produced by the artificial neural network after statistical estimation is applied thereto. In addition, the tsunami prediction unit repeats calculation to acquire the probability distribution of the height of a wave, and calculates the flooding rate by using the height of a wave which is selected from the acquired probability distribution.