基于决策树算法的山洪灾害三维动态预测模型

A Three Dimensional Dynamic Prediction Model for Flash Floods Based on Decision Tree Algorithm

随着我国近几年来科学技术的飞速发展,有些地区积累了大量的水文及河道地形数据,如何有效利用这些数据来提高山洪灾害预测的准确性,实现山洪灾害预警的可视化,成为目前山洪灾害监控预警领域研究的热点问题之一。论文将决策树C4.5算法、Unity 3D三维建模技术和马斯京根法相互融合,通过超声波对相关河道河床信息进行采样,利用监测河段的水位、流量等数据,提出洪水三维动态预测演进模型,并结合短信模块设计了一套基于决策树算法的山洪灾害三维动态预测系统,实现了山洪灾情演进的可视化展示和危险性预测。经初步测试,该系统具有较强的可视交互性与信息警示准确性的优点,能避免或减轻山洪灾害对人民生命财产的损失。

With the rapid development of science and technology of our country in recent years,some regions have accumulated a lot of hydrology and river topography data,how to effectively use these data to improve the accuracy of flood prediction and visualization of flood warning,Which becomes one of the research hotspots in the field of flood disaster monitoring and early warning. In this paper,using decision tree C4.5 algorithm,Unity 3D modeling technology,Muskingum method,through the Ultrasonic detection technology to sample the relevant river bed information,using the data from the monitoring of the river water level,flow,the evolution model of 3D dynamic flood forecasting is put forward by combining the SMS module design a three dimensional dynamic prediction system of mountain flood disaster based on decision tree algorithm. This system realizes visual display of flood disaster evolution and fatalness prediction. Through preliminary tests,the system has the advantages of strong visual interaction and information warning accuracy,and can avoid or mitigate the loss of life and property by flash floods.