基于气象数据深度挖掘的雷电灾害自动化监测系统设计

Design of an automated monitoring system for lightning disasters based on deep mining of meteorological data

为提高雷电灾害的预防和应对能力,设计基于气象数据深度挖掘的雷电灾害自动化监测系统。通过360度雷电传感器获取当前环境中云闪、地闪电磁波脉冲频率与次数数据,利用定位模块获取雷电发生经纬度信息,将气象数据经过Oracle10数据库存储后,使用深度挖掘算法中的Clus⁃tream算法,建立雷电灾害自动化监测框架,将气象数据输入到该框架内,使用多机器学习单元对气象数据训练集进行学习,得到基分类器,通过该分类器挖掘气象数据内的雷电灾害监测结果,实现雷电灾害自动化监测。实验结果表明,该系统具备较强的云闪电磁波采集能力和监测雷电灾害能力,应用性强。

In order to improve the ability of lightning disaster prevention and response,an automatic lightning disaster monitoring system based on deep mining of meteorological data is designed.The pulse frequency and frequency data of cloud flash and ground flash electromagnetic wave in the current environment are obtained through the 360 degree lightning sensor.The longitude and latitude information of lightning occurrence is obtained by using the positioning module.After the meteorological data is stored in the Oracle10 database,the automatic monitoring framework of lightning disaster is established by using the Clustream algorithm in the deep mining algorithm.Meteorological data is input into the framework,and multiple machine learning units are used to learn the training set of meteorological data,and a base classifier is obtained.Through this classifier,lightning disaster monitoring results in meteorological data are mined to realize automatic lightning disaster monitoring.The experiment result shows that the system has strong ability of collecting cloud and lightning wave and monitoring lightning disaster.