摘要
采用铱星SBD卫星通信业务,基于支持向量机算法,设计了一套专门服务于林业部门的信息采集、数据分析及林火识别的森林管护系统,实现了CO浓度、CO2浓度、烟雾浓度、空气温湿度的自动采集与林区无死角实时远距离无线传输,在10 m范围内可有效识别森林火灾发生的可能性。采集终端硬件基于STM32处理器、铱星Q9602模块、LTC2950-1开关机控制、DS2781电源管理以及两级降压结构,软件在自动模式及手动模式下均可实现数据的实时采集与远程传输,并在远程服务器上进行数据分析与林火发生可能性识别。经实验验证:系统可实时采集5种微环境参数,卫星传输成功率为100%,10 m范围内林火正确识别率为80%左右。2014年开始,系统在内蒙古自治区呼伦贝尔市海拉尔区林区的森林管护工作中得到应用测试。
With the analysis of domestic and abroad present situation of forest protection and fire monitoring,based on the Iridium SBD communication services and support vector machine algorithm,an information collecting and data analyzing system for forest management and protection was designed,which was able to collect concentration of carbon monoxide,carbon dioxide,smoke air temperature and air humidity automatically,transfer the wireless remotely in anywhere of forest,and recognize forest fire probability effectively in a certain range. The hardware of monitoring terminal consisted of STM32 CPU,Iridium Q9602 module,LTC2950-1,DS2781 and two-stage power voltage step down system. The software made the terminal be able to collect and transfer on automatic mode or manual mode,analyze data and recognize fire probability on the remote server. It was proved by experiment that the system could collect five micro-environment parameters,transfer data by using Iridium with the success rate of100% and recognize fire probability above the rate of 80%. The system had been applied and tested in the forest fire work of Hailar District, Hulunbeier, Inner Mongolia Province. It can improve the monitoring mechanism of the patrol personnel and the efficiency of forest resources survey.
出处
《农业机械学报》
EI
CAS
CSCD
北大核心
2016年第1期324-330,共7页
Transactions of the Chinese Society for Agricultural Machinery
基金
中央高校基本科研业务费专项资金项目(BLX2013009)
北京市教委共建项目
北京市共建项目
关键词
森林管护
铱星通信
Q9602模块
支持向量机
林火识别
forest management and protection
Iridium communication
Q9602 module
support vector machine
forest fire recognition