摘要
为增强辽宁省农业旱灾风险综合管理水平,对其2005-2016年的农业旱灾脆弱性运用随机权神经网络法进行评价,并检验分析了传统BP网络、RBE网络与随机权神经网络模型的准确性与可靠性。结果显示:经济的快速增强是驱动辽宁省农业旱灾脆弱性的主要趋势,研究期间该区域旱灾脆弱性整体呈现出减弱的变化趋势;相对于其他两种模型,随机权神经网络具有更好的可靠性与准确性,可为区域旱灾风险综合管理和防控提供一定参考价值。
In order to enhance the comprehensive management level of agricultural drought risk in Liaoning Province,the vulnerability of agricultural drought from 2005 to 2016 was evaluated by using stochastic weight neural network method,and the accuracy and reliability of traditional BP network,RBE network and stochastic weight neural network models were tested and analyzed. The results show that the rapid economic growth is the main trend driving the agricultural drought vulnerability in Liaoning Province,during the study period,the drought vulnerability in this region showed a decreasing trend;compared with the other two models,the stochastic weighted neural network has better reliability and accuracy,which can provide a certain reference value for regional drought risk integrated management and control.
作者
王琳
WANG Lin(Dashiqiao Water Conservancy Affairs Center,Yingkou 115100,China)
出处
《黑龙江水利科技》
2019年第3期4-8,共5页
Heilongjiang Hydraulic Science and Technology
关键词
旱灾风险
脆弱性评价
神经网络
辽宁省
drought risk
vulnerability assessment
neural network
Liaoning Province
