基于FCE和SVM融合的线路典型冰风灾害算法分析

目前,国内外对于线路覆冰和风耦合作用的灾害分析较少,因此,提出了一种基于模糊综合评价法(fuzzy comprehensive evaluation,FCE)和支持向量机(support vector machine,SVM)融合的线路典型冰风灾害分析算法。通过分析典型冰风灾害影响因子及类型,借助模糊综合评价法提取了关键的灾害影响指标,并对风速和风向关键指标进行修正。在提取的温度、相对湿度、风速、风向和地貌5类致灾相关程度高指标的基础上,提出了采用径向基RBF核函数的非线性SVM小样本灾害分析模型。通过历史的冰风故障和非冰风故障数据建立训练样本和测试样本,仿真结果表明,建立的模糊综合评价和支持向量机融合的冰风灾害模型可有效分析判断冰风灾害发生的概率,实现了对冰风灾害小样本数据的可靠分析。

Risk Assessment of Hail Disaster on Cotton——A Case Study in Anhui Province

[Objective] The aim was to reveal the pattern of hail disaster risk for cotton in Anhui Province. [Method] Based on database of risk evaluation on hail disaster for cotton in Anhui Province, statistical inference and comprehensive indices measurement were made use of for extraction of disastrous index of hail （H） and loss rate of cotton （L）. Hail vulnerability curves in four stages （seedling, bud, boll and boll opening stages） of cotton were fitted by Matlab and SPSS software. Risk evaluation on hail disaster of cotton in Anhui Province was conducted and the related map was drawn on basis of yearly model of loss rate （by hail disaster）, as follows： Sa=n∑i=1Si-a∏i=1Si=1-（1-Si） [Result] In Anhui Province, high risk area by hail concentrates in northeast regions and mountainous regions in west of Anhui, and lower risk area is distributed in middle and down streams of Yangtze River and mountainous area in south of Anhui. [Conclusion] The research explored evaluation methods on hail risks based on the limited information, and could provide references for risk evaluation on hail disaster in other regions.