针对历史灾情资料缺乏,采用传统统计模型进行风险分析精度不高的缺陷,根据正态信息扩散原理,引入能够优化处理小样本和体现风险评估模糊不确定性的模糊风险计算模型,对辽宁省自20世纪90年代以来的旱灾受灾指数进行分析。计算结果表明:辽宁省的旱灾几乎年年发生,受损程度主要集中在20%-40%,受损10%以上旱灾发生概率为92.92%,结果符合实际。该方法简单易行,分析结果意义清楚,对防灾减灾具有一定的指导作用。
In general, the precision of risk analysis of traditional statistics model is not high for the lack of enough information of historical disasters recording. Considering of this limits, the fuzzy risk assessment model was introduced according to normal information diffusion principle. This model has the advantages of being able to optimally solve assessment problems with few samples and reflect the fuzzy uneeaainty property of risk assessment,so we applied this model to assess the drought index since the 1990s in Liaoning province. The results showed that the drought occurred almost every year,the stricken degree mainly concentrated between 20% and 40% ,the occurrence probability above 0.1 was 92.92%, which fitted the actual conditions. This method was simple and easy to use, the results were clear and has guidance significance for preventing drought.