甘肃省地区农业旱灾脆弱性评价

Agricultural Drought Vulnerability Evaluation of Gansu Province

农业旱灾脆弱性评价指标呈现多维及互不相关特性,利用混合蛙跳算法投影寻踪模型对农业旱灾脆弱性进行综合评价是一种新的研究方法.引入细菌觅食算法趋化操作至基本混合蛙跳算法中,强化了个体在解空间的精细化搜索能力,有效提高了算法优化投影寻踪模型的精度和稳定性.投影寻踪模型根据样本数据寻求最佳投影方向,进而获取指标影响权重系数,客观上降低了人为赋权对评价结果的影响.使用该方法对甘肃省14个地区主要农业旱灾脆弱性指标数据分析评价,进一步验证了其可行性与有效性.

Address to the multidimensionality and the uncorrelation of agricultural drought vulner-ability evaluation indicators a new method, projection pursuit based on the improved shuffled frog leaping algorithm, is used to evaluate the agricultural drought vulnerability.By combining the chemotaxis operation of bacteria foraging algorithm with the shuffled frog leaping algorithm, the method enhances the individual refinement searching ability within solution domain and improves the optimizing accuracy and stability of the projection pursuit model. The proposed method searches for the optimal projection direction and acquires evaluation index weights basing on characteristics of sample data, accordingly,reducing the interference of artificial empowerment with evaluation results of objectivity.The feasibility and validity of the method are verified by the analyzing and evaluating the agricultural drought vulnerability data in 14 regions of Gansu province.