云南省水资源-经济-社会-水生态协调度评价

Coordination degree evaluation on "water resources-economy-society-water ecology" of Yunnan Province

为科学评价云南省2006～2016年水资源-经济-社会-水生态系统协调度水平,引入阴-阳对优化(YYPO)算法、投影寻踪(PP)和正态云模型(CM),构建了YYPO-PP-CM水资源-经济-社会-水生态协调度评价模型。从水资源、经济、社会、水生态系统中遴选出20个指标来构建水资源-经济-社会-水生态系统协调度评价指标体系和等级标准,采用云模型正向发生器来计算各分级评价指标隶属度;选取了8个标准测试函数对YYPO算法的优化性能进行仿真验证,并与粒子群优化(PSO)算法、布谷鸟搜索(CS)算法等4种传统优化算法的性能进行对比。基于PP基本原理,利用云南省2006～2016年水资源-经济-社会-水生态指标数据构造投影寻踪优化目标函数,通过YYPO-PP给出各评价指标权重,同时,根据隶属度矩阵和权重矩阵来计算水资源-经济-社会-水生态协调度评价的分级确定度,并进行评价分析,最后将评价结果与投影寻踪法、模糊评价法的结果进行比较。结果表明:①YYPO算法的寻优精度优于PSO、CS等传统优化算法,具有较好的开发、探索平衡能力和全局极值寻优能力。②YYPO-PP-CM模型将云南省2006～2007年水资源-经济-社会-水生态协调度评价为"极不协调",2008～2009年为"不协调",2010年为"基本协调",2011～2014年为"较协调",2015～2016年为"协调",表明近10 a来云南省水资源-经济-社会-水生态系统协调度水平持续提升。③评价结果与投影寻踪法、模糊评价法的评价结果基本一致。④YYPO-PP-CM模型兼具客观性、模糊性和随机性,既能客观确定评价指标权重,反映水资源-经济-社会-水生态协调度评价分级的定性概念,又能反映隶属程度的不确定性,具有良好的应用价值。

In order to scientifically evaluate the coordination degree of 'water resources-economy-society-aquatic ecosystem' in Yunnan from 2006 to 2016, we introduced YYPO algorithm, projection pursuit(PP) and cloud model(CM) to construct YYPO-PP-CM water resources-economic-social-aquatic ecological coordination degree evaluation model. Twenty indicators were selected from water resources, economy, society and aquatic ecosystem to construct an index system and grading standard. CM forward generator was used to calculate membership of each grade of evaluation index. Eight standard test functions were chosen to verify the optimization performance of YYPO algorithm, and compared it with the performance of other four traditional optimization algorithms including PSO algorithm and CS algorithm. Based on the principle of PP, we employed data from the system to construct PP optimized objective function;the weight of each evaluation index was given by YYPO-PP, grading degree was calculated and analyzed according to membership matrix and weight matrix, and the result was compared to that of the PP method and fuzzy evaluation method. The results show that:① YYPO is superior than traditional methods such as PSO and CS in accuracy, being good at developing and exploring balance and global extreme optimization. ②Yunnan Province is evaluated as 'extremely uncoordinated' in its 'water resources-economy-society-aquatic ecosystem' between 2006 and 2007, 'Uncoordinated' between 2008 and 2009, 'Roughly Coordinated' for 2010, 'Coordinated' between 2011 and 2014, and 'well coordinated' between 2015 and 2016, which indicated that for the past ten years, its coordination degree was gradually improved. ③ evaluation results were in consistent with the results obtained by PP method and fuzzy evaluation method. ④ YYPO-PP-CM model is a combination of objectivity, fuzziness and randomness, not only can it determine the weight of evaluation indexes, reflecting the qualitative concept for 'water resources-economy-society-aquatic ecology' coordination and evaluation grading certainty, but also has certain application value.