摘要
为使区域水资源优化配置决策过程更加合理有效,从系统净收益、供水保证率、工程投资费用和系统环境影响等层面建立水资源配置决策系统。基于离差最大化算法(Maximum Deviation,MD)和线性分派法(Linear Assignment Method,LAM)建模思想,考虑决策方案原属性值的差异性及其属性本身重要度,利用EW型区间距离研究决策方案属性值的总离差和最大,进而构建了水资源配置方案优选的IMD-LAM决策算法,并与优势关系排序决策算法和加权离差最大化决策算法进行对比验证。实例研究表明:该算法可有效避免决策方案属性值差异性较小及区分度不高的缺陷,提高了决策指标属性信息的利用率,使得评价结果更具层次性与可靠性。
In order to make the decision-making process of optimal allocation of regional water resources more reasonable and effective,the paper established the decision-making system of water resources allocation from system net income,water supply guarantee rate,project investment cost and system environmental impact.On the basic of the modeling idea with Maximum Deviation algorithm(MD)and Linear Assignment Method(LAM),considering the difference of original attribute value for decision scheme and importance degree itself,the IMD-LAM decision-making algorithm for optimizing water resources allocation scheme is constructed in this paper,as well as use EW Interval Distance to study the total deviation and maximum of attribute value of decision schemes.Moreover,the proposed algorithm is validated with dominance ranking decision-making algorithm and weighted deviation maximization decision-making algorithm.Case study shows that the algorithm can effectively avoid the shortcomings of small difference in attribute values and low discrimination of schemes,which has improved the attribute information of decision indicators and make the evaluation results more hierarchical and reliable.
作者
王飞
谭征
王威
WANG Fei;TAN Zheng;WANG Wei(Key Laboratory of Human Settlements and Energy Conservation,Nanyang Institute of Technology,Nanyang 473004,China;Institute of Earthquake Resistance and Disaster Reduction,Beijing University of Technology,Beijing 100124,China)
出处
《给水排水》
CSCD
北大核心
2021年第S02期466-472,共7页
Water & Wastewater Engineering
基金
河南省高等学校重点科研项目(20A560017)
河南省科技攻关项目(202102310246)
河南省科技攻关项目(212102210384)
河南省档案科技项目计划(2021-R-20)
南阳理工学院交叉学科项目(230080)
关键词
水资源配置
优选决策
改进离差最大化
线性分配法
Water resources allocation
Decision making
Improving maximization deviation(IMD)
Linear assignment method(LAM)