基于主成分分析法的输电网规划方案综合决策  被引量:100

Comprehensive Decision-Making of Alternative Transmission Network Planning Based on Principal Component Analysis

在线阅读下载全文

作  者:聂宏展[1] 聂耸[2] 乔怡[3] 吕盼[3] 

机构地区:[1]东北电力大学电气工程学院,吉林省吉林市132012 [2]华北电力大学控制与计算机工程学院,河北省保定市071003 [3]新疆电力设计院,新疆维吾尔自治区乌鲁木齐市830002

出  处:《电网技术》2010年第6期134-138,共5页Power System Technology

摘  要:针对目前解决输电网规划方案综合决策问题的主流方法完全依赖专家评判确定指标权重具有主观随意性大的弊端,将多元统计学中的主成分分析法应用于输电网规划方案综合决策问题。主成分分析法可以用少数几个不相关的主成分替代原来相互关联的众多指标即主成分,是对原指标数据信息的综合与简化,同时根据主成分对系统总方差贡献率确定主成分权重。这种依赖客观数据特征确定权重的客观赋权方法,降低了评价过程中的主观因素和不确定因素。最后结合算例介绍了采用主成分分析法求解输电网规划方案综合决策问题的具体步骤。算例结果证明采用主成分分析法决策过程科学,决策结果合理、可信。At present in the comprehensive decision- making of alternative power transmission network planning, there is drawback of subjectivity and arbitrariness in the determined index weights because these methods completely depend on expert evaluation. To solve this problem, the principal component analysis (PCA) in multivariate statistics is applied to comprehensive decision-making of transmission network planning, because PCA can utilize a few of uncorrelated principal components to substitute original interdependent indices, i.e., the principal components integrate and simplify original index data information, meanwhile according to the principal components' contribution rate to total variance of the system the weights of principal components can be determined. The influences of subjective factors and uncertain factors during the evaluation are reduced by such an objective weight-determining method that determines weights by objective data. Combining with calculation example, concrete procedures to solve comprehensive decision-making for alternative power transmission planning by PCA are presented in detail. Calculation results show that the decision-making based on PCA is satisfied and the decisionmaking result is credible.

关 键 词:优化规划 综合决策 主成分分析法 

分 类 号:TM715[电气工程—电力系统及自动化]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象