耿鹏云,安磊,王鑫

• 1:国网冀北电力有限公司经济技术研究院
• 2:国网冀北电力有限公司建设部

摘要(Abstract)：

针对目前输电工程造价技术指标过多,影响因素比较复杂,导致输电工程造价估算困难,设计概算审查难以达到理想效果的问题。建立基于数据挖掘技术的输变电工程造价预测模型,其采用数据挖掘技术来判断不同工程技术指标对工程造价所造成的影响级别,同时能够自动查询错误、异常或者不合理的数据,降低了人为因素的影响,并通过支持向量机来对样本数据进行样本学习,从而建立输变电工程造价预测模型。测试结果表明,该模型预测结果相对误差低,其能够准确预测输变电工程的造价,且对于造价预算具有一定的参考价值。

关键词(KeyWords)： 数据挖掘;输电工程;造价预测模型;支持向量机;样本学习;概算审查

Abstract:

Keywords:

基金项目(Foundation):

作者(Author): 耿鹏云,安磊,王鑫

DOI: 10.16652/j.issn.1004-373x.2018.04.040

参考文献(References)：

• [1]杨永明,王燕,范秀君,等.基于灰关联-神经网络的电力工程造价估算[J].重庆大学学报,2013,36(11):15-20.YANG Yongming,WANG Yan,FAN Xiujun,et al.Cost estimation of power engineering based on grey relational neural network[J].Journal of Chongqing University,2013,36(11):15-20.
• [2]张吉刚,梁娜.基于改进BP模型的我国社会物流总成本预测[J].统计与决策,2014(6):61-63.ZHANG Jigang,LIANG Na.The total cost prediction of China′s social logistics based on the improved BP model[J].Statistics&decision,2014(6):61-63.
• [3]周成杰.基于BP神经网络的工程投资估算方法探讨[J].铁路工程造价管理,2015,30(5):6-9.ZHOU Chengjie.Discussion of project investment estimation method based on BP neural network[J].Railway engineering cost management,2015,30(5):6-9.
• [4]郝胜兰.基于模糊神经网络的房产软件项目成本估算研究[D].大连:大连海事大学,2012.HAO Shenglan.Research on the cost estimation of real estate software project based on fuzzy neural network[D].Dalian:Dalian Maritime University,2012.
• [5]王苏斌,郑海涛,邵谦谦,等.SPSS统计分析[M].北京:机械工业出版社,2003.WANG Subin,ZHENG Haitao,SHAO Qianqian,et al.SPSS statistical analysis[M].Beijing:China Machine Press,2003.
• [6]郭涛,马林东,葛智平.基于EMD和神经网络的短期电力负荷预测[J].电子设计工程,2013,21(21):105-106.GUO Tao,MA Lindong,GE Zhiping.Forecasting of shortterm power load based on EMD and neural network[J].Electronic design engineering,2013,21(21):105-106.
• [7]马立新,郑晓栋,尹晶晶.基于粗糙特征量的短期电力负荷预测[J].电子科技,2016,29(1):40-43.MA Lixin,ZHENG Xiaodong,YIN Jingjing.Short-term load forecasting based on rough characteristic-component algorithm[J].Electronic science and technology,2016,29(1):40-43.
• [8]杜军岗,魏汝祥,刘宝平.基于PSO优化LS-SVM的小样本非线性协整检验与建模研究[J].系统工程理论与实践,2014,34(9):2322-2331.DU Jungang,WEI Ruxiang,LIU Baoping.Nonlinear cointegration test and error correction modeling based on LS-SVM optimized by PSO in small sample[J].Systems engineering-theory&practice,2014,34(9):2322-2331.
• [9]ACHIMUGU P,SELAMAT A.A hybridized approach for prioritizing software requirements based on K-means and evolutionary algorithms[J].Computational intelligence applications in modeling and control,2015,575:73-79.
• [10]CHEN S,XU Z,TANG Y.A hybrid clustering algorithm based on fuzzy C-means and improved particle swarm optimization[J].Arabian journal for science and engineering,2014,39(12):8875-8887.