摘要
针对输电工程造价预测精度不高的问题,提出了一种基于T分布随机邻域嵌入降维(TSNE)及改进灰色关联的输电工程造价预测模型。根据输电工程造价的分部与组合原理,构建了输电工程造价预测指标体系。提出了基于TSNE可视化降维的工程量指标约减方法,并建立了基于改进灰色关联的量价双维时空相似特征优选模型。根据造价数据的正态分布统计特性,提出了基于abm的优选样本工程造价均值强化方法,兼顾施工生产力水平差异,使预测结果更好地反映输电工程造价实际。仿真结果表明,110kV、220kV、330kV、500kV输电工程造价预测值与结算实际值的误差分别为3.19%、2.91%、2.74%、2.56%,均小于5%的误差要求,方法具有良好的预测效果。
Aiming at the problem of low prediction accuracy of transmission project cost, a transmission project cost prediction model based on TSNE and improved grey correlation was established. Firstly, according to the principle of division and combination of transmission project cost, the prediction index system of transmission project cost was constructed. Secondly, the reduction method of engineering quantity index based on TSNE visual dimension reduction was proposed, and the optimization model of two-dimensional space-time similar characteristics of quantity and price based on improved grey correlation degree was established. Finally, according to the statistical characteristics of normal distribution of cost data, the method of abm was put forward to strengthen the average value of project cost of the optimal samples, taking into account the difference of construction productivity level, so that the prediction results can better reflect the actual cost of transmission project. The simulation results show that the errors between the predicted value of 110kV,220kV,330kV and 500kV transmission project cost and the actual value of settlement are 3.19%,2.91%,2.74% and 2.56% respectively, which are less than the error requirement of 5%,the method has good prediction effect.
作者
宋晓华
汪鹏
张露
寇英芳
SONG Xiao-hua;WANG Peng;ZHANG Lu;KOU Ying-fang(School of Economics and Management,North China Electric Power University,Beijing 102206,China;Beijing Key Laboratory of New Energy and Low-Carbon Development,North China Electric Power University,Beijing 102206,China;State Grid Beijing Electric Power Company,Beijing 100031,China)
出处
《计算机仿真》
北大核心
2022年第12期116-120,共5页
Computer Simulation
基金
国家自然科学基金(72074074)。
关键词
降维
改进灰色关联
输电工程造价
预测
Dimension reduction
Improved grey correlation
Transmission project cost
Prediction
作者简介
宋晓华(1971-),女(汉族),黑龙江省五常市人,博导、教授,主要研究领域为电力技术经济。;汪鹏(1989-),男(汉族),陕西省延安市人,博士研究生,主要研究领域为电力技术经济。;张露(1999-),女(汉族),贵州省贵阳市人,硕士研究生,主要研究领域为电力技术经济。;寇英芳(1972-),女(汉族),天津人,高级工程师,主要研究领域为工程及物资管理。
