张波,杨轶涵,胡锡燎,姜霓裳
ZHANG Bo,YANG Yihan,HU Xiliao,JIANG Nichang

• 1:国网浙江省电力有限公司经济技术研究院
• 2:浙江大学软件学院

摘要(Abstract)：

为了进行输变电工程中目标工程量的评估，提出基于隔离森林-MLP（多层感知机）的工程量智能评估方法。运用皮尔逊相关系数和置换特征重要性算法，对各工程量与相关技术类特征的关系进行了探究和特征筛选。在此基础上，提出一种基于隔离森林的数据合理性筛选方法，利用5年的历史数据训练基于MLP的神经网络预测模型，并在测试集上验证了利用所提方法进行工程量评估的准确性。算例仿真表明，所提特征筛选方法可以提升神经网络的评估精度，同时所提方法可以给出工程量合理性的粗估计与智能评估。该评估方法可用于自动化评估工程量合理性，促进电网基建数字化转型。
To evaluate quantities in power transmission and transformation projects, an intelligent evaluation method of quantity threshold based on isolated forest and multi-layer perceptron is proposed. Firstly, the Pearson correlation coefficient and permutation feature importance are applied to explore the relationship between various quantities and related technical features and select the features. On this basis, a neural network predictive model is trained by using five-year historical data. Aiming at the problem that the historical data may contain a lot of noises, a data rationality selection method based on isolation forest is proposed and the accuracy of using the proposed to evaluate quantities is verified in test set. The case simulation shows that the proposed feature selection method can improve the evaluation accuracy of neural network. At the same time, the proposed method can provide both rough estimation and intelligent evaluation on the rationality of quantities and can automatically evaluate the quantity rationality for promoting the digital transformation of power grid infrastructure.

关键词(KeyWords)： 输变电工程量;合理性阈值;隔离森林;多层感知机网络;皮尔逊相关性;岭回归;置换特征重要性;自动化评估
power transmission and transformation quantities;rationality threshold;isolation forest;multi-layer perceptron network;Pearson correlation coefficient;ridge regression;permutation feature importance;automated assessment

Abstract:

Keywords:

基金项目(Foundation): 国网浙江省电力有限公司科技项目（5211JY20001V）

作者(Author): 张波,杨轶涵,胡锡燎,姜霓裳
ZHANG Bo,YANG Yihan,HU Xiliao,JIANG Nichang

DOI: 10.19585/j.zjdl.202308008

参考文献(References)：

• [1] ALBRECHT A J.Measuring application development productivity[C]//Proceedings of the Joint SHARE/GUIDE/IBM Application Development Symposium.[S.l.:s.n.],1979:83-92.
• [2]梅生伟，朱建全.智能电网中的若干数学与控制科学问题及其展望[J].自动化学报，2013,39(2):119-131.MEI Shengwei,ZHU Jianquan.Mathematical and control scientific issues of smart grid and its prospects[J]. Acta Automatica Sinica,2013,39(2):119-131.
• [3]马韬韬，郭创新，朱传柏，等.一种电网智能控制中心架构研究[J].电力科学与技术学报，2009,24(2):15-19.MA Taotao,GUO Chuangxin,ZHU Chuanbai,et al. A novel architecture of power grid control center[J].Journal of Electric Power Science and Technology,2009,24(2):15-19.
• [4]解武辉.电网建筑工程对象化分类及工程量计算方法[C]//电网工程造价管理优秀论文集.[出版地不详：出版者不详]，2011:295-301.XIE Wuhui.Objectified classification and engineering quantity calculation method of construction project in power grid[C]//Excellent Proceedings of Cost Management of Power Grid Engineering.[S.l.:s.n.],2011:295-301.
• [5]俞敏，王愿翔，闫园，等.架空线路改造工程造价的组合预测方法[J].电力科学与技术学报，2020,35(1):24-30.YU Min,WANG Yuanxiang,YAN Yuan,et al.A combinational forecasting method for predicting the cost of an overhead line reconstruction project[J].Journal of Electric Power Science and Technology,2020,35(1):24-30.
• [6]肖立华，张博，胡伟，等.基于机器学习的电网工程量计价预测模型[J].沈阳工业大学学报，2021,43(3):241-246.XIAO Lihua,ZHANG Bo,HU Wei,et al. Forecasting model for power grid engineering quantity pricing based on machine learning[J]. Journal of Shenyang University of Technology,2021,43(3):241-246.
• [7]陈光冲.基于历史中标数据的投标价合理性识别系统研究[D].南京：东南大学，2017.CHEN Guangchong.Research on bid price rationality identification system based on historical bid winning data[D].Nanjing:Southeast University,2017.
• [8] LIU F T,TING K M,ZHOU Z H.Isolation forest[C]//2008 Eighth IEEE International Conference on Data Mining,December 15-19,2008,Pisa,Italy:413-422.
• [9]胡洁.高维数据特征降维研究综述[J].计算机应用研究，2008,25(9):2601-2606.HU Jie. Survey on feature dimension reduction for highdimensional data[J].Application Research of Computers,2008,25(9):2601-2606.
• [10]牛勇，李华鹏，刘阳惠，等.超高维数据特征筛选方法综述[J].应用概率统计，2021,37(1):69-110.NIU Yong,LI Huapeng,LIU Yanghui,et al.Overview of feature screening methods for ultra-high dimensional data[J].Chinese Journal of Applied Probability and Statistics,2021,37(1):69-110.
• [11]陈宇韶，唐振军，罗扬，等.皮尔森优化结合Xgboost算法的股价预测研究[J].信息技术，2018,42(9):84-89.CHEN Yushao,TANG Zhenjun,LUO Yang,et al. Research on stock price prediction based on Xgboost algorithm with Pearson optimization[J].Information Technology,2018,42(9):84-89.
• [12] BREIMAN L. Random forests[J]. Machine Learning,2001,45(1):5-32.
• [13] ROSENBLATT F.The perceptron:a probabilistic model for information storage and organization in the brain[J].Psychological Review,1958,65(6):386-408.
• [14] POPESCU M C,BALAS V E,POPESCU L P,et. al.Multilayer perceptron and neural networks[J]. Wseas Transactions on Circuits and System,2009,8(7):579-588.
• [15] IAN G,YOSHUA B.深度学习[M].北京：人民邮电出版社，2017.
• [16] HORNIK K,STINCHCOMBE M,WHITE H. Multilayer feedforward networks are universal approximators[J].Neural Networks,1989,2(5):359-366.
• [17] CSáJI B C.Approximation with artificial neural networks[D].Budapest,Hungary:Etvs Lornd University,2001.