天然气是一种绿色高效的能源,准确预测其需求量对于天然气政策制定、生产和贸易等均具有重大意义。传统的回归分析、灰色模型等方法对天然气需求量预测仅仅考虑时间因素,无法保证预测的准确性。近年来,基于人工神经网络的天然气需求量预测被证实是一种高效准确的方法,但目前研究主要集中于算法优化,而对于天然气需求量影响因素的研究则较少。基于此,采用灰色关联度法（GRA）、平均影响值法（MIV）和主成分分析法（PCA）对天然气需求量影响因素进行了特征筛选,以简化和优化神经网络模型,提高预测的准确性和精度。研究结果表明：① 三种方法均能显著降低神经网络预测误差,其中MIV法效果最佳,预测的平均相对误差从9.077%降至0.983%;② 较之于传统的灰色预测模型,通过三种方法特征筛选的BP神经网络模型对2019-2025年的天然气需求量预测结果基本一致,而灰色模型预测结果偏大,表明所建立模型预测精度较高,结果可靠,可以用于对天然气需求量的中长期预测。
Natural gas is a kind of green and efficient energy. To accurately predict its demand is of great significance to the policy making, production, and trade of natural gas. However, in some traditional prediction methods of both regression analysis and grey model, only time factor is considered, resulting that the prediction accuracy cannot be guaranteed. In recent years, artificial neural network has been proved to be efficient and accurate in natural-gas prediction, but the current research mainly focuses on algorithm optimization, even less on the influence factors of natural-gas demand. So, BP neural network model was established to qualitatively analyzed these factors. Then, feature selection was done on the influence factors by means of grey correlation degree method (GRA), average influence value method (MIV), and principal component analysis (PCA), so as to simplify and optimize the neural network model to improve the prediction accuracy. Results show that (1) these three methods can significantly reduce the prediction error in neural network, among which MIV method as the best one may decrease the average relative prediction error from 9.077% to 0.983%; and (2) the prediction results of natural-gas demand from 2019 to 2025 by the BP neural network model and feature selection by the three methods are basically accordant, while those by the traditional grey model are larger, indicating that this model has high prediction accuracy and reliable prediction results and can be used for predicting the middle- and long-term natural-gas demand.
[1] 杨建红. 中国天然气市场可持续发展分析[J]. 天然气工业,2018,38(4):145-152.
YANG Jianhong.Analysis of sustainable development of natural gas market in China[J]. Natural Gas Industry,2018,38(4):145-152.
[2] 侯玉民. PG市天然气需求预测[D]. 成都:西南石油大学,2014.
HOU Yumin.Prediction of natural gas demand in PG city[D]. Chengdu:Southwest Petroleum University,2014.
[3] 亢亿林. LS州天然气需求预测研究[D]. 成都:西南石油大学,2014.
KANG Yilin.Research on natural gas demand forecast in LS city[D]. Chengdu:Southwest Petroleum University,2014.
[4] 王安,王文虎,胡娇,等. 最优组合预测模型在中国天然气需求预测中的应用[J]. 平顶山学院学报,2018,033(5):32-39.
WANG An,WANG Wenhu,HU Jiao,et al.The Optimal Combined Forecasting Model and its Application in Forecasting the Natural Gas Demand in China[J]. Journal of Pingdingshan University,2018,033(5):32-39.
[5] 李洪兵,曾轶. 灰色回归组合新模型在城市天然气需求预测中的应用[J]. 天然气技术与经济,2020,14(2):72-77.
LI Hongbin,ZENG Yi.A new compound model of gray regression and its application to predicting urban natural-gas demand[J]. Natural Gas Technology and Economy,2020,14(2):72-77.
[6] GEEM Z W,ROPER W E.Energy demand estimation of South Korea using artificial neural network[J]. Energy Policy,2009,37(10):4049-4054.
[7] 罗东坤,徐平. 基于改进BP神经网络的天然气需求预测[J]. 油气田地面工程,2008,(7):20-21.
LUO Dongkun,XU Ping.Forecasting of Natural Gas Demand Based on Improved BP Neural Network[J]. Oil-Gasfield Surface Engineering,2008,(7):20-21.
[8] 冯雪,张金锁,邹绍辉,等. 基于RBF 神经网络非线性集成模型的天然气需求预测[J]. 统计与决策. 2015: 91-93.
FENG Xue,ZHANG Jinsuo,ZOU Shaohui,et al.Prediction of natural gas demand based on RBF neural network nonlinear ensemble model[J]. Statistics and Decision,2015: 91-93.
[9] 邹绍辉,丁治立. 基于DDE-BAG的中国天然气需求预测模型[J]. 中国矿业,2018,27(08):65-72.
ZOU Shaohui,DING Yeli.China natural gas demand forecast model based on DDE-BAG[J]. China Mining Magazine,2018,27(08):65-72.
[10] 陶阳威,孙梅,王小芳. 基于改进的BP神经网络的中国能源需求预测研究[J]. 山西财经大学学报,2010,(S2):3-5.
TAO Yangwei,SUN Mei,WANG Xiaofang.China energy demand forecast based on improved BP neural network[J]. Journal of ShanXi Finance and Economics University,2010,(S2):3-5.
[11] 叶倩. 城市天然气需求预测研究及应用[D]. 重庆:重庆大学,2010.
YE Qian.City natural gas demand forecast research and application[D]. Chongqing:Chongqing University,2010.
[12] XU J,CHEN Y,XIE T,et al.Prediction of triaxial behavior of recycled aggregate concrete using multivariable regression and artificial neural network techniques[J]. Construction and Building Materials,2019,226: 534-554.
[13] CHENG Q,QI Z,ZHANG G,et al.Robust modelling and prediction of thermally induced positional error based on grey rough set theory and neural networks[J]. The International Journal of Advanced Manufacturing Technology,2016,83(5):753-764.
[14] LI J Y,MEN C,QI J F,et al.Impact factor analysis,prediction,and mapping of soil corrosion of carbon steel across China based on MIV-BP artificial neural network and GIS[J]. Journal of Soils and Sediments,2020,20(8):3204-3216.
[15] YAN H,ZHANG J,ZHOU N,et al.Application of hybrid artificial intelligence model to predict coal strength alteration during CO2 geological sequestration in coal seams[J]. Science of The Total Environment,2020,711: 135029.
[16] DOUKAS H,PAPADOPOULOU A,SAVVAKIS N,et al.Assessing energy sustainability of rural communities using Principal Component Analysis[J]. Renewable and Sustainable Energy Reviews,2012,16(4):1949-1957.
[17] PARHIZKAR T,RAFIEIPOUR E,PARHIZKAR A.Evaluation and improvement of energy consumption prediction models using principal component analysis based feature reduction[J]. Journal of Cleaner Production,2021,279: 123866.
[18] 邓宏伟,刘桥. 四川天然气需求影响因素分析及灰色预测[J]. 现代商贸工业,2018,(24):10-12.
DENG Hongwei,LIU Qiao.Analysis on factors of Sichuan natural gas demand and gray prediction[J]. Modern Business Trade Industry,2018,(24):10-12.
[19] 张东旭. 关于国内外天然气资源经济评价的对比分析[J]. 现代商业,2020,564(11):61-62.
ZHANG Dongxu.Comparative analysis on economic evaluation of natural gas resources at home and abroad[J]. Modern Business,2020,564(11):61-62.
[20] 张吉军,李洪兵,孙逸林,等. 基于复合权重的天然气需求量组合预测模型的构建[J]. 天然气技术与经济,2021,15(2):57-63.
ZHANG Jijun,LI Hongbing,SUN Yilin,et al.Coweightbased combination forecast model of natural-gas demand and its application[J]. Natural Gas Technology and Economy,2021,15(2):57-63.
