摘要
针对当前高能耗产业用电量预测模型对用电数据质量依赖较高且时序分析能力不佳,导致用电量预测准确率较低、计算耗时较长的问题,提出基于自回归分布滞后的高能耗产业用电量预测方法。采用大数据挖掘技术中的K均值算法完成数据分析,应用ADF检测对数据展开平稳性检测,选取因果关系检验方法完成指标数据的检测与关联分析,获取预测模型指标。应用自回归分布滞后方法,构建高能耗产业用电量预测模型。实验结果表明,该方法的数据整理分析能力较强,进一步提升了用电量预测结果的准确性,可降低用电量预测耗时。
In response to the current high dependence of electricity consumption prediction models on the quality of electricity consumption data and poor time series analysis ability,which leads to low accuracy and long calculation time in electricity consumption prediction,a high energy consumption industry electricity consumption prediction method based on Autoregressive Distributed Lag(ARDL)is proposed.The K-means algorithm in big data mining technology is adopted to complete data analysis,and the ADF detection is applied to perform stationarity detection on the data.Nextly,causal relationship testing methods are selected to complete indicator data detection and correlation analysis,and obtain predictive model indicators.The ARDL method is used to construct a prediction model for electricity consumption in high-energy consuming industries.The experiment results show that the method has strong data organization and analysis capabilities,therefore,further improving the accuracy of electricity consumption prediction results and reducing the time consumption of electricity consumption prediction.
作者
刘逾尔
陈显枝
陈思琦
荀超
刘沙沙
LIU Yu-er;CHEN Xian-zhi;CHEN Si-qi;XUN Chao;LIU Sha-sha(State Grid Fuzhou Power Supply Company,Fuzhou 350000,China;State Grid Fujian Electric Power Co.,Ltd.,Fuzhou 350000,China;Beijing Guodiantong Network Technology Co.,Ltd.,Beijing 100000,China)
出处
《信息技术》
2025年第4期141-146,共6页
Information Technology
关键词
K均值算法
高能耗产业
自回归分布滞后
用电量分析
K-means algorithm
high energy consuming industries
Autoregressive Distributed Lag
electricity consumption analysis
