摘要
传统方法在综合能源业务潜力预测时,未进行数据标准化处理,能源预测值与实际结果不符,为此,本文设计一种基于时序分析的综合能源业务潜力预测模型。通过异常值剔除、缺失值补足、数据标准化和时间序列编写等步骤完成样本数据预处理过程;通过分解时间序列判断预处理后数据时间序列平稳性,构建综合能源业务潜力预测模型,输入测试数据到模型中实现综合能源业务潜力预测。结果表明,应用本模型后,能得出综合能源业务潜力值,且与系统给出的数据接近,证明模型可用性和有效性。
In the traditional method,the data is not standardized and the predicted value is inconsistent with the actual results.There-fore,this paper designs a comprehensive energy business potential prediction model based on time series analysis.Through the steps of outlier elimination,missing value complement,data standardization and time series compilation,the sample data preprocessing process is completed.By decomposing the time series to judge the stability of the preprocessed data time series,the comprehensive energy business potential prediction model is constructed,and the test data are input into the model to realize the comprehensive energy business potential prediction.The results show that the potential value of comprehensive energy business can be obtained after applying the model,which is close to the data given by the sys-tem,which proves the availability and effectiveness of the model.
作者
王卫公
江晶晶
朱佳盈
WANG Wei-gong;JIANG Jing-jing;ZHU Jia-ying(State Grid Shanghai Shibei Electric Power Supply Company,Shanghai 200072 China)
出处
《自动化技术与应用》
2022年第3期137-140,157,共5页
Techniques of Automation and Applications
关键词
时序分析
综合能源业务潜力
预测模型
缺失值补足
time series analysis
comprehensive energy business potential
prediction model
missing value complement