基于SARIMA-LSTM的区域用电消耗预测研究
Research on Regional Electricity Consumption Forecasting Based on SARIMA-LSTM
郭斌 1熊显名1
作者信息
- 1. 桂林电子科技大学光电工程学院,广西 桂林 541000
- 折叠
摘要
针对季节性、节假日因素对区域用电消耗的影响,并为更好地管理和规划特定区域的电力供应提供有价值的见解.提出了一种混合模型,结合SARIMA和LSTM来预测区域用电消耗.提出的SARIMA-LSTM模型旨在捕捉电力消耗数据中的时间依赖性和季节性模式.利用历史用电数据,并采用SARIMA来捕捉线性依赖性,同时采用LSTM来捕捉非线性和长期依赖性,将两个方法的结果进行叠加,该混合模型使用特定区域的用电数据进行训练和评估.分别与SARIMA和LSTM单一模型相比,结果表明,SARIMA-LSTM模型在准确预测区域用电消耗方面优于单一的SARIMA和LSTM模型.
Abstract
Aiming to investigates the impact of seasonal and holiday factors on regional electricity consumption and pro-vides valuable insights for better managing and planning electricity supply in specific regions.A hybrid model combining SARIMA and LSTM is proposed to forecast regional electricity consumption.The proposed SARIMA-LSTM model aims to capture both the temporal dependencies and seasonal patterns in electricity consumption data.Historical electricity con-sumption data is utilized,with SARIMA capturing linear dependencies and LSTM capturing nonlinear and long-term depen-dencies.The results of both methods are combined in the hybrid model,which is trained and evaluated using electricity data from a specific region.
关键词
SARIMA模型/长短期记忆网络/区域用电消耗Key words
SARIMA model/LSTM/regional electricity consumption引用本文复制引用
基金项目
国家重点研发计划(2022YFF0605502)
国家科技重大专项(2017ZX02101007-003)
国家自然科学基金(61965005)
国家自然科学基金(62205076)
广西自然科学基金(2019GXNSFDA185010)
广西壮族自治区重点研发计划(AB22035047)
上海市在线检测与控制技术重点实验室开放基金(ZX2021104)
出版年
2024
NETL
NSTL
万方数据