Multi-featured power load forecasting based on VMD-SSA-BiLSTM

曹 帅; 尹 杰; 李艺丰; 石璐杉; 赵玉林; 丁超杰; 周 霞; 刘贵宇

doi:10.3969/j.issn.1009-1831.2024.06.014

您当前的位置：

首页 >

文章列表页 >

Multi-featured power load forecasting based on VMD-SSA-BiLSTM

更新时间：2026-03-05

- Multi-featured power load forecasting based on VMD-SSA-BiLSTM
- Vol. 26, Issue 6, (2024)
- 作者机构：
  
  1. 国网江苏省电力有限公司,南京,210008
  2. 国电南瑞科技股份有限公司,南京,211000
  3. 南京邮电大学先进技术研究院,南京,210023
- 作者简介：
- 基金信息：
- DOI：10.3969/j.issn.1009-1831.2024.06.014
  CLC：
- Published：2024
- 稿件说明：
移动端阅览
曹帅, 尹杰, 李艺丰, et al. Multi-featured power load forecasting based on VMD-SSA-BiLSTM[J]. 2024, 26(6).
DOI：

曹帅, 尹杰, 李艺丰, et al. Multi-featured power load forecasting based on VMD-SSA-BiLSTM[J]. 2024, 26(6). DOI： 10.3969/j.issn.1009-1831.2024.06.014.

摘要

为了全面发掘负荷数据中的时序信息和天气信息

以提高电力负荷预测的精准度

提出一种基于变分模态分解（variation-al mode decomposition

VMD）和双向长短期记忆神经网络（bi-direction long short-term memory

BiLSTM）相耦合的多维时序电力负荷预测方法。首先通过对多维天气信息和时序信息进行相关性分析

选取相关性高的特征向量作为输入

同时利用VMD将原始负荷数据分解为不同频率的本征模态函数（intrinsic mode function

IMF）

然后将IMF和相关性高的特征向量输入到经过麻雀搜索算法（sparrow search algorithm

SSA）优化的BiLSTM神经网络进行预测

最后叠加IMF的预测值

得到最终电力负荷预测值

并进行实例验证。与BiLSTM、VMD-BiLSTM模型相比

VMD-SSA_BiLSTM模型能够充分挖掘数据中的时序信息和天气信息

可以提升多维负荷数据的预测精度。

Abstract

To fully explore the timing and weather information in load data and improve the accuracy of power load prediction

a neural network based on variational mode decomposition（VMD）and bi-directional long short-term memory（BiLSTM）is proposed. Multi-dimensional sequential power load forecasting method leverages the strengths of VMD and BiLSTM to improve the accuracy of power load prediction. Firstly

through correlation analysis of multi-dimensional weather information and time sequence information

feature vectors with high correlation are selected as inputs. Meanwhile

VMD is used to decompose the original load data into intrinsic mode functions（IMF）of different frequencies. Then

the IMF and feature vector with high correlation are input to BiLSTM neural network optimized by sparrow search algorithm（SSA）for prediction. Finally

the predicted value of IMF is superimposed to obtain the final predicted value of power load. Load forecasting data set of 2016 electrical mathematical contest in modeling is used as an example to verify. Compared with BiLSTM and VMD-BiLSTM model

VMD-SSA-BiLSTM model can fully mine timing and weather information in data

and improve the prediction accuracy of multidimensional load data.

关键词

Keywords

references

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Design and implementation of virtual power plant system supporting dynamic aggregation of equipment

Short-term power load forecasting based on load decomposition and identification

Load forecasting and adjustable potential analysis of silicon carbide enterprises based on CNN-Attention-BiLSTM

Study on the influence of environmental climate change on the load of fishery electrical equipment

Short-term load probability forecasting based on VaR and integrated neural network quantile regression

Related Author

王孟强

黄澍

孟洪民

诸葛雪迎

刘泽三

徐哲男

李子钰

朱俊澎

Related Institution

国网信息通信产业集团有限公司

国网河南省电力公司经济技术研究院

河海大学电气与动力工程学院

国网甘肃省电力公司兰州供电公司

国网甘肃省电力公司

AI问答

⁰