A power load forecasting method using cosine similarity and a graph convolutional network
To address the challenges of existing power load forecasting models， which struggle to deeply extract spatiotemporal correlation features and exhibit weak generalization capabilities—failing to simultaneously manage both short-term and long-term forecasting—this study proposes a multi-user power loa...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
zhejiang electric power
2025-01-01
|
| Series: | Zhejiang dianli |
| Subjects: | |
| Online Access: | https://zjdl.cbpt.cnki.net/WKE3/WebPublication/paperDigest.aspx?paperID=8c8abcbf-8259-4108-ac39-48caede8c4ea |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823857103016558592 |
|---|---|
| author | JI Shan JIANG Wei JING Xin |
| author_facet | JI Shan JIANG Wei JING Xin |
| author_sort | JI Shan |
| collection | DOAJ |
| description | To address the challenges of existing power load forecasting models， which struggle to deeply extract spatiotemporal correlation features and exhibit weak generalization capabilities—failing to simultaneously manage both short-term and long-term forecasting—this study proposes a multi-user power load forecasting method using cosine similarity and a global-local collaborative graph convolutional network. First， cosine similarity is utilized to learn similar patterns between load data from different nodes， allowing for the deep extraction of spatiotemporal correlation features. Second， a collaborative modeling approach is applied to static global factors and dynamic local factors that influence power load trends， enhancing the model’s generalization ability. Finally， extensive experiments on a real-world dataset demonstrate the method’s effectiveness and robustness in forecasting both short-term and long-term load series. |
| format | Article |
| id | doaj-art-f20643a135304d4c91237bdf4b85c99d |
| institution | Kabale University |
| issn | 1007-1881 |
| language | zho |
| publishDate | 2025-01-01 |
| publisher | zhejiang electric power |
| record_format | Article |
| series | Zhejiang dianli |
| spelling | doaj-art-f20643a135304d4c91237bdf4b85c99d2025-02-12T00:54:58Zzhozhejiang electric powerZhejiang dianli1007-18812025-01-01441687510.19585/j.zjdl.2025010071007-1881（2025）01-0068-08A power load forecasting method using cosine similarity and a graph convolutional networkJI Shan0JIANG Wei1JING Xin2Hangzhou Data Far East Technology Pty. Ltd.， Hangzhou 310007， ChinaCenter of Technology and Innovation State Grid Zhejiang Electric Power Co.， Ltd.， Hangzhou 310007， ChinaHangzhou Data Far East Technology Pty. Ltd.， Hangzhou 310007， ChinaTo address the challenges of existing power load forecasting models， which struggle to deeply extract spatiotemporal correlation features and exhibit weak generalization capabilities—failing to simultaneously manage both short-term and long-term forecasting—this study proposes a multi-user power load forecasting method using cosine similarity and a global-local collaborative graph convolutional network. First， cosine similarity is utilized to learn similar patterns between load data from different nodes， allowing for the deep extraction of spatiotemporal correlation features. Second， a collaborative modeling approach is applied to static global factors and dynamic local factors that influence power load trends， enhancing the model’s generalization ability. Finally， extensive experiments on a real-world dataset demonstrate the method’s effectiveness and robustness in forecasting both short-term and long-term load series.https://zjdl.cbpt.cnki.net/WKE3/WebPublication/paperDigest.aspx?paperID=8c8abcbf-8259-4108-ac39-48caede8c4eamulti-node power load forecastingspatiotemporal correlation featurescosine similaritygraph convolutional |
| spellingShingle | JI Shan JIANG Wei JING Xin A power load forecasting method using cosine similarity and a graph convolutional network Zhejiang dianli multi-node power load forecasting spatiotemporal correlation features cosine similarity graph convolutional |
| title | A power load forecasting method using cosine similarity and a graph convolutional network |
| title_full | A power load forecasting method using cosine similarity and a graph convolutional network |
| title_fullStr | A power load forecasting method using cosine similarity and a graph convolutional network |
| title_full_unstemmed | A power load forecasting method using cosine similarity and a graph convolutional network |
| title_short | A power load forecasting method using cosine similarity and a graph convolutional network |
| title_sort | power load forecasting method using cosine similarity and a graph convolutional network |
| topic | multi-node power load forecasting spatiotemporal correlation features cosine similarity graph convolutional |
| url | https://zjdl.cbpt.cnki.net/WKE3/WebPublication/paperDigest.aspx?paperID=8c8abcbf-8259-4108-ac39-48caede8c4ea |
| work_keys_str_mv | AT jishan apowerloadforecastingmethodusingcosinesimilarityandagraphconvolutionalnetwork AT jiangwei apowerloadforecastingmethodusingcosinesimilarityandagraphconvolutionalnetwork AT jingxin apowerloadforecastingmethodusingcosinesimilarityandagraphconvolutionalnetwork AT jishan powerloadforecastingmethodusingcosinesimilarityandagraphconvolutionalnetwork AT jiangwei powerloadforecastingmethodusingcosinesimilarityandagraphconvolutionalnetwork AT jingxin powerloadforecastingmethodusingcosinesimilarityandagraphconvolutionalnetwork |