A hybrid prediction method for short‐term load based on temporal convolutional networks and attentional mechanisms
Abstract Accurate power load prediction is an important guide for power system planning and operation. High‐ or low‐load prediction results will affect the operation of the power system. In recent years, deep learning technology represented by convolution neural network (CNN) and transformer has bee...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2024-03-01
|
| Series: | IET Generation, Transmission & Distribution |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/gtd2.12798 |
| _version_ | 1797278878841962496 |
|---|---|
| author | Min Li Hangwei Tian Qinghui Chen Mingle Zhou Gang Li |
| author_facet | Min Li Hangwei Tian Qinghui Chen Mingle Zhou Gang Li |
| author_sort | Min Li |
| collection | DOAJ |
| description | Abstract Accurate power load prediction is an important guide for power system planning and operation. High‐ or low‐load prediction results will affect the operation of the power system. In recent years, deep learning technology represented by convolution neural network (CNN) and transformer has been proved to be suitable for power load prediction. This paper proposes a new short‐term power load hybrid forecasting model, called channel enhanced attention (CEA) and temporal convolutional network (TCN)‐based transformer comprehensive forecasting model. This method combines the short‐term feature extraction ability of TCN with the long‐term dependent capture ability of transformer for short‐term load forecasting. And the CEA designed in this study is added to improve the prediction accuracy. On the same dataset, the designed model predicts power load mean square errors of 0.056 and 0.146 for the next 24 h and the next week, respectively, which is 0.002 to 0.073 and 0.012 to 0.024 lower than the baseline model. The experimental results show that the hybrid short‐term power load prediction model proposed in this paper is significantly better than the existing methods. The predicted curve is in agreement with the actual charge change, which provides a good guidance for short‐term power load prediction. |
| first_indexed | 2024-03-07T16:15:28Z |
| format | Article |
| id | doaj.art-fa4f8271f83048c5884a5459ffb772da |
| institution | Directory Open Access Journal |
| issn | 1751-8687 1751-8695 |
| language | English |
| last_indexed | 2024-03-07T16:15:28Z |
| publishDate | 2024-03-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Generation, Transmission & Distribution |
| spelling | doaj.art-fa4f8271f83048c5884a5459ffb772da2024-03-04T11:27:49ZengWileyIET Generation, Transmission & Distribution1751-86871751-86952024-03-0118588589810.1049/gtd2.12798A hybrid prediction method for short‐term load based on temporal convolutional networks and attentional mechanismsMin Li0Hangwei Tian1Qinghui Chen2Mingle Zhou3Gang Li4Shandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan ChinaShandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan ChinaShandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan ChinaShandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan ChinaShandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan ChinaAbstract Accurate power load prediction is an important guide for power system planning and operation. High‐ or low‐load prediction results will affect the operation of the power system. In recent years, deep learning technology represented by convolution neural network (CNN) and transformer has been proved to be suitable for power load prediction. This paper proposes a new short‐term power load hybrid forecasting model, called channel enhanced attention (CEA) and temporal convolutional network (TCN)‐based transformer comprehensive forecasting model. This method combines the short‐term feature extraction ability of TCN with the long‐term dependent capture ability of transformer for short‐term load forecasting. And the CEA designed in this study is added to improve the prediction accuracy. On the same dataset, the designed model predicts power load mean square errors of 0.056 and 0.146 for the next 24 h and the next week, respectively, which is 0.002 to 0.073 and 0.012 to 0.024 lower than the baseline model. The experimental results show that the hybrid short‐term power load prediction model proposed in this paper is significantly better than the existing methods. The predicted curve is in agreement with the actual charge change, which provides a good guidance for short‐term power load prediction.https://doi.org/10.1049/gtd2.12798attentiondeep learningshort‐term load forecastingtemporal convolutional networktime series |
| spellingShingle | Min Li Hangwei Tian Qinghui Chen Mingle Zhou Gang Li A hybrid prediction method for short‐term load based on temporal convolutional networks and attentional mechanisms IET Generation, Transmission & Distribution attention deep learning short‐term load forecasting temporal convolutional network time series |
| title | A hybrid prediction method for short‐term load based on temporal convolutional networks and attentional mechanisms |
| title_full | A hybrid prediction method for short‐term load based on temporal convolutional networks and attentional mechanisms |
| title_fullStr | A hybrid prediction method for short‐term load based on temporal convolutional networks and attentional mechanisms |
| title_full_unstemmed | A hybrid prediction method for short‐term load based on temporal convolutional networks and attentional mechanisms |
| title_short | A hybrid prediction method for short‐term load based on temporal convolutional networks and attentional mechanisms |
| title_sort | hybrid prediction method for short term load based on temporal convolutional networks and attentional mechanisms |
| topic | attention deep learning short‐term load forecasting temporal convolutional network time series |
| url | https://doi.org/10.1049/gtd2.12798 |
| work_keys_str_mv | AT minli ahybridpredictionmethodforshorttermloadbasedontemporalconvolutionalnetworksandattentionalmechanisms AT hangweitian ahybridpredictionmethodforshorttermloadbasedontemporalconvolutionalnetworksandattentionalmechanisms AT qinghuichen ahybridpredictionmethodforshorttermloadbasedontemporalconvolutionalnetworksandattentionalmechanisms AT minglezhou ahybridpredictionmethodforshorttermloadbasedontemporalconvolutionalnetworksandattentionalmechanisms AT gangli ahybridpredictionmethodforshorttermloadbasedontemporalconvolutionalnetworksandattentionalmechanisms AT minli hybridpredictionmethodforshorttermloadbasedontemporalconvolutionalnetworksandattentionalmechanisms AT hangweitian hybridpredictionmethodforshorttermloadbasedontemporalconvolutionalnetworksandattentionalmechanisms AT qinghuichen hybridpredictionmethodforshorttermloadbasedontemporalconvolutionalnetworksandattentionalmechanisms AT minglezhou hybridpredictionmethodforshorttermloadbasedontemporalconvolutionalnetworksandattentionalmechanisms AT gangli hybridpredictionmethodforshorttermloadbasedontemporalconvolutionalnetworksandattentionalmechanisms |