Multimodal Feature Extraction and Fusion Deep Neural Networks for Short-Term Load Forecasting
Precise and reliable forecasting of short-term electricity load is essential to the development of smart grids. Particularly, deep neural networks (DNNs) are widely utilized for the prediction of short-term electricity load due to their automatic feature extraction ability. However, these available...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9218993/ |
| _version_ | 1818933646817492992 |
|---|---|
| author | Zhengmin Kong Chenggang Zhang He Lv Feng Xiong Zhuolin Fu |
| author_facet | Zhengmin Kong Chenggang Zhang He Lv Feng Xiong Zhuolin Fu |
| author_sort | Zhengmin Kong |
| collection | DOAJ |
| description | Precise and reliable forecasting of short-term electricity load is essential to the development of smart grids. Particularly, deep neural networks (DNNs) are widely utilized for the prediction of short-term electricity load due to their automatic feature extraction ability. However, these available stacked deep-learning models may lose some temporal features or spatial features of original input data. To capture more comprehensive information, in this article, we present an integration scheme based on empirical mode decomposition (EMD), similar day methods, and DNNs to perform short-term load forecasting. It is especially worth noting that the electricity price is also an important factor for load variation, which is considered in our proposed scheme. Specifically, there are two primary layers: a feature extraction layer and a forecasting layer. In the feature extraction layer, EMD is applied to decompose load time series into several components, which are arranged into the 2-D input matrix of the convolutional neural network (CNN). Both the output vectors of the CNN and the raw load sequences are fed into the long short-term memory (LSTM) layer. Therefore, the whole EMD based CNN-LSTM approach extracts multimodal spatial-temporal features from input data. Meanwhile, the electricity price data is utilized to obtain multimodal spatial-temporal features in the same way. Additionally, the day and hour information and loads of similar days are to augment extra features for prediction. In the forecasting layer, the forecasting task is accomplished through a fully-connected neural network based on the outputs of the feature extraction layer. Leveraging these techniques enables our proposed scheme to extract more latent features, which significantly improve the accuracy. In order to demonstrate the performance of our proposed scheme, related experiments are conducted on actual data from the electricity market in Singapore. Compared to other available models, our proposed scheme is superior in graphic and numerical results. |
| first_indexed | 2024-12-20T04:51:42Z |
| format | Article |
| id | doaj.art-3af8fcb0f2fc4ce383306f13c12d004f |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-20T04:51:42Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-3af8fcb0f2fc4ce383306f13c12d004f2022-12-21T19:52:50ZengIEEEIEEE Access2169-35362020-01-01818537318538310.1109/ACCESS.2020.30298289218993Multimodal Feature Extraction and Fusion Deep Neural Networks for Short-Term Load ForecastingZhengmin Kong0https://orcid.org/0000-0001-9257-181XChenggang Zhang1https://orcid.org/0000-0003-0349-4854He Lv2Feng Xiong3https://orcid.org/0000-0002-8711-7548Zhuolin Fu4School of Electrical Engineering and Automation, Wuhan University, Wuhan, ChinaSchool of Electrical Engineering and Automation, Wuhan University, Wuhan, ChinaEnOS Cloud Platform Algorithm~Department, Envision Energy Ltd., Shanghai, ChinaSchool of Electrical Engineering and Automation, Wuhan University, Wuhan, ChinaSchool of Electrical Engineering and Automation, Wuhan University, Wuhan, ChinaPrecise and reliable forecasting of short-term electricity load is essential to the development of smart grids. Particularly, deep neural networks (DNNs) are widely utilized for the prediction of short-term electricity load due to their automatic feature extraction ability. However, these available stacked deep-learning models may lose some temporal features or spatial features of original input data. To capture more comprehensive information, in this article, we present an integration scheme based on empirical mode decomposition (EMD), similar day methods, and DNNs to perform short-term load forecasting. It is especially worth noting that the electricity price is also an important factor for load variation, which is considered in our proposed scheme. Specifically, there are two primary layers: a feature extraction layer and a forecasting layer. In the feature extraction layer, EMD is applied to decompose load time series into several components, which are arranged into the 2-D input matrix of the convolutional neural network (CNN). Both the output vectors of the CNN and the raw load sequences are fed into the long short-term memory (LSTM) layer. Therefore, the whole EMD based CNN-LSTM approach extracts multimodal spatial-temporal features from input data. Meanwhile, the electricity price data is utilized to obtain multimodal spatial-temporal features in the same way. Additionally, the day and hour information and loads of similar days are to augment extra features for prediction. In the forecasting layer, the forecasting task is accomplished through a fully-connected neural network based on the outputs of the feature extraction layer. Leveraging these techniques enables our proposed scheme to extract more latent features, which significantly improve the accuracy. In order to demonstrate the performance of our proposed scheme, related experiments are conducted on actual data from the electricity market in Singapore. Compared to other available models, our proposed scheme is superior in graphic and numerical results.https://ieeexplore.ieee.org/document/9218993/Short-term load forecastingempirical mode decompositionsimilar day methodsdeep neural networkstransitional forecasting schemeelectricity price |
| spellingShingle | Zhengmin Kong Chenggang Zhang He Lv Feng Xiong Zhuolin Fu Multimodal Feature Extraction and Fusion Deep Neural Networks for Short-Term Load Forecasting IEEE Access Short-term load forecasting empirical mode decomposition similar day methods deep neural networks transitional forecasting scheme electricity price |
| title | Multimodal Feature Extraction and Fusion Deep Neural Networks for Short-Term Load Forecasting |
| title_full | Multimodal Feature Extraction and Fusion Deep Neural Networks for Short-Term Load Forecasting |
| title_fullStr | Multimodal Feature Extraction and Fusion Deep Neural Networks for Short-Term Load Forecasting |
| title_full_unstemmed | Multimodal Feature Extraction and Fusion Deep Neural Networks for Short-Term Load Forecasting |
| title_short | Multimodal Feature Extraction and Fusion Deep Neural Networks for Short-Term Load Forecasting |
| title_sort | multimodal feature extraction and fusion deep neural networks for short term load forecasting |
| topic | Short-term load forecasting empirical mode decomposition similar day methods deep neural networks transitional forecasting scheme electricity price |
| url | https://ieeexplore.ieee.org/document/9218993/ |
| work_keys_str_mv | AT zhengminkong multimodalfeatureextractionandfusiondeepneuralnetworksforshorttermloadforecasting AT chenggangzhang multimodalfeatureextractionandfusiondeepneuralnetworksforshorttermloadforecasting AT helv multimodalfeatureextractionandfusiondeepneuralnetworksforshorttermloadforecasting AT fengxiong multimodalfeatureextractionandfusiondeepneuralnetworksforshorttermloadforecasting AT zhuolinfu multimodalfeatureextractionandfusiondeepneuralnetworksforshorttermloadforecasting |