A Deep-Learning-Based Meta-Modeling Workflow for Thermal Load Forecasting in Buildings: Method and a Case Study
This paper proposes a meta-modeling workflow to forecast the cooling and heating loads of buildings at individual and district levels in the early design stage. Seven input variables, with large impacts on building loads, are selected for designing meta-models to establish the MySQL database. The lo...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-02-01
|
| Series: | Buildings |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-5309/12/2/177 |
| _version_ | 1797482169392693248 |
|---|---|
| author | Yuhao Zhou Yumin Liang Yiqun Pan Xiaolei Yuan Yurong Xie Wenqi Jia |
| author_facet | Yuhao Zhou Yumin Liang Yiqun Pan Xiaolei Yuan Yurong Xie Wenqi Jia |
| author_sort | Yuhao Zhou |
| collection | DOAJ |
| description | This paper proposes a meta-modeling workflow to forecast the cooling and heating loads of buildings at individual and district levels in the early design stage. Seven input variables, with large impacts on building loads, are selected for designing meta-models to establish the MySQL database. The load profiles of office, commercial, and hotel models are simulated with EnergyPlus in batches. A sequence-to-sequence (Seq2Seq) model based on the deep-learning method of a one-dimensional convolutional neural network (1D-CNN) is introduced to achieve rapid forecasting of all-year hourly building loads. The method performs well with the load effective hour rate (LEHR) of around 90% and MAPE less than 10%. Finally, this meta-modeling workflow is applied to a district as a case study in Shanghai, China. The forecasting results well match the actual loads with R<sup>2</sup> of 0.9978 and 0.9975, respectively, for the heating and cooling load. The LEHR value of all-year hourly forecasting loads is 98.4%, as well as an MAPE of 4.4%. This meta-modeling workflow expands the applicability of building-physics-based methods and improves the time resolution of conventional data-driven methods. It shows small forecasting errors and fast computing speed while meeting the required precision and convenience of engineering in the building early design stage. |
| first_indexed | 2024-03-09T22:26:28Z |
| format | Article |
| id | doaj.art-d642f2e58bf24e23aaae7cfdad2ec065 |
| institution | Directory Open Access Journal |
| issn | 2075-5309 |
| language | English |
| last_indexed | 2024-03-09T22:26:28Z |
| publishDate | 2022-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Buildings |
| spelling | doaj.art-d642f2e58bf24e23aaae7cfdad2ec0652023-11-23T19:06:09ZengMDPI AGBuildings2075-53092022-02-0112217710.3390/buildings12020177A Deep-Learning-Based Meta-Modeling Workflow for Thermal Load Forecasting in Buildings: Method and a Case StudyYuhao Zhou0Yumin Liang1Yiqun Pan2Xiaolei Yuan3Yurong Xie4Wenqi Jia5Huadian Electric Power Research Institute Co., Ltd., Hangzhou 310030, ChinaSchool of Mechanical Engineering, Tongji University, Shanghai 201804, ChinaSchool of Mechanical Engineering, Tongji University, Shanghai 201804, ChinaSchool of Mechanical Engineering, Tongji University, Shanghai 201804, ChinaHuadian Electric Power Research Institute Co., Ltd., Hangzhou 310030, ChinaDepartment of Mechanical Engineering, Texas A&M University, College Station, TX 77840, USAThis paper proposes a meta-modeling workflow to forecast the cooling and heating loads of buildings at individual and district levels in the early design stage. Seven input variables, with large impacts on building loads, are selected for designing meta-models to establish the MySQL database. The load profiles of office, commercial, and hotel models are simulated with EnergyPlus in batches. A sequence-to-sequence (Seq2Seq) model based on the deep-learning method of a one-dimensional convolutional neural network (1D-CNN) is introduced to achieve rapid forecasting of all-year hourly building loads. The method performs well with the load effective hour rate (LEHR) of around 90% and MAPE less than 10%. Finally, this meta-modeling workflow is applied to a district as a case study in Shanghai, China. The forecasting results well match the actual loads with R<sup>2</sup> of 0.9978 and 0.9975, respectively, for the heating and cooling load. The LEHR value of all-year hourly forecasting loads is 98.4%, as well as an MAPE of 4.4%. This meta-modeling workflow expands the applicability of building-physics-based methods and improves the time resolution of conventional data-driven methods. It shows small forecasting errors and fast computing speed while meeting the required precision and convenience of engineering in the building early design stage.https://www.mdpi.com/2075-5309/12/2/177building load forecastingmeta-modelingdeep learningCNNSeq2Seqdistrict |
| spellingShingle | Yuhao Zhou Yumin Liang Yiqun Pan Xiaolei Yuan Yurong Xie Wenqi Jia A Deep-Learning-Based Meta-Modeling Workflow for Thermal Load Forecasting in Buildings: Method and a Case Study Buildings building load forecasting meta-modeling deep learning CNN Seq2Seq district |
| title | A Deep-Learning-Based Meta-Modeling Workflow for Thermal Load Forecasting in Buildings: Method and a Case Study |
| title_full | A Deep-Learning-Based Meta-Modeling Workflow for Thermal Load Forecasting in Buildings: Method and a Case Study |
| title_fullStr | A Deep-Learning-Based Meta-Modeling Workflow for Thermal Load Forecasting in Buildings: Method and a Case Study |
| title_full_unstemmed | A Deep-Learning-Based Meta-Modeling Workflow for Thermal Load Forecasting in Buildings: Method and a Case Study |
| title_short | A Deep-Learning-Based Meta-Modeling Workflow for Thermal Load Forecasting in Buildings: Method and a Case Study |
| title_sort | deep learning based meta modeling workflow for thermal load forecasting in buildings method and a case study |
| topic | building load forecasting meta-modeling deep learning CNN Seq2Seq district |
| url | https://www.mdpi.com/2075-5309/12/2/177 |
| work_keys_str_mv | AT yuhaozhou adeeplearningbasedmetamodelingworkflowforthermalloadforecastinginbuildingsmethodandacasestudy AT yuminliang adeeplearningbasedmetamodelingworkflowforthermalloadforecastinginbuildingsmethodandacasestudy AT yiqunpan adeeplearningbasedmetamodelingworkflowforthermalloadforecastinginbuildingsmethodandacasestudy AT xiaoleiyuan adeeplearningbasedmetamodelingworkflowforthermalloadforecastinginbuildingsmethodandacasestudy AT yurongxie adeeplearningbasedmetamodelingworkflowforthermalloadforecastinginbuildingsmethodandacasestudy AT wenqijia adeeplearningbasedmetamodelingworkflowforthermalloadforecastinginbuildingsmethodandacasestudy AT yuhaozhou deeplearningbasedmetamodelingworkflowforthermalloadforecastinginbuildingsmethodandacasestudy AT yuminliang deeplearningbasedmetamodelingworkflowforthermalloadforecastinginbuildingsmethodandacasestudy AT yiqunpan deeplearningbasedmetamodelingworkflowforthermalloadforecastinginbuildingsmethodandacasestudy AT xiaoleiyuan deeplearningbasedmetamodelingworkflowforthermalloadforecastinginbuildingsmethodandacasestudy AT yurongxie deeplearningbasedmetamodelingworkflowforthermalloadforecastinginbuildingsmethodandacasestudy AT wenqijia deeplearningbasedmetamodelingworkflowforthermalloadforecastinginbuildingsmethodandacasestudy |