Modeling of Building Energy Consumption by Integrating Regression Analysis and Artificial Neural Network with Data Classification
With the constant expansion of the building sector as a major energy consumer in the modern world, the significance of energy-efficient building systems cannot be more emphasized. Most of the buildings are now equipped with an electric dashboard to record consumption data which presents a significan...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-11-01
|
| Series: | Buildings |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-5309/10/11/198 |
| _version_ | 1797549006826504192 |
|---|---|
| author | Iffat Ridwana Nabil Nassif Wonchang Choi |
| author_facet | Iffat Ridwana Nabil Nassif Wonchang Choi |
| author_sort | Iffat Ridwana |
| collection | DOAJ |
| description | With the constant expansion of the building sector as a major energy consumer in the modern world, the significance of energy-efficient building systems cannot be more emphasized. Most of the buildings are now equipped with an electric dashboard to record consumption data which presents a significant scope of research by utilizing those data in energy modeling. This paper investigates conventional regression modeling in building energy estimation and proposes three models with data classifications to improve their performance. The proposed models are regression models and an artificial neural network model with data classification for predicting hourly or sub-hourly energy usage in four different buildings. Energy data is collected from a building energy simulation program and existing buildings to develop the models for detailed analysis. Data classification is recommended according to the system operating schedules of the buildings and models are tested for their performance in capturing the data trends resulting from those schedules. Proposed regression models and an ANN model with the recommended classification show very accurate results in estimating energy demand compared to conventional regression models. Correlation coefficient and root mean squared error values improve noticeably for the proposed models and they can potentially be utilized for energy conservation purposes and energy savings in the buildings. |
| first_indexed | 2024-03-10T15:08:39Z |
| format | Article |
| id | doaj.art-080067d3b8704d3b8ae15d05418e32c3 |
| institution | Directory Open Access Journal |
| issn | 2075-5309 |
| language | English |
| last_indexed | 2024-03-10T15:08:39Z |
| publishDate | 2020-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Buildings |
| spelling | doaj.art-080067d3b8704d3b8ae15d05418e32c32023-11-20T19:31:04ZengMDPI AGBuildings2075-53092020-11-01101119810.3390/buildings10110198Modeling of Building Energy Consumption by Integrating Regression Analysis and Artificial Neural Network with Data ClassificationIffat Ridwana0Nabil Nassif1Wonchang Choi2Department of Civil and Architectural Engineering and Construction Management, University of Cincinnati, Cincinnati, OH 45221, USADepartment of Civil and Architectural Engineering and Construction Management, University of Cincinnati, Cincinnati, OH 45221, USADepartment of Architectural Engineering, Gachon University, Seongnam-si 1342, KoreaWith the constant expansion of the building sector as a major energy consumer in the modern world, the significance of energy-efficient building systems cannot be more emphasized. Most of the buildings are now equipped with an electric dashboard to record consumption data which presents a significant scope of research by utilizing those data in energy modeling. This paper investigates conventional regression modeling in building energy estimation and proposes three models with data classifications to improve their performance. The proposed models are regression models and an artificial neural network model with data classification for predicting hourly or sub-hourly energy usage in four different buildings. Energy data is collected from a building energy simulation program and existing buildings to develop the models for detailed analysis. Data classification is recommended according to the system operating schedules of the buildings and models are tested for their performance in capturing the data trends resulting from those schedules. Proposed regression models and an ANN model with the recommended classification show very accurate results in estimating energy demand compared to conventional regression models. Correlation coefficient and root mean squared error values improve noticeably for the proposed models and they can potentially be utilized for energy conservation purposes and energy savings in the buildings.https://www.mdpi.com/2075-5309/10/11/198building energy consumptioncomputational modelingdata classificationregression modelartificial neural network modeloptimization |
| spellingShingle | Iffat Ridwana Nabil Nassif Wonchang Choi Modeling of Building Energy Consumption by Integrating Regression Analysis and Artificial Neural Network with Data Classification Buildings building energy consumption computational modeling data classification regression model artificial neural network model optimization |
| title | Modeling of Building Energy Consumption by Integrating Regression Analysis and Artificial Neural Network with Data Classification |
| title_full | Modeling of Building Energy Consumption by Integrating Regression Analysis and Artificial Neural Network with Data Classification |
| title_fullStr | Modeling of Building Energy Consumption by Integrating Regression Analysis and Artificial Neural Network with Data Classification |
| title_full_unstemmed | Modeling of Building Energy Consumption by Integrating Regression Analysis and Artificial Neural Network with Data Classification |
| title_short | Modeling of Building Energy Consumption by Integrating Regression Analysis and Artificial Neural Network with Data Classification |
| title_sort | modeling of building energy consumption by integrating regression analysis and artificial neural network with data classification |
| topic | building energy consumption computational modeling data classification regression model artificial neural network model optimization |
| url | https://www.mdpi.com/2075-5309/10/11/198 |
| work_keys_str_mv | AT iffatridwana modelingofbuildingenergyconsumptionbyintegratingregressionanalysisandartificialneuralnetworkwithdataclassification AT nabilnassif modelingofbuildingenergyconsumptionbyintegratingregressionanalysisandartificialneuralnetworkwithdataclassification AT wonchangchoi modelingofbuildingenergyconsumptionbyintegratingregressionanalysisandartificialneuralnetworkwithdataclassification |