Simplified neural network model design with sensitivity analysis and electricity consumption prediction in a commercial building. Kim, M. K., Cha, J., Lee, E., Pham, V. H., Lee, S., & Theera-Umpon, N. Energies, March, 2019. Publisher: MDPI AG![Simplified neural network model design with sensitivity analysis and electricity consumption prediction in a commercial building [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex 4 downloads With growing urbanization, it has become necessary to manage this growth smartly. Specifically, increased electrical energy consumption has become a rapid urbanization trend in China. A building model based on a neural network was proposed to overcome the difficulties of analytical modelling. However, increased amounts of data, repetitive computation, and training time become a limitation of this approach. A simplified model can be used instead of the full order model if the performance is acceptable. In order to select effective data, Mean Impact Value (MIV) has been applied to select meaningful data. To verify this neural network method, we used real electricity consumption data of a shopping mall in China as a case study. In this paper, a Bayesian Regularization Neural Network (BRNN) is utilized to avoid overfitting due to the small amount of data. With the simplified data set, the building model showed reasonable performance. The mean of Root Mean Square Error achieved is around 10% with respect to the actual consumption and the standard deviation is low, which reflects the model’s reliability. We also compare the results with our previous approach using the Levenberg–Marquardt back propagation (LM-BP) method. The main difference is the output reliability of the two methods. LM-BP shows higher error than BRNN due to overfitting. BRNN shows reliable prediction results when the simplified neural network model is applied.
@article{Kim2019,
title = {Simplified neural network model design with sensitivity analysis and electricity consumption prediction in a commercial building},
volume = {12},
issn = {19961073},
url = {http://www.scopus.com/inward/record.url?eid=2-s2.0-85065539954%7B%5C&%7DpartnerID=MN8TOARS},
doi = {10.3390/en12071201},
abstract = {With growing urbanization, it has become necessary to manage this growth smartly. Specifically, increased electrical energy consumption has become a rapid urbanization trend in China. A building model based on a neural network was proposed to overcome the difficulties of analytical modelling. However, increased amounts of data, repetitive computation, and training time become a limitation of this approach. A simplified model can be used instead of the full order model if the performance is acceptable. In order to select effective data, Mean Impact Value (MIV) has been applied to select meaningful data. To verify this neural network method, we used real electricity consumption data of a shopping mall in China as a case study. In this paper, a Bayesian Regularization Neural Network (BRNN) is utilized to avoid overfitting due to the small amount of data. With the simplified data set, the building model showed reasonable performance. The mean of Root Mean Square Error achieved is around 10\% with respect to the actual consumption and the standard deviation is low, which reflects the model’s reliability. We also compare the results with our previous approach using the Levenberg–Marquardt back propagation (LM-BP) method. The main difference is the output reliability of the two methods. LM-BP shows higher error than BRNN due to overfitting. BRNN shows reliable prediction results when the simplified neural network model is applied.},
number = {7},
journal = {Energies},
author = {Kim, Moon Keun and Cha, Jaehoon and Lee, Eunmi and Pham, Van Huy and Lee, Sanghyuk and Theera-Umpon, Nipon},
month = mar,
year = {2019},
note = {Publisher: MDPI AG},
keywords = {Bayesian regularization neural network, Building modelling, Energy management, Mean impact value, Simplified model},
}
Downloads: 4
{"_id":"kBuTYq8zNvYBZ7p23","bibbaseid":"kim-cha-lee-pham-lee-theeraumpon-simplifiedneuralnetworkmodeldesignwithsensitivityanalysisandelectricityconsumptionpredictioninacommercialbuilding-2019","downloads":4,"creationDate":"2019-03-28T12:07:04.728Z","title":"Simplified neural network model design with sensitivity analysis and electricity consumption prediction in a commercial building","author_short":["Kim, M. K.","Cha, J.","Lee, E.","Pham, V. H.","Lee, S.","Theera-Umpon, N."],"year":2019,"bibtype":"article","biburl":"https://api.zotero.org/groups/2168152/items?key=VCdsaROd5deDY3prqqG8kI0c&format=bibtex&limit=100","bibdata":{"bibtype":"article","type":"article","title":"Simplified neural network model design with sensitivity analysis and electricity consumption prediction in a commercial building","volume":"12","issn":"19961073","url":"http://www.scopus.com/inward/record.url?eid=2-s2.0-85065539954%7B%5C&%7DpartnerID=MN8TOARS","doi":"10.3390/en12071201","abstract":"With growing urbanization, it has become necessary to manage this growth smartly. Specifically, increased electrical energy consumption has become a rapid urbanization trend in China. A building model based on a neural network was proposed to overcome the difficulties of analytical modelling. However, increased amounts of data, repetitive computation, and training time become a limitation of this approach. A simplified model can be used instead of the full order model if the performance is acceptable. In order to select effective data, Mean Impact Value (MIV) has been applied to select meaningful data. To verify this neural network method, we used real electricity consumption data of a shopping mall in China as a case study. In this paper, a Bayesian Regularization Neural Network (BRNN) is utilized to avoid overfitting due to the small amount of data. With the simplified data set, the building model showed reasonable performance. The mean of Root Mean Square Error achieved is around 10% with respect to the actual consumption and the standard deviation is low, which reflects the model’s reliability. We also compare the results with our previous approach using the Levenberg–Marquardt back propagation (LM-BP) method. The main difference is the output reliability of the two methods. LM-BP shows higher error than BRNN due to overfitting. BRNN shows reliable prediction results when the simplified neural network model is applied.","number":"7","journal":"Energies","author":[{"propositions":[],"lastnames":["Kim"],"firstnames":["Moon","Keun"],"suffixes":[]},{"propositions":[],"lastnames":["Cha"],"firstnames":["Jaehoon"],"suffixes":[]},{"propositions":[],"lastnames":["Lee"],"firstnames":["Eunmi"],"suffixes":[]},{"propositions":[],"lastnames":["Pham"],"firstnames":["Van","Huy"],"suffixes":[]},{"propositions":[],"lastnames":["Lee"],"firstnames":["Sanghyuk"],"suffixes":[]},{"propositions":[],"lastnames":["Theera-Umpon"],"firstnames":["Nipon"],"suffixes":[]}],"month":"March","year":"2019","note":"Publisher: MDPI AG","keywords":"Bayesian regularization neural network, Building modelling, Energy management, Mean impact value, Simplified model","bibtex":"@article{Kim2019,\n\ttitle = {Simplified neural network model design with sensitivity analysis and electricity consumption prediction in a commercial building},\n\tvolume = {12},\n\tissn = {19961073},\n\turl = {http://www.scopus.com/inward/record.url?eid=2-s2.0-85065539954%7B%5C&%7DpartnerID=MN8TOARS},\n\tdoi = {10.3390/en12071201},\n\tabstract = {With growing urbanization, it has become necessary to manage this growth smartly. Specifically, increased electrical energy consumption has become a rapid urbanization trend in China. A building model based on a neural network was proposed to overcome the difficulties of analytical modelling. However, increased amounts of data, repetitive computation, and training time become a limitation of this approach. A simplified model can be used instead of the full order model if the performance is acceptable. In order to select effective data, Mean Impact Value (MIV) has been applied to select meaningful data. To verify this neural network method, we used real electricity consumption data of a shopping mall in China as a case study. In this paper, a Bayesian Regularization Neural Network (BRNN) is utilized to avoid overfitting due to the small amount of data. With the simplified data set, the building model showed reasonable performance. The mean of Root Mean Square Error achieved is around 10\\% with respect to the actual consumption and the standard deviation is low, which reflects the model’s reliability. We also compare the results with our previous approach using the Levenberg–Marquardt back propagation (LM-BP) method. The main difference is the output reliability of the two methods. LM-BP shows higher error than BRNN due to overfitting. BRNN shows reliable prediction results when the simplified neural network model is applied.},\n\tnumber = {7},\n\tjournal = {Energies},\n\tauthor = {Kim, Moon Keun and Cha, Jaehoon and Lee, Eunmi and Pham, Van Huy and Lee, Sanghyuk and Theera-Umpon, Nipon},\n\tmonth = mar,\n\tyear = {2019},\n\tnote = {Publisher: MDPI AG},\n\tkeywords = {Bayesian regularization neural network, Building modelling, Energy management, Mean impact value, Simplified model},\n}\n\n","author_short":["Kim, M. K.","Cha, J.","Lee, E.","Pham, V. H.","Lee, S.","Theera-Umpon, N."],"key":"Kim2019","id":"Kim2019","bibbaseid":"kim-cha-lee-pham-lee-theeraumpon-simplifiedneuralnetworkmodeldesignwithsensitivityanalysisandelectricityconsumptionpredictioninacommercialbuilding-2019","role":"author","urls":{"Paper":"http://www.scopus.com/inward/record.url?eid=2-s2.0-85065539954%7B%5C&%7DpartnerID=MN8TOARS"},"keyword":["Bayesian regularization neural network","Building modelling","Energy management","Mean impact value","Simplified model"],"metadata":{"authorlinks":{}},"downloads":4},"search_terms":["simplified","neural","network","model","design","sensitivity","analysis","electricity","consumption","prediction","commercial","building","kim","cha","lee","pham","lee","theera-umpon"],"keywords":["bayesian regularization neural network","building modelling","energy management","mean impact value","simplified model"],"authorIDs":[],"dataSources":["syJjwTDDM32TsM2iF"]}