Enerji Tüketim Tahmini İçin Farklı Regresyon Algoritmalarının Analizi

Öz

Ekonomi, sanat, eğlence ve sanayi gibi farklı alanlarda enerji tüketen merkezlerde enerji tüketim verilerinin büyümesine bağlı olarak, üretilmesi gereken enerji miktarı önümüzdeki yıllarda daha da artacaktır. Sığınak ve konutlarda enerji tüketimini en aza indirmek için elektronik cihazlardan oluşan yapay zeka destekli konut sistemlerinde artış olduğu gözlemlendi. Küresel ısınma, sera gazı emisyonları, karbondioksit, kimyasal çözücüler, radyasyon gibi çevresel faktörlerin artması göz önüne alındığında, enerjinin verimli kullanımına yönelik çalışmalar artırılmalıdır. Bu amaçla Amerika Birleşik Devletleri bölgesel iletişim kuruluşu PJM Interconnection LLC (PJM)'nin internet sitesinden elde edilen Asya bölgesinin Mega Watt cinsinden saatlik veri tüketimini Dominion Virginia Power (DOM) verisi olarak gösteren veri seti kullanılmıştır. Bu veri seti üzerinde son zamanlarda popüler olan XGBoost, LSTM algoritmaları, klasik Lineer regresyon ve Ransac algoritmaları DOM veri seti kullanılarak karşılaştırılmıştır. Karşılaştırma sırasında kullanılan veri setinin ölçeklenmiş ve ölçeklenmemiş versiyonu arasındaki eğitim ve test sonuçları arasındaki farklar incelenmiştir.

Anahtar Kelimeler

• LSTM
• XGBoost
• enerji tüketimi
• Ransac
• Doğrusal regresyon

LSTM, XGBoost, energy consumption, Ransac, Linear regression.

Abstract

Depending on the growth of energy consumption data in centers consuming energy in different fields such as economy, art, entertainment and industry, the amount of energy that needs to be produced will increase even more in the coming years. It has been observed that there has been an increase in artificial intelligence supported housing systems consisting of electronic devices in order to minimize energy consumption in shelters and residences. Considering the increase in environmental factors such as global warming, greenhouse gas emissions, carbon dioxide, chemical solvents, and radiation, studies should be increased for the efficient use of energy. For this purpose, the data set showing the hourly data consumption in Mega Watts of the Dominion Virginia Power (DOM) data of the Asian region obtained from the website of the United States regional communication organization PJM Interconnection LLC (PJM) was used. On this dataset, the recently popular XGBoost, LSTM algorithms, classical Linear regression and Ransac algorithms are compared using the DOM dataset. The differences between the training and test results between the scaled and unscaled version of the data set used during the comparison were examined.

Keywords

• LSTM
• XGBoost
• energy consumption
• Ransac
• Linear regression

Kaynakça

1. Antanasijević, Davor, Viktor Pocajt, Mirjana Ristić, and Aleksandra Perić-Grujić. 2015. “Modeling of Energy Consumption and Related GHG (Greenhouse Gas) Intensity and Emissions in Europe Using General Regression Neural Networks.” Energy 84 (May): 816–24. https://doi.org/10.1016/j.energy.2015.03.060.
2. Arghira, Nicoleta, Lamis Hawarah, Stéphane Ploix, and Mireille Jacomino. 2012. “Prediction of Appliances Energy Use in Smart Homes.” Energy 48 (1): 128–34. https://doi.org/10.1016/j.energy.2012.04.010.
3. Bagnasco, A, F Fresi, M Saviozzi, F Silvestro, and A Vinci. 2015. “Electrical Consumption Forecasting in Hospital Facilities: An Application Case.” Energy and Buildings 103 (September): 261–70. https://doi.org/10.1016/j.enbuild.2015.05.056.
4. Bahar, Nur H.A., Michaela Lo, Made Sanjaya, Josh Van Vianen, Peter Alexander, Amy Ickowitz, and Terry Sunderland. 2020. “Meeting the Food Security Challenge for Nine Billion People in 2050: What Impact on Forests?” Global Environmental Change 62 (May): 102056. https://doi.org/10.1016/j.gloenvcha.2020.102056.
5. Bhati, Abhishek, Michael Hansen, and Ching Man Chan. 2017. “Energy Conservation through Smart Homes in a Smart City: A Lesson for Singapore Households.” Energy Policy 104 (May): 230–39. https://doi.org/10.1016/j.enpol.2017.01.032.
6. Carvalho, Monica, Danielle Bandeira de Mello Delgado, Karollyne Marques de Lima, Marianna de Camargo Cancela, Camila Alves dos Siqueira, and Dyego Leandro Bezerra de Souza. 2021. “Effects of the COVID‐19 Pandemic on the Brazilian Electricity Consumption Patterns.” International Journal of Energy Research 45 (2): 3358–64.
7. Change, Climate. 2014. “Synthesis Report. Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change.[Core Writing Team, RK Pachauri and LA Meyer.” IPCC, Geneva, Switzerland.
8. Chen, Tianqi, and Carlos Guestrin. 2016. “Xgboost: A Scalable Tree Boosting System.” In Proceedings of the 22nd Acm Sigkdd International Conference on Knowledge Discovery and Data Mining, 785–94.

9. Deb, Chirag, Fan Zhang, Junjing Yang, Siew Eang Lee, and Kwok Wei Shah. 2017. “A Review on Time Series Forecasting Techniques for Building Energy Consumption.” Renewable and Sustainable Energy Reviews 74: 902–24.
10. Derpanis, Konstantinos G. 2010. “Overview of the RANSAC Algorithm.” Image Rochester NY 4 (1): 2–3.
11. Edwards, Richard E, Joshua New, and Lynne E Parker. 2012. “Predicting Future Hourly Residential Electrical Consumption: A Machine Learning Case Study.” Energy and Buildings 49 (June): 591–603. https://doi.org/10.1016/j.enbuild.2012.03.010.
12. Enn, Rosa. 2015. “Impact of Climate Change and Human Activity on the Eco-Environment. An Analysis of the Xisha Islands.” Island Studies Journal 10 (2): 263–64.
13. Fischler, Martin A, and Robert C Bolles. 1981. “Random Sample Consensus: A Paradigm for Model Fitting with Applications to Image Analysis and Automated Cartography.” Communications of the ACM 24 (6): 381–95.
14. Friedman, Jerome H. 2002. “Stochastic Gradient Boosting.” Computational Statistics & Data Analysis 38 (4): 367–78.
15. García, Sebastián, Antonio Parejo, Enrique Personal, Juan Ignacio Guerrero, Félix Biscarri, and Carlos León. 2021. “A Retrospective Analysis of the Impact of the COVID-19 Restrictions on Energy Consumption at a Disaggregated Level.” Applied Energy 287: 116547.
16. Ghiani, Emilio, Marco Galici, Mario Mureddu, and Fabrizio Pilo. 2020. “Impact on Electricity Consumption and Market Pricing of Energy and Ancillary Services during Pandemic of COVID-19 in Italy.” Energies 13 (13): 3357.
17. Günay, M Erdem. 2016. “Forecasting Annual Gross Electricity Demand by Artificial Neural Networks Using Predicted Values of Socio-Economic Indicators and Climatic Conditions: Case of Turkey.” Energy Policy 90: 92–101.
18. Guo, Siyue, Da Yan, Shan Hu, and Yang Zhang. 2021. “Modelling Building Energy Consumption in China under Different Future Scenarios.” Energy 214 (January): 119063. https://doi.org/10.1016/j.energy.2020.119063.
19. Hu, Liyang, Chao Wang, Zhirui Ye, and Sheng Wang. 2021. “Estimating Gaseous Pollutants from Bus Emissions: A Hybrid Model Based on GRU and XGBoost.” Science of The Total Environment 783: 146870.
20. Ilbeigi, Marjan, Mohammad Ghomeishi, and Ali Dehghanbanadaki. 2020. “Prediction and Optimization of Energy Consumption in an Office Building Using Artificial Neural Network and a Genetic Algorithm.” Sustainable Cities and Society 61: 102325.
21. Jung, Hyun Chul, Jin Sung Kim, and Hoon Heo. 2015. “Prediction of Building Energy Consumption Using an Improved Real Coded Genetic Algorithm Based Least Squares Support Vector Machine Approach.” Energy and Buildings 90 (March): 76–84. https://doi.org/10.1016/j.enbuild.2014.12.029.
22. Ketkar, Nikhil, and Eder Santana. 2017. Deep Learning with Python. Vol. 1. Springer.
23. Li, Yan. 2019. “Prediction of Energy Consumption: Variable Regression or Time Series? A Case in China.” Energy Science & Engineering 7 (6): 2510–18. https://doi.org/10.1002/ese3.439.
24. Lü, Xiaoshu, Tao Lu, Charles J Kibert, and Martti Viljanen. 2015. “Modeling and Forecasting Energy Consumption for Heterogeneous Buildings Using a Physical–Statistical Approach.” Applied Energy 144 (April): 261–75. https://doi.org/10.1016/j.apenergy.2014.12.019.
25. Mitchell, Rory, and Eibe Frank. 2017. “Accelerating the XGBoost Algorithm Using GPU Computing.” PeerJ Computer Science 3 (July): e127. https://doi.org/10.7717/peerj-cs.127.
26. Moletsane, Phenyo Phemelo, Tebogo Judith Motlhamme, Reza Malekian, and Dijana Capeska Bogatmoska. 2018. “Linear Regression Analysis of Energy Consumption Data for Smart Homes.” In 2018 41st International Convention on Information and Communication Technology, Electronics and Microelectronics (MIPRO), 0395–99. IEEE. https://doi.org/10.23919/MIPRO.2018.8400075.
27. Moreno, M, Benito Úbeda, Antonio Skarmeta, and Miguel Zamora. 2014. “How Can We Tackle Energy Efficiency in IoT BasedSmart Buildings?” Sensors 14 (6): 9582–9614. https://doi.org/10.3390/s140609582.
28. O’Neill, Zheng, and Charles O’Neill. 2016. “Development of a Probabilistic Graphical Model for Predicting Building Energy Performance.” Applied Energy 164 (February): 650–58. https://doi.org/10.1016/j.apenergy.2015.12.015.
29. Protić, Milan, Fahman Fathurrahman, and Miomir Raos. 2019. “Modelling Energy Consumption of the Republic of Serbia Using Linear Regression and Artificial Neural Network Technique.” Tehnicki Vjesnik - Technical Gazette 26 (1): 135–41. https://doi.org/10.17559/TV-20180219142019.
30. Roldán-Blay, Carlos, Guillermo Escrivá-Escrivá, Carlos Álvarez-Bel, Carlos Roldán-Porta, and Javier Rodríguez-García. 2013. “Upgrade of an Artificial Neural Network Prediction Method for Electrical Consumption Forecasting Using an Hourly Temperature Curve Model.” Energy and Buildings 60 (May): 38–46. https://doi.org/10.1016/j.enbuild.2012.12.009.
31. Wang, Jian Qi, Yu Du, and Jing Wang. 2020. “LSTM Based Long-Term Energy Consumption Prediction with Periodicity.” Energy 197: 117197.
32. Wei, Yixuan, Xingxing Zhang, Yong Shi, Liang Xia, Song Pan, Jinshun Wu, Mengjie Han, and Xiaoyun Zhao. 2018. “A Review of Data-Driven Approaches for Prediction and Classification of Building Energy Consumption.” Renewable and Sustainable Energy Reviews 82: 1027–47.
33. Zhou, Jian, Enming Li, Mingzheng Wang, Xin Chen, Xiuzhi Shi, and Lishuai Jiang. 2019. “Feasibility of Stochastic Gradient Boosting Approach for Evaluating Seismic Liquefaction Potential Based on SPT and CPT Case Histories.” Journal of Performance of Constructed Facilities 33 (3): 4019024.