Prediction model of public houses’ heating systems:a comparison of support vector machine methodand random forest method

• Autorzy: Perekrest Andrii , Chenchevoi Vladimir , Chencheva Olga , Kovalenko Alexandr , Kushch-Zhyrko Mykhailo , Kalizhanova Aliya , Amirgaliyev Yedilkhan

Identyfikatory

DOI

10.35784/iapgos.3032

Warianty tytułu

PL

Model prognozowania systemów grzewczych budynków użyteczności publicznej: porównanie metody support vector machine i random forest

• Języki publikacji: EN

Abstrakty

EN

Data analysis and predicting play an important role in managing heat-supplying systems. Applying the models of predicting the systems’ parameters is possible for qualitative management, accepting appropriate decisions relating control that will be aimed at increasing energy efficiency and decreasing the amount of the consumed power source, diagnosing and defining non-typical processes in the functioning of the systems. The article deals with comparing two methods of ma-chine learning: random forest (RF) and support vector machine (SVM) for predicting the temperature of the heat-carrying agent in the heating system based on the data of electronic weather-dependent controller. The authors use the following parameters to compare the models: accuracy, source cost and the opportunity to interpret the results and non-obvious interrelations. The time spent for defining the optimal hyperparameters and conducting the SVM model training is deter-mined to exceed significantly the data of the RF parameter despite the close meanings of the root mean square error (RMSE). The change from 15-min data to once-a-minute ones is done to improve the RF model accuracy. RMSE of the RF model on the test data equals 0.41°С. The article studies the importance of the contribution of variables to the prediction accuracy.

PL

Analiza danych i prognozowanie odgrywają ważną rolę w zarządzaniu systemami zaopatrzenia w ciepło. Wykorzystanie modeli do przewidywania parametrów systemu jest możliwe do zarządzania jakością, podejmowania odpowiednich decyzji sterujących, które będą miały na celu poprawę efektywności energetycznej i zmniejszenie ilości zużywanego źródła energii elektrycznej, diagnozowania i wykrywania nietypowych procesów w funkcjonowaniu systemu. W artykule porównano dwie metody uczenia maszynowego: Random Forest (RF) i Support Vector Machine (SVM) do przewidywania temperatury czynnika grzewczego w systemie grzewczym na podstawie danych elektronicznego regulatora pogodowego. Do porównania modeli autorzy wykorzystują następujące parametry: dokładność, koszt początkowy oraz możliwość interpretacji wyników i nieoczywistych zależności. Ustalono, że czas poświęcony na wyznaczenie optymalnych hiperparametrów i wytrenowanie modelu SVM znacznie przekracza dane parametru RF, pomimo zbliżonych wartości błędu średniokwadratowego (RMSE). Zmiana z danych 15-minutowych na dane raz na minutę została dokonana w celu poprawy dokładności modelu RF. RMSE modelu RF z danych testowych wynosi 0,41°C. W pracy zbadano znaczenie wkładu zmiennych w dokładność prognozy.

Słowa kluczowe

EN

building heat supply; random forest; support vector machine

PL

zaopatrzenie w ciepło budynku; random forest; metoda wektorów wspierających

• Wydawca: Wydawnictwo Politechniki Lubelskiej
• Czasopismo: Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska
• Rocznik: 2022
• Tom: T. 12, nr 3
• Strony: 34--39
• Opis fizyczny: Bibliogr., 30 poz., tab., wykr.

Twórcy

autor

Perekrest Andrii

• Kremenchuk Mykhailo Ostrohradskyi National University, Department of Computer Engineering and Electronics, Kremenchuk,Ukraine

• https://orcid.org/0000-0002-7728-9020

autor

Chenchevoi Vladimir

• Kremenchuk Mykhailo Ostrohradskyi National University, Department of Systems of Automatic Control and Electric Drive, Kremenchuk, Ukraine

• https://orcid.org/0000-0002-6478-3767

autor

Chencheva Olga

• Kremenchuk Mykhailo Ostrohradskyi National University, Department of Computer Engineering and Electronics, Kremenchuk,Ukraine

• https://orcid.org/0000-0002-5691-7884

autor

Kovalenko Alexandr

• Cherkasy State Technological University, Cherkasy, Ukraine

• https://orcid.org/0000-0002-5073-3507

autor

Kushch-Zhyrko Mykhailo

• Kremenchuk Mykhailo Ostrohradskyi National University, Department of Computer Engineering and Electronics, Kremenchuk,Ukraine

• https://orcid.org/0000-0001-9622-9114

autor

Kalizhanova Aliya

• University of Power Engineering and Telecommunications, Almaty, Kazakhstan

• https://orcid.org/0000-0002-5979-9756

autor

Amirgaliyev Yedilkhan

• Institute of Information and Computational Technologies MES CS RK, Almaty, Kazakhstan

• https://orcid.org/0000-0002-6528-0619

Bibliografia

• [1] Ahmad M. V. et al.: Tree-based ensemble methods for predicting PV power generation and their comparison with support vector regression. Energy 164, 2018, 465–474 [http://doi.org/10.1016/j.energy.2018.08.207].
• [2] Ahmad, M. V. et al.: Predictive modelling for solar thermal energy systems: A comparison of support vector regression, random forest, extra trees and regression trees. Journal of Cleaner Production 203, 2018, 810–821 [http://doi.org/10.1016/j.jclepro.2018.08.207].
• [3] Ahmad M. W. et al.: Trees vs Neurons: Comparison between random forest and ANN for high-resolution prediction of building energy consumption. Energy and Buildings 147, 2017, 77–89 [http://doi.org/10.1016/j.enbuild.2017.04.038].
• [4] Azarov A. D. et al.: Class of numerical systems for pipe-line bit sequential development of multiple optoelectronic data streams. Proc. SPIE 4425, 2001, 406-409.
• [5] Azarov A.D. et al.: Static and dynamic characteristics of the self-calibrating multibit ADC analog components. Proc. SPIE 8698, 2012, 86980N.
• [6] Breiman L.: Out-of-bag estimation. Tech. rep. University of California, 1996 [https://www.stat.berkeley.edu/~breiman/OOBestimation.pdf].
• [7] Breiman L.: Random Forests. Machine Learning 45(1), 2001, 5–32 [http://doi.org/10.1023/A:1010933404324].
• [8] Dong B. et al.: Applying support vector machines to predict building energy consumption in tropical region. Energy and Buildings 37(5), 2005, 545–553 [http://doi.org/10.1016/j.enbuild.2004.09.009].
• [9] Esen H. et al.: Modelling of a new solar air heater through least-squares support vector machines. Expert Systems with Applications 36(7), 2009, 10673–1068 [http://doi.org/10.1016/j.eswa.2009.02.045].
• [10] Geng Y. et al.: Building energy performance diagnosis using energy bills and weather data. Energy and Buildings 172, 2018, 181–191 [http://doi.org/10.1016/j.enbuild.2018.04.047].
• [11] Kaczmarek C. et al.: Measurement of pressure sensitivity of modal birefringence of birefringent optical fibers using a Sagnac interferometer. Optica Applicata 45(1), 2015, 5–14.
• [12] Kukharchuk V. V. et al.: Method of magneto-elastic control of mechanic rigidity in assemblies of hydropower units. Proc. SPIE 10445, 2017, 104456A.
• [13] Kukharchuk V. V. et al.: Noncontact method of temperature measurement based on the phenomenon of the luminophor temperature decreasing. Proc. SPIE 10031, 2016, 100312F.
• [14] Kukharchuk V. V. et al.: Discrete wavelet transformation in spectral analysis of vibration processes at hydropower units. Przegląd Elektrotechniczny 93(5), 2017, 65–68.
• [15] Kvyetnyy R. et al.: Blur recognition using second fundamental form of image surface. Proc. SPIE 9816, 2015, 98161A.
• [16] Kvyetnyy R. et al.: Method of image texture seg-mentation using Laws' energy measures. Proc. SPIE 10445, 2017, 1044561.
• [17] Kvyetnyy R. et al.: Modification of fractal coding algorithm by a combination of modern technologies and parallel computations. Proc. SPIE 9816, 2015, 98161R.
• [18] Osadchuk A. et al.: Pressure transducer of the on the basis of reactive properties of transistor structure with negative resistance. Proc. SPIE 9816, 2015, 98161C.
• [19] Osadchuk O. et al.: The Generator of Superhigh Frequencies on the Basis Silicon Germanium Heterojunction Bipolar Transistors. 13th International Conference on Modern Problems of Radio Engineering, Telecommunications and Computer Science (TCSET), 2016, 336 – 338.
• [20] Paluszyska A.: Structure mining and knowledge extraction from random forest with applications to The Cancer Genome Atlas project. 2017. [https://rawgit.com/geneticsMiNIng/BlackBoxOpener/master/randomForestExpl ainer_Master_thesis.pdf].
• [21] Parfenenko Yu. V. et al.: Prediction the heat consumption of social and public sector buildings using neural networks. Radio Electronics, Computer Science, Control 2, 2015, 41–46.
• [22] Perekrest A. et al.: Key Performance Indicators Assessment Methodology Principles Adaptation for Heating Systems of Administrative and Residential Buildings. IEEE Problems of Automated Electrodrive. Theory and Practice (PAEP), 2020, 1–4.
• [23] Perekrest A. et al.: Complex information and technical solutions for energy management of municipal energetics. Proc. SPIE 10445, 2017, 1044567 [http://doi.org/10.1117/12.2280962].
• [24] Ruiz L. G. B. et al.: Energy consumption forecasting based on Elman neural networks with evolutive optimization. Expert Systems with Applications 92, 2018, 380–389 [http://doi.org/10.1016/j.eswa.2017.09.059].
• [25] Smolarz A. et al.: Fuzzy controller for a lean premixed burner. Przegląd Elektrotechniczny 86(7), 2010, 287–289.
• [26] Vapnik V., Chapelle O.: Bounds on error expectation for support vector machines. Neural Computation 12 (9), 2000, 2013–2036 [http://doi.org/10.1162/089976600300015042].
• [27] Vedmitskyi Y. G. et al.: New non-system physical quantities for vibration monitoring of transient processes at hydropower facilities, integral vibratory accelerations. Przegląd Elektrotechniczny 93(3), 2017, 69–72.
• [28] Wei Y. et al.: A review of data-driven approaches for prediction and classification of building energy consumption. Renewable and Sustainable Energy Reviews 82, 2018, 1027–1047 [http://doi.org/10.1016/j.rser.2017.09.108].
• [29] Wójcik W. et al.: Concept of application of signals from fiber-optic system for flame monitoring to control separate pulverized coal burner. Proc. SPIE 5484, 2004, 427–431.
• [30] Wójcik W. et al.: Vision based monitoring of coal flames source. Przegląd Elektrotechniczny 84(3), 2008, 241–243.

Uwagi

Opracowanie rekordu ze środków MEiN, umowa nr SONP/SP/546092/2022 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2022-2023).