Neural Network for Forecasting Energy Consumption Load of a Railway Marshalling Yard
https://doi.org/10.30932/1992-3252-2019-17-3-6-15
Abstract
A multilayer neural network has been designed to forecast average daily energy consumption of a railway marshalling yard. The suggested model comprises a multilayer perceptron using 22 inputs, the n-th number of hidden layers and one output. The number of hidden layers in the neural network and neurons in them was chosen experimentally. A comparative selection of activation functions and training methods has allowed for all other parameters to achieve a minimum average relative error.
Two types of loads corresponding to holidays (non-working) and working days were identified. An additional input node with binary coding and two nodes for coding the season were introduced due to a certain repeatability characterizing samples of prediction of loads of energy consumption of the marshalling yard depending on type of a day and on a season. As accounting of the dependence of the forecast on load values in previous days and years (dynamic dependencies) is most important factor, this neural network takes into account the average daily energy consumption during four days of the current period, preceding
the forecasted date, and the average daily power consumption during four days prior to this date during last three years.
As a result, considering all factors and experimentally selected parameters of the neural network, the minimum resulting error of MAPE is about 1,4 %, which shows the advantage of the developed neural network in comparison with two other methods of solution of the problem, suggested by other researchers.
About the Authors
V. N. Gridin
Center for Information Technologies in Design of Russian Academy of Sciences
Russian Federation
Russian Federation
D.Sc. (Eng), professor, scientific director
Moscow
V. V. Doenin
Center for Information Technologies in Design of Russian Academy of Sciences
Russian Federation
Russian Federation
D.Sc. (Eng), senior researcher, professor of Russian University of Transport
Moscow
V. S. Panishchev
Center for Information Technologies in Design of Russian Academy of Sciences
Russian Federation
Russian Federation
Ph.D. (Eng), senior researcher
Moscow
I. D. Bysov
Southwest State University
Russian Federation
Russian Federation
Master’s student
Kursk,
References
1. Akhmetyanov, R. R., Delegodina, L. A., Kopylova, N. P. [et al]. The tasks of energy consumption forecasting in the integrated AMR system of Novosibirsk Scientific Center [Zadachi prognozirovaniya energopotrebleniya v integrirovannoi ASKUE Novosibirskogo nauchnogo tsentra]. Energosnabzhenie, 2007, Iss. 1, pp. 42–47.
2. Akhmetyanov, R. R., Delegodina, L. A., Kopylova, N. P. [et al]. The tasks of energy consumption forecasting in the integrated AMR system of Novosibirsk Scientific Center [Zadachi prognozirovaniya energopotrebleniya v integrirovannoi ASKUE Novosibirskogo nauchnogo tsentra]. Energosnabzhenie, 2007, Iss. 1, pp. 42–47.
3. Akhmetyanov, R. R., Delegodina, L. A., Kopylova, N. P. [et al]. The multiplicative model of the seasonal energy consumption of enterprises [Multiplikativnaya model sezonnogo energopotrebleniya predpriyatii]. Avtometriya, 2008, Iss. 3, pp. 106–118.
4. Akhmetyanov, R. R., Delegodina, L. A., Kopylova, N. P. [et al]. The multiplicative model of the seasonal energy consumption of enterprises [Multiplikativnaya model sezonnogo energopotrebleniya predpriyatii]. Avtometriya, 2008, Iss. 3, pp. 106–118.
5. Akhmetyanov, R. R., Delegodina, L. A., Kopylova, N. P. [et al]. Using non-standard seasonal models of autoregression and integrated moving average in resource saving problems [Ispolzovanie nestandartnykh sezonnykh modelei avtoregresii i prointegrirovannogo skolzyashchego srednego v zadachakh resursosberezheniya]. Avtometriya, 2008, Iss. 4, pp. 28–41.
6. Akhmetyanov, R. R., Delegodina, L. A., Kopylova, N. P. [et al]. Using non-standard seasonal models of autoregression and integrated moving average in resource saving problems [Ispolzovanie nestandartnykh sezonnykh modelei avtoregresii i prointegrirovannogo skolzyashchego srednego v zadachakh resursosberezheniya]. Avtometriya, 2008, Iss. 4, pp. 28–41.
7. Bysov, I. D., Panishchev, V. S. Using a neural network to predict the loads of energy consumption of an enterprise [Ispolzovanie neironnoi seti dlya prognozirovaniya nagruzok energeticheskogo potrebleniya predpriyatiya]. Tendetsii razvitiya nauki i obrazovaniya, 2017, Iss. 33–1, pp. 17–18.
8. Bysov, I. D., Panishchev, V. S. Using a neural network to predict the loads of energy consumption of an enterprise [Ispolzovanie neironnoi seti dlya prognozirovaniya nagruzok energeticheskogo potrebleniya predpriyatiya]. Tendetsii razvitiya nauki i obrazovaniya, 2017, Iss. 33–1, pp. 17–18.
9. Kingma, D. P., Ba, J. Adam: A Method for Stochastic Optimization. Cornell University Library, 2014. [Electronic resource]: https://arxiv.org/pdf/1412.6980.pdf. Last accessed 27.02.2019.
10. Kingma, D. P., Ba, J. Adam: A Method for Stochastic Optimization. Cornell University Library, 2014. [Electronic resource]: https://arxiv.org/pdf/1412.6980.pdf. Last accessed 27.02.2019.
11. Gridin, V. N., Solodovnikov, V. I., Karnakov, V. V. Selection of initial values and optimization of neural network parameters [Vybor nachalnykh znachenii i optimizatsii parametrov neironnoi seti]. Novie informatsionnie tekhnologii v avtomatizirovannykh sistemakh, 2016, Iss. 19, pp. 270–273.
12. Gridin, V. N., Solodovnikov, V. I., Karnakov, V. V. Selection of initial values and optimization of neural network parameters [Vybor nachalnykh znachenii i optimizatsii parametrov neironnoi seti]. Novie informatsionnie tekhnologii v avtomatizirovannykh sistemakh, 2016, Iss. 19, pp. 270–273.
13. Müller, A., Guido, S. Introduction to machine learning using Python. Trans. from English. Moscow, Williams, 2017, 480 p.
14. Müller, A., Guido, S. Introduction to machine learning using Python. Trans. from English. Moscow, Williams, 2017, 480 p.
15. Raschka, S. Python Machine Learning. Trans. from English. Moscow, DMK Press, 2017, 418 p.
16. Raschka, S. Python Machine Learning. Trans. from English. Moscow, DMK Press, 2017, 418 p.
17. McKinney, W. Python for Data Analysis: Data Wrangling with Pandas, NumPy, and IPython [Russian title: Python and data analysis]. Trans. from English. Moscow, DMK Press, 2015, 482 p.
18. McKinney, W. Python for Data Analysis: Data Wrangling with Pandas, NumPy, and IPython [Russian title: Python and data analysis]. Trans. from English. Moscow, DMK Press, 2015, 482 p.
19. Solomakho, K. L. Application of the principal component method for predicting the volume of energy consumption of a power sales company. Ph.D. (Eng) thesis [Primenenie metoda glavnykh component dlya prognozirovaniya ob’emov energopotrebleniya energosbytovogo predpriyatiya. Dis… kand. tekh. nauk]. Southern Urals State University, Chelyabinsk, 2015 [Electronic resource]: https://www.susu.ru/sites/default/files/dissertation/03_text_diss_5.pdf. Last accessed 27.02.2019.
20. Solomakho, K. L. Application of the principal component method for predicting the volume of energy consumption of a power sales company. Ph.D. (Eng) thesis [Primenenie metoda glavnykh component dlya prognozirovaniya ob’emov energopotrebleniya energosbytovogo predpriyatiya. Dis… kand. tekh. nauk]. Southern Urals State University, Chelyabinsk, 2015 [Electronic resource]: https://www.susu.ru/sites/default/files/dissertation/03_text_diss_5.pdf. Last accessed 27.02.2019.
21. Osovsky, S. Neural networks for information processing [Neironnie seti dlya obrabotki informatsii]. Moscow, Finansy i statistika, 2002, 344 p.
22. Osovsky, S. Neural networks for information processing [Neironnie seti dlya obrabotki informatsii]. Moscow, Finansy i statistika, 2002, 344 p.
23. Gridin, V. N., Doenin, V. V., Solodovnikov, V. I., Panishchev, V. S., Trufanov, M. I. On the issue of building an intellectual subsystem for analyzing and predicting the operation of the marshalling yard [K voprosu postroeniya intellektualnoi podsistemy analiza i prognozarovaniya raboty sortirovochnogo uzla]. Informatsionnie tekhnologii i vychislitelnie sistemy, 2017, Iss. 4, pp. 95–103.
24. Gridin, V. N., Doenin, V. V., Solodovnikov, V. I., Panishchev, V. S., Trufanov, M. I. On the issue of building an intellectual subsystem for analyzing and predicting the operation of the marshalling yard [K voprosu postroeniya intellektualnoi podsistemy analiza i prognozarovaniya raboty sortirovochnogo uzla]. Informatsionnie tekhnologii i vychislitelnie sistemy, 2017, Iss. 4, pp. 95–103.
25. Gridin, V. N., Doenin, V. V., Panishchev, V. S. On construction of an intelligent subsystem for analyzing the parameters of a marshalling hub. World of Transport and Transportation, Vol. 15, 2017, Iss. 4, pp. 6–19.
26. Gridin, V. N., Doenin, V. V., Panishchev, V. S. On construction of an intelligent subsystem for analyzing the parameters of a marshalling hub. World of Transport and Transportation, Vol. 15, 2017, Iss. 4, pp. 6–19.
Review
For citations:
Gridin V.N., Doenin V.V., Panishchev V.S., Bysov I.D. Neural Network for Forecasting Energy Consumption Load of a Railway Marshalling Yard. World of Transport and Transportation. 2019;17(3):6-15. https://doi.org/10.30932/1992-3252-2019-17-3-6-15
Views:
664
Email this article 