Segmentation of heterogeneous objects in forecasting problems of energy consumption

About the authors:

Andrey S. Bezrukov, Post-graduate student, Department of Software Development, Institute of Mathematics and Computer Sciences, Tyumen State University
Marina S. Vorobyova, Cand. Sci. (Tech.), Associate Professor, Department of Software, University of Tyumen; m.s.vorobeva@utmn.ru

Abstract:

The peculiarities of the forecasting problem of energy consumption are considered in the paper; the main methods and solution models for forecasting problems are presented. The accuracy of the obtained forecast depending on the qualitative pre-processing of the indicators under investigation is important for the forecasting problems of energy consumption. A new method for data pre-processing of energy consumption objects is introduced for ranking and segmentation tasks in the construction of energy consumption forecasts. A mathematical model of heterogeneous objects and the approach to the modeling of heterogeneous objects using association rules are suggested. The generation of association rules is performed using the apparatus of rough sets. The resulting model allows the adequate representation of a complex object to be obtained for further processing, analysis and forecasting. The results of the study are used while investigating the dependencies in the construction of forecast models of energy consumption, preliminary preparation of indicators for forecasting problem of energy consumption of complex objects.

References:

1.             Ajuev B. Rynki jelektrojenergii i ih realizacija v EJeS Rossii [Electricity Markets and Their Implementation in the UES of Russia]. Ekaterinburg: URO RAN, 2007. 107 p.
(In Russian)

2.             Bezrukov A. S., Vorobyeva M. S. Modelirovanie processa ocenki pokazatelej odnorodnyh i neodnorodnyh obyektov [Simulation of the Evaluation Process of Indicators of Homogeneous and Heterogeneous Objects] // Vestnik Tjumenskogo Gosudarstvennogo Universiteta [Tyumen State University Herald]. 2014. No 7. Pp. 218-225. (In Russian)

3.             Bezrukov A. S. Primenenie iskusstvennyh nejronnyh setej dlja prognozirovanija dinamiki pokazatelej objekta upravlenija na primere sportivnyh sooruzhenij [Application of Artificial Neural Networks to Predict the Dynamics of the Control Object on the Example of Sports Facilities] // Matematicheskoe i informacionnoe modelirovanie: sbornik nauchnyh trudov [Mathematical and Information Modeling: Collection of Scientific Papers]. Tyumen: Izdatel'stvo Tjumenskogo gosudarstvennogo universiteta [Tyumen State University Publishing House]. 2013. Issue 13. Pp. 49-54. (In Russian)

4.             Boldyrev M. V. Reshenie zadach s primeneniem nechetkoj logiki [Solving problems with the use of fuzzy logic] // Jenergosberezhenie, avtomatizacija v promyshlennosti, intellektual'nye zdanija i ASUTP [Energy-saving, Industrial Automation, Process Control Systems and Intelligent Building]. 2010. Vol. 5. Pp. 5-7. (In Russian)

5.             Dem'janenko T. S., Mohov V. G. Prognozirovanie potreblenija jelektricheskoj jenergii na optovom rynke jelektrojenergii i moshhnosti [Predicting Consumption of Electric Power in the Wholesale Electricity Market] // Vestnik Juzhno-Ural'skogo gosudarstvennogo universiteta. Serija: Jekonomika i menedzhment [South Ural State University Herald. Economics and Management]. 2014. Vol. 8. No 2. (In Russian)

6.             Kirpichnikova I. M., Saplin L. A., Solomaho K. L. Prognozirovanie obyomov potreblenija jelektrojenergii [Energy Consumption Forecasting] // Vestnik JuUrGU. Serija: Jenergetika [South Ural State University Herald. Power Engineering]. 2014. No 2. Pp. 16-25. (In Russian)

7.             Sitnikov D. E. Metod poiska logicheskih pravil v jekspertnyh sistemah na osnove associativnyh zavisimostej mezhdu diskretnymi priznakami [Search Method of Logical Rules in Expert Systems Based on Associative Relationships between Discrete Signs] // Zbіrnik naukovih prac' Harkіvs'kogo unіversitetu Povіtrjanih sil [Proceedings of Kharkiv Air Force University]. 2014. No 2. Pp. 147-150. (In Russian)

8.             Talalaev A. A. Analiz jeffektivnosti primenenija iskusstvennyh nejronnyh setej dlja reshenija zadach raspoznavanija, szhatija i prognozirovanija [An analysis of the Effectiveness of Artificial Neural Networks to Solve the Problems of Recognition, Compression and Prediction] // Iskusstvennyj intellekt i prinjatie reshenij [Artificial Intelligence and Decision Making]. 2008. Vol. 2. Pp. 24-33. (In Russian)

9.             Chuchueva I. A. Model' jekstrapoljacii vremennyh rjadov po vyborke maksimal'nogo podobija [Time Series Extrapolation Model Based on Maximum Likeness Set] // Informacionnye tehnologii [Information Technologies]. 2010. No 12. Pp. 43-47.
(In Russian)

10.         Shherbakov M. V. Specifika primenenija intellektual'nyh modelej analiza dannyh dlja povyshenija jenergeticheskoj jeffektivnosti [Specific of Specificity of Application of Intelligent Data Analysis Models for Energy Efficiency] // Izvestija Volgogradskogo gosudarstvennogo tehnicheskogo universiteta. Aktual'nye problemy upravlenija, vychislitel'noj tehniki i informatiki v tehnicheskih sistemah [Proceedings of Volgograd State Technical University. Actual Problems of Management, Computer Science and Informatics in Technical Systems]. 2010. No 11 (71). Issue 9. Pp. 72-76. (In Russian)