Prediction and analysis of power consumption and power loss at industrial facilities

THE PURPOSE. Conduct a study to improve the reliability of forecasting the magnitude of power consumption and power losses at an industrial enterprise.

METHODS. Methods are used to determine and predict the parameters of consumption and losses of electricity at industrial facilities.

RESULTS. To clarify the magnitude of electricity losses, it is proposed to use coefficients that take into account the type of load curves and show the ratio of the values of the sum of the squares of currents (powers) of the variable load curve and the values of the sum of average currents (powers), that is. the ratio of power losses during load operation according to variable and uniform schedules (K_graph), as well as a coefficient that takes into account the topology of the circuit (K_top). The study of radial and main circuits of networks was carried out and the losses of electricity were determined using the proposed coefficients. The values of equivalent resistances of shop circuits of networks of various topologies are calculated. The operational data of the section of the workshop network are given. It was revealed that with a constant technological process, an increase in the equivalent resistance of the network circuit is due to an increase in the resistance of the contacts of switching devices installed on the lines. The value of the estimated supply of electricity was determined using the parameter of the average value of the equivalent resistance. At the same time, the error in calculating the estimated supply in relation to the actual annual supply of electricity amounted to 2,63%. According to the retrospective values of the average equivalent resistance of the circuit, it is possible to determine the predicted value of this parameter using the average value of the coefficient of change in the equivalent resistance. These characteristics of the scheme are recommended to be used in the assessment and forecasting of losses and the estimated supply of electricity, which will increase the reliability of the predicted parameters for industrial facilities.

1. Gracheva E.I., Shakurova Z.M., Abdullazyanov R.E. Comparative analysis of the most common deterministic methods for determining electricity losses in workshop networks. Problems of energy. 2019;5:87-96.

2. Gracheva E.I., Gorlov A.N., Shakurova Z.M. Analysis and evaluation of energy savings in internal plant power supply systems. Problems of energy. 2020;2:65-74.

3. Gracheva EI, Naumov OV, Gorlov AN, Shakurova Z.M. Algorithms and probabilistic models of the parameters of the functioning of intra-factory power supply. Problems of energy. 2021;1:93-104.

4. Safin AR, Khusnutdinov RR, Kopylov AM, Maksimov VV, et al. Development of a method for topological optimization of electrical machines based on a genetic algorithm. Bulletin of KSEU. 2018;4 (40):77-85.

5. Petrov TI. Modification of the genetic algorithm for complex topological optimization of the rotor of synchronous motors. Energy Problems. 2021;23(3):70-79.

6. Serebryakov NA. Application of the adaptive ensemble neural network method for shortterm forecasting of power consumption of the electrical complex of district electrical networks. Omsk Scientific Bulletin. 2021;1(175):39-45.

7. Kivchun O.R. The method of complexing the procedures of rank analysis in the management of power consumption of objects of the regional electrotechnical complex. Izvestia of the TulGU. Technical science. 2021;21:200-209.

8. Manikovsky A.S. Analysis of methods for forecasting power consumption. Collected scientific papers DONIZhT. 2021;60:23-30.

9. Zubkov S.D. Technocenological analysis of the enterprise's power consumption. Innovative science. 2020;2:20-22.

10. Khujasaidov JKh, Rusina AG, Matrenin PV, Dmitriev SA, et al. Development of power consumption forecasting models based on time series in isolated power systems. ESiK. 2020;3 (48):23-27.

11. Dziuba AP. Model of demand management for electricity of industrial enterprises connected to the networks of electricity producers. Bulletin of SurGU. 2021;1(31):12-25.

12. Lyakhomsky AV, Shadrin AA. Predictive model of power consumption based on a fully connected neural network. Electrotechnical and information complexes and systems. 2022;1. (18):107-113.

13. Zhaosu Meng, Huike Sun, Xi Wang. Forecasting Energy Consumption Based on SVR and Markov Model: A Case Study of China. Front. Environ. Sci., 06 April 2022 Sec. Environmental Economics and Management.

14. S. B. Taieb, J. W. Taylor, and R. J. Hyndman. Hierarchical probabilistic forecasting of electricity demand with smart meter data. Journal of the American Statistical Association, 2020.

15. T. Ahmad and H. Chen. Deep learning for multi-scale smart energy forecasting. Energy, vol.175. c.98112, 2019.

16. World Bank, Electric Power Consumption, The World Bank Group, Washington, D.C., USA, 2021. https://data.worldbank.org/indicator/EG.USE.ELEC.KH.PC.

17. Soluyanov Yu.I., Fedotov A.I., Galitsky Yu.Ya., Chernova N.V., Akhmetshin A.R. Actualization of the normative values of the specific electrical load of apartment buildings in the Republic of Tatarstan. Electricity. 2021;6: 62–71.

18. Soluyanov YuI, Fedotov AI, Akhmetshin AR, Soluyanov VI, et al. Analysis of the actual electrical loads of multi-apartment residential buildings in the Moscow region. Industrial Energy. 2022:4:20-28.