CC BY
7
DEVELOPMENT OF SOFTWARE FOR FORECASTING RATIONAL ENERGY CONSUMPTION AT INDUSTRIAL ENTERPRISES Rakhmonov I.U.1, Kurbonov N.N.2, Kholikhmatov B.B.3
1Rakhmonov Ikromjon Usmonovich - Doctor of Technical Science (DSc), Head of Department;
2Kurbonov Nurbek Nurullo ugli - doctoral Student; 3Kholikhmatov Bakhriddin Berdi ugli - Assistant,
DEPARTMENT OF POWER SUPPLY, TASHKENT STATE TECHNICAL UNIVERSITY, TASHKENT, REPUBLIC OF UZBEKISTAN
Abstract: nowadays, planning electricity consumption is important for improving its efficiency and, as a result, increasing the competitiveness of manufactured products by reducing the share of electricity costs in the cost of production. There are several software programs that differ from each other in terms of calculation methodology, interface, and technical capabilities. Despite the abundance of such software, the lack of some required features has created the need for new software with the required features. As a result of scientific research, new software has been developed, and this article discusses features that are different from existing software. Keywords: forecasting, application, plan, efficiency, consumption, demand, LSTM, GMDH, ARIMA, indicators.
РАЗРАБОТКА ПРОГРАММНОГО ОБЕСПЕЧЕНИЯ ПО ПРОГНОЗИРОВАНИЮ РАЦИОНАЛЬНОГО ЭНЕРГОПОТРЕБЛЕНИЯ НА ПРОМЫШЛЕННЫХ
ПРЕДПРИЯТИЯХ
1 2 3
Рахмонов И.У. , Курбонов Н.Н. , Холихматов Б.Б.
1Рахмонов Икромжон Усмонович - доктор технических наук, заведующий кафедрой; 2Курбонов Нурбек Нурулло угли - докторант;
3Холихматов Бахриддин Берди угли - ассистент, кафедра электроснабжения, Ташкентский государственный технический университет, г. Ташкент, Республика Узбекистан
Аннотация: в настоящее время прогнозирование потребления электроэнергии имеет важное значение для повышения его эффективности и, как следствие, повышения конкурентоспособности выпускаемой продукции за счет снижения доли затрат на электроэнергию в себестоимости продукции. Существует несколько программ, отличающихся друг от друга методикой расчета, интерфейсом и техническими возможностями. Несмотря на обилие таких программных обеспечений, отсутствие некоторых необходимых функций создало потребность в новом программном обеспечении с необходимыми функциями. В результате научных исследований было разработано новое программное обеспечение, и в этой статье рассматриваются функции, которые отличаются от существующих программных обеспечений. Ключевые слова: прогнозирование, приложение, план, эффективность, потребление, LSTM, GMDH, ARIMA, индикаторы, программное обеспечение.
Forecasting the consumption of electrical energy is important to improve its efficiency and, as a result, to improve the competitiveness of manufactured products by reducing the share of electricity costs in the cost of production. When determining the forecast indicators of electricity consumption by industrial enterprises, it is advisable to use modern high-precision forecasting methods.
There is several forecasting software available today. Their calculation methodology, the principle of operation, and results printing are different. At the same time, the field of application is different.
Among them, we will list some of the most used software today.
1. Workday Adaptive Planning - Workday Adaptive Planning provides business planning through corporate performance management (CPM), sales planning, and workforce planning. Workday Adaptive Planning leads the way for people in companies to collaborate, gain insights, and make smarter decisions faster. Powerful modeling for any size organization, yet so easy for everybody who plans [1].
2. GMDH Shell electricity forecasting software is quite easy to use thanks to a number of applicable templates designed for various industries including electricity. The simplification doesn't reduce its wide capabilities though, so once you master the program in full you'll be able to fine-tune any template to your exact preferences [2].
3. Forecast Pro is used across virtually all industries and puts sophisticated forecasting techniques into anyone's hands. It is powerful & accurate, yet easy-to-use and quick to implement — you can be up and running in just days, or even hours [3].
4. LoadFor™ is a self-learning system based on machine learning. On the basis of historical electricity load, historical meteorological data and meteorological forecasts, the system is not only able to predict the electricity load, but can also automatically and continuously calibrate and improve its predictions as it is fed with more data. In addition, online power measurements can be used as input (if available) to increase forecast accuracy [4].
In contrast to this software, the newly developed forecasting software has the following advantages:
1. Direct link- prepared software has a section for creating daily, monthly, quarterly, and annual reports of the enterprise directly, and it is not necessary to enter the initial data separately for forecasting. The program is directly linked to incoming data. The user only needs to select the forecast period.
2. External data reception function - it is possible to perform forecasting based on external data.
3. Extensive methodology - the software makes predictions based on three complementary methods at the same time. The obtained results display the results of the software method with the least error and provide an opportunity to see other results. These methods are based on models such as ARIMA, GMDH, LSTM.
Fig. 1. Model development window for predicting electricity consumption by workshops and departments of an industrial
enterprise
Fig. 2. The window for analyzing the results offorecasting the power consumption of workshops and divisions of an industrial
enterprise
This part of a single automated program is based on the models and methods developed for the prediction problem. According to the interface shown in Figure 1, the user first selects a prediction model. For the 3 cases under consideration, the integrated moving average autoregression method, the principal component method, and the artificial neural population method are selected. Based on these methods, forecasting models are developed.
At the same time, the task is to develop a one- or multi-factor forecasting model. In a single-factor forecasting model, the main input parameter is the history of electricity consumption data, in a multi-factorial model, the factors that affect electricity consumption; form the initial database. In this case, the data is taken from the INPUTBASE.db database and the calculations are based on the forecasting models developed by the authors of the article.
According to the interface shown in Figure 2, errors in the methods used in forecasting in tabular and graphical form can be identified.
Fig. 3. The result window of the selected method for predicting the power consumption of workshops and divisions of an
industrial enterprise
In the interface shown in Figure 3, the forecasting method, its accuracy, and the forecast indicators determined using the forecasting model are presented in tabular and graphical form; the user can download them in a convenient form (pdf, xlsx, docx). Conclusion
In conclusion, various methods and software are used in planning processes such as energy production and consumption. Today, there are several forecasting software, and their working technology and methodology are also different. However, the presence of functions that connect directly to the input data, perform a forecast based on several models and compare them, and print the result with the highest accuracy, makes the new software important.
References / Список литературы
1. [Electronic Resource]. URL: https://softwareconnect.com/epm/adaptive-planning/ (date of access: 07.10.2022).
2. [Electronic Resource]. URL: https://gmdhsoftware.com/electricity-load-forecasting-software/?ysclid=l8ibceh0rr561562581/ (date of access: 07.10.2022).
3. [Electronic Resource]. URL: https://www.forecastpro.com/forecast-pro-landing-page/?gclid=EAIaIQobChMIs6yw7Nmx-gIV5kWRBR0GlQYPEAAYAiAAEgIJ-_D_BwE/ (date of access: 07.10.2022).
4. [Electronic Resource]. URL: https://gmdhsoftware.com/business-forecasting-software/?ysclid=l8iaerd3nk445988641/ (date of access: 07.10.2022).
5. William J. Bezdek. JoelMaleport Robert Z Olshan. Live, Virtual & Constructive Simulation for Real Time Rapid Prototyping, Experimentation and Testing using Network Centric Operations.
6. Rakhmonov I. U., Zhalilova D.A. Ratsionalizatsiya rezhima raboty ventilyatsionnykh, vodosnabzhayushchikh i osvetitel'nykh ustanovok na predpriyatiyakh tekstil'noy promyshlennosti // Nauchno-metodicheskiy zhurnal "Academy". № 8 (71), 2021. Dekabr'. Str. 13-15.
7. Rakhmonov I.U., Toirov M.M. Naivygodneyshiye rezhimy energoyemkikh potrebiteley promyshlennykh predpriyatiy s razlichnym tekhnologicheskim protsessom // Izdatel'stvo «Problemy nauki» " European science", 2021. № 6 (62). Dekabr'. Str. 17-19.
8. Rakhmonov I. U., Nazhimova A.M. Otsenka vliyaniya energeticheskikh, tekhnologicheskikh i ekspluatatsionnykh faktorov na pokazateli udel'nogo raskhoda elektroenergii na yedinitsu vypuskayemoy produktsii // Nauchno-metodicheskiy zhurnal "Problemy nauki". № 8 (67), 2021. Noyabr'. Str. 20-22.
9. Rakhmonov I.U., Ziyavuddinov A.F. Issledovaniye zakonomernosti izmeneniya parametrov elektropotrebleniya promyshlennykh predpriyatiy // Nauchno-metodicheskiy zhurnal "Problemy sovremennoy nauki i obrazovaniya", 2021. № 9 (166). Str. 17-20.
