CC BY
4
DETERMINATION OF THE FACTORS AFFECTING THE FORECASTING OF ELECTRICITY CONSUMPTION OF MACHINE-BUILDING ENTERPRISES Rakhmonov I.U.1, Ushakov V/Y^2, Niyozov N.N.3
'Rakhmonov Ikromjon Usmonovich - Doctor of Technical Sciences, Head of the Department,
DEPARTMENT OF POWER SUPPLY, TASHKENT STATE TECHNICAL UNIVERSITY, TASHKENT, REPUBLIC OF UZBEKISTAN; 2Ushakov Vasily Yakovlevich - Doctor of Technical Sciences, Professor, SCHOOL OF POWER ENGINEERING TOMSK POLYTECHNIC UNIVERSITY, TOMSK; 3Niyozov Numon Nizomiddinovich - Doctor of Philosophy in Technical Sciences (PhD), Assistant Professor,
DEPARTMENT OF POWER SUPPLY, TASHKENT STATE TECHNICAL UNIVERSITY, TASHKENT, REPUBLIC OF UZBEKISTAN
Abstract: the article deals with the issue of determining the factors influencing the forecasting of electricity consumption, using the example of a machine-building enterprise. The number of factors affecting electricity was determined based on the expert assessment method. Factors with a high correlation between electricity and factors were determined by constructing a correlation matrix.
Keywords: electricity, factors, forecasting, correlation, evaluation, expert, algorithm, product, device.
ОПРЕДЕЛЕНИЕ ФАКТОРОВ, ВЛИЯЮЩИХ НА ПОТРЕБЛЕНИЕ ЭЛЕКТРОЭНЕРГИИ ПРЕДПРИЯТИЯМИ МАШИНОСТРОЕНИЯ, ПРИ
ПРОГНОЗИРОВАНИИ
12 3
Рахмонов И.У. , Ушаков В.Я. , Ниёзов Н.Н.
'Рахмонов Икромжон Усмонович - доктор технических наук, заведующий кафедрой, кафедра электроснабжения, Ташкентский государственный технический университет, г. Ташкент, Республика Узбекистан; 2Ушаков Василий Яковлевич - доктор технических наук, профессор, Инженерная школа энергетики Томский политехнический университет, г. Томск; 3Ниёзов Нумон Низомиддинович - доктор философии по техническим наукам, доцент,
кафедра электроснабжения, Ташкентский государственный технический университет, г. Ташкент, Республика Узбекистан
Аннотация: в статье рассматривается вопрос определения факторов, влияющих на прогнозирование потребления электроэнергии, на примере машиностроительного предприятия. Количество факторов, влияющих на электроэнергию, определялось на основе метода экспертной оценки. Факторы с высокой корреляцией между электроэнергией и факторами определялись путем построения корреляционной матрицы.
Ключевые слова: электроэнергия, факторы, прогнозирование, корреляция, оценка, эксперт, алгоритм, продукт, устройство.
UDC 62'.3''.'2
The electricity consumed in the technological process of industrial enterprises is influenced by a number of factors, and some of these factors have a greater and some less influence. Moreover, depending on the influence of these factors, it can be constant and seasonal. Factors affecting electricity can be divided into technological, energy and meteorological ones [1].
It is known that the fulfillment of bilateral tasks is the main criterion for predicting electricity consumption [2; 3; 4; 8; 10]. The first is that the error between the actual and forecast indicators in the forecast model being developed is small, and the second is to reduce the amount of fines paid to the power supply organization due to errors in determining the forecast indicators of electricity consumption [5; 6; 7].
To fulfill the conditions indicated above, when developing a forecast model for electricity consumption, energy (electricity consumption, rest, work or holidays, etc. nomenclature), affecting electricity consumption and meteorological (weather temperature, day length) factors, and develop forecasting models taking into account the main influencing factors among them.
Based on this, the number of factors affecting the electricity consumption of the Uz Truck & Bus Motors JV was determined on the basis of the peer review method, in which 7 experienced and qualified specialists of the enterprise took part in a questionnaire formed to determine the factors affecting electricity [9; 10; 11; 2; 5].
The experts who took part in the survey, as factors affecting electricity, selected 7 factors that can affect electricity, and assessed the level of influence of factors based on their experience and qualifications according to a 10-point rating system (Table 1).
Wnpor = f (AW, T, П, Кдам , , Kдав , ) (1) According to the recommendations of the experts, based on the data of Table 1, the factors affecting electricity are reflected in Figure 1.
The dependence function of the factors affecting the electricity consumption of the research object is given below:
Wnpor = f (AW, T, П, Кдам , , Kдав , )
Table 1. The results of the developed method of expert evaluation in determining the factors affecting electricity
^^^^ Expert Factors name Expert 1 Expert 2 Expert 3 Expert 4 Expert 5 Expert 6 Expert 7 Results ol collective expert evaluation
Waste of electricity in workshops 7 4 6 8 4 7 6 6
Holidays 3 5 5 8 7 10 5 6,1
Length of day 6 8 5 4 7 6 3 5,7
Working days 5 6 8 3 8 9 5 6,3
Product size 6 10 9 7 5 8 6 7,3
Loads of devices 5 7 10 7 8 7 8 7,4
Weather temperature 6 6 8 10 7 4 7 6,9
FACTORS
Weather temperature, Т ^ > f Working days, К
Product size, П Day length Klen
Loads of devices, Q Weekends/holidays, К
Waste of electricity in workshops, AW
Fig. 1. Factors affecting electricity consumption
In order to scientifically substantiate the factors affecting electricity consumption determined above based on the expert evaluation method, it is considered appropriate to determine the degree of influence of each factor on electricity by constructing a correlation matrix. Determination of the main influencing factors is performed based on the algorithm presented in Figure 2.
From the algorithm presented in Figure 2, it can be seen that the input data column (the number of factors) is determined using the Len function, and the number of factors is received separately in the form of a list. Each of these factors is subjected to regression analysis, and factors with a correlation coefficient greater than 0.6 are stored in a separate F(n) set and printed after iterations.
Fig. 2. Algorithm for determining the factors affecting electricity, block diagram
For example, factor 1 is taken and regression analysis is performed. Then it is checked whether the value of the correlation coefficient is greater than or less than 0,6. If the value of the coefficient is greater than 0,6, F is stored in the set, if it is less than 0.6, it is forgotten.
Then it is checked what factor the incoming factor (now factor 1) is. Verification is carried out by comparison with the total number of factors identified initially. If it is not the last factor, the order is incremented by one and the next factor goes through the same process. After the last factor, the print function prints to the screen the results of factors with a correlation coefficient value higher than 0,6.
Factors W Т П Qupload Kwork Kperiod Kholidays AW
W 1,00
Т 0,72 1,00
П 0,75 0,021 1,00
Qupload 0,79 0,011 0,28 1,00
Kwork 0,49 0,48 0,61 0,424 1,00
Kperiod 0,42 0,621 0,51 -0,376 0,0192 1,00
Kholidays 0,36 0,43 0,74 -0,05 -0,38 0,051 1,00
AW 0,81 0,46 0,32 0,683 0,044 0,142 0,784 1,00
As can be seen from the correlation matrix (Table 2), among the factors influencing the consumption of electricity in the machine-building industry, there are 4 factors with correlation coefficients higher than 0,6, which are included in the forecasting model: temperature; size goods; waste of electricity in workshops; device loading. It is considered appropriate to use these factors in forecasting the parameters of electricity consumption of machine-building enterprises.
References / Список литературы
1. Надтока И.И. Многофакторное моделирование электропотребления промышленного предприятия Текст. / И.И. Надтока // Известия вузов. Электромеханика, 1998. № 2/3. С. 72-74.
2. Rakhmonov I.U., Ushakov V.Ya., Niyozov N.N., Kurbonov N.N., Mamutov М. Energy saving in industry // Energy Systems Research 2021. E3S Web of Conferences 289, 07014 (2021). doi.org/10.1051/e3sconf/202128907014.
3. Воронов И.В., Политое Е.А., Ефременко В.М. Обзор типов исскуственных нейронных сетей и методов их обучения // Вестник КузГТУ, 2007. № 3. С. 38-42.
4. Сеетозарое В.В. Основы статистической обработки результатов измерений. Учебное пособие. М. Изд.: МИФИ, 2005, 36-40 с.
5. Кирпичникоеа И.М. Исследование методов прогнозирования электропотребления сбытового предприятия / И.М. Кирпичникова, К.Л. Соломахо // Электротехнические системы и комплексы, 2014. № 3. С. 39-43.
6. Рахмоное И. У. Кора металлургия корхоналарида электр энргия истеъмоли самарадорлигини ошириш: дисс. Техника фанлари буйича фалсафа доктори (PhD), 2018. 108 бет.
7. Rakhmonov I. U., Niyozov N.N., Najimova A.M. Improving the reliability of electrical equipment in rural areas // Science and Education in Karakalpakstan, 2020. № 1. Pp. 51-57 (05.00.00 № 27).
8. Рахмоное И.У., Жалилоеа Д.А. Рационализация режима работы вентиляционных, водоснабжающих и осветительных установок на предприятиях текстильной промышленности // Научно-методический журнал "Academy". № 8 (71). Декабрь, 2021. Стр. 13-15.
9. Рахмоное И. У., Тоирое М.М. Наивыгоднейшие режимы энергоемких потребителей промышленных предприятий с различным технологическим процессом // Издательство «Проблемы науки». " European science". № 6 (62). Декабрь, 2021. Стр. 17-19.
10. Рахмоное И.У., Нажимоеа А.М. Оценка влияния энергетических, технологических и эксплуатационных факторов на показатели удельного расхода электроэнергии на единицу выпускаемой продукции // Научно-методический журнал "Проблемы науки". № 8 (67). Ноябрь, 2021. Стр. 20-22.
11. Рахмоное И.У., Зияеуддиное А.Ф. Исследование закономерности изменения параметров электропотребления промышленных предприятий // Научно-методический журнал "Проблемы современной науки и образования", 2021. № 9 (166). Октябрь 2021. Стр. 17-20.
