KR20220157080A - Method and apparatuses for prediction for predicting electrical generation of special day - Google Patents

Abstract

According to an embodiment of the present invention, a power demand prediction method of a power demand prediction device for a special day comprises the following steps of: predicting power demand on a special day to be predicted by applying power demand data of a past special day to a fuzzy linear regression analysis model; determining whether temperature sensitivity is corrected based on a forecast temperature of the special day to be predicted; calculating temperature sensitivity of the special day according to the determination result to correct the power demand of the predicted special day; and estimating and reflecting the amount of BTM solar power generation to the corrected power demand of the special day.

Description

Apparatus and method for predicting power demand on special days

The present invention relates to an apparatus and method for predicting electric power demand on a special day, and more specifically, to determine the electric power demand on a special day reflecting the influence of temperature that varies by hour and the variability of electric power demand by BTM (Behind-the-Meter) solar power generators. It relates to a prediction device and method.

Power demand forecasting technology is an essential element for stable and economical power system operation. In particular, it is expected that the importance of short-term electricity demand forecasting will increase for the efficient use of the power system.

In the case of such a short-term power demand forecast, hourly power demand for the next day is predicted for the operation of the power market and power system. Here, the power demand may be divided into Monday, Tuesday through Friday, Saturday, Sunday, special day, and special light load period according to the daily power demand pattern. Special days refer to statutory holidays and temporary holidays, and special days refer to the 7 days before and after, including New Year's Day and Chuseok, and the concentrated period of summer vacation.

Compared to the power demand on normal days, the power demand on special days is characterized by relatively low power demand, irregular patterns of power demand, and a limited number of past data that can be used for prediction. In addition, in the case of a special day following the lunar calendar, the position of the special day on the solar calendar changes every year, so seasonal characteristics may be reflected differently. Due to these characteristics, power demand on special days is excluded from neural network-based prediction model learning or predicted separately from power demand on normal days.

Various methods such as expert techniques, neural networks, relative coefficients, and fuzzy linear regression analysis are applied to predict electricity demand on special days. It is relatively high compared to the error of forecasting power demand per day, and it needs improvement.

Electricity demand changes under the influence of social and economic factors in the medium and long term, and changes under the influence of the weather in the short term. In particular, among several factors, temperature has a great influence on the use of cooling and heating loads. In addition, as the supply of solar power generators increases, behind-the-meter (BTM) solar power generators, which do not measure power generation in real time, are increasing, which increases the uncertainty of power demand forecasting. However, the prediction method based on fuzzy linear regression analysis currently in use has a disadvantage in that it does not reflect the effect of temperature that varies over time and the variability of power demand by BTM solar power generators. 1 is a diagram showing an error in predicting power demand on a special day. As shown in FIG. 1, the error in predicting power demand on special days tends to increase year by year.

Therefore, in order to improve the accuracy of electricity demand forecasting on special days, it is required to develop an algorithm for forecasting hourly electricity demand on special days that can reflect the influence of temperature that varies by hour and the influence of BTM solar power generators that increase by year.

In order to solve the above problems, an apparatus and method for predicting power demand on a special day, reflecting the influence of hourly temperature and the variability of power demand by BTM (Behind-the-Meter) photovoltaic generators, are provided.

However, the object of the present invention is not limited to the above-mentioned matters, and other objects not mentioned will be clearly understood by those skilled in the art from the description below.

In order to achieve the above object, a method for predicting power demand of an apparatus for predicting power demand for special days according to an embodiment of the present invention applies power demand data of past special days to a fuzzy linear regression analysis model to predict special days. predicting power demand; determining whether temperature sensitivity is corrected based on the predicted temperature of the special day to be predicted; Calculating the temperature sensitivity of the special day according to the determination result and correcting the electricity demand of the predicted special day; and predicting and reflecting the amount of BTM photovoltaic power generation to the corrected electricity demand on the special day.

In one embodiment, the step of determining whether to correct the temperature sensitivity may include determining that temperature sensitivity correction is necessary when the predicted temperature of the special day to be predicted does not fall within the cooling/heating reference temperature of 18°C to 26°C. do.

In one embodiment, the calculating the temperature sensitivity of the special day and correcting the predicted electricity demand on the special day may include calculating the basic electricity demand on the special day; normalizing power demand using the basic power demand; Calculating temperature sensitivity for each time interval through regression analysis of normalized electricity demand and temperature; and correcting power demand on a special day based on the calculated temperature sensitivity for each time section.

In one embodiment, the step of correcting the power demand on the special day is characterized by correcting the power demand on the special day according to Equation 1 below.

[Equation 1]

는 기온 차이만큼 보정된 특수일 t시의 전력수요예측 값, L_t는 특수일 t시의 전력수요예측 값,

는 특수일 p구간의 기온민감도,

은 m월 p구간의 기본 전력수요,

는 특수일 t시의 예보기온,

는 특수일 이전 평일 4일의 t시 기온 평균값,

는 과거 3개년의 동일 특수일 t시의 기온 평균값,

는 과거 3개년의 동일 특수일 이전 평일 4일의 t시 기온 평균값을 의미한다.In Equation 1, p means a time interval, m is the month where the special day is located, t is the time,

is the predicted power demand at special day t corrected by the temperature difference, L _t is the predicted power demand at special day t,

is the temperature sensitivity of the special day p section,

is the basic electricity demand for the period p in month m,

is the predicted temperature at time t on a special day,

is the average temperature at time t of the 4 weekdays prior to the special day,

is the average temperature at time t on the same special day in the past three years,

means the average temperature at time t of 4 weekdays prior to the same special day in the past 3 years.

In one embodiment, the step of predicting and reflecting the BTM solar power generation amount to the corrected power demand of the special day is to reflect the BTM solar power generation amount to the corrected power demand of the special day according to Equation 2 below. to be characterized

[Equation 2]

는 BTM 태양광 발전기의 영향이 반영된 특수일 t시간의 전력수요예측 값,

는 기온에 의한 영향이 반영된 특수일 t시간의 전력수요예측 값,

는 특수일의 BTM 태양광 설비용량 추정치,

는 설비용량으로 정규화 된 특수일 t시의 BTM 태양광 발전량 추정치,

는 설비용량으로 정규화된 과거 3개년의 동일 특수일 t시의 BTM 태양광 발전량 추정치평균값을 나타낸다.In Equation 2, t is time,

is the predicted power demand at time t on a special day reflecting the influence of the BTM photovoltaic generator,

is the predicted power demand at time t on a special day reflecting the influence of temperature,

is the estimated BTM solar installed capacity on a special day,

is the estimated BTM solar power generation capacity at special day t, normalized to installed capacity,

represents the estimated average value of BTM photovoltaic power generation at time t on the same special day for the past three years normalized to installed capacity.

An apparatus for predicting power demand on a special day according to another embodiment of the present invention includes a memory for storing power demand data and a fuzzy linear regression analysis model on a special day in the past; a power demand prediction unit for predicting power demand on a special day to be predicted by applying the power demand data of the past special day to a fuzzy linear regression analysis model; a temperature correction unit that determines whether or not to correct temperature sensitivity based on the predicted temperature of the special day to be predicted, calculates the temperature sensitivity of the special day according to the determination result, and corrects electricity demand on the predicted special day; and a BTM solar power generation amount reflection unit that predicts and reflects the BTM solar power generation amount to the corrected electricity demand on a special day.

According to the apparatus and method for predicting power demand on a special day according to the present invention, it is possible to accurately predict power demand by reflecting the influence of hourly temperature and the variability of power demand by BTM (Behind-the-Meter) solar power generators. .

1 is a diagram showing an error in predicting power demand on a special day.
2 is a block diagram for explaining a schematic configuration of an apparatus for predicting power demand on a special day according to an embodiment of the present invention.
3 is a block diagram illustrating a configuration of a control unit of an apparatus for predicting power demand on a special day according to some embodiments.
4 is a flowchart illustrating a method for predicting power demand on a special day according to an embodiment of the present invention.
5 is a flowchart illustrating a method for predicting power demand on a special day based on a fuzzy linear regression model according to an embodiment of the present invention.
6 is a flowchart illustrating temperature correction according to an embodiment of the present invention.

Since the present invention can have various changes and various embodiments, specific embodiments will be described in detail with reference to the drawings. However, this is not intended to limit the present invention to specific embodiments, and should be understood to include all modifications, equivalents, or substitutes included in the spirit and technical scope of the present invention. Like reference numerals have been used for like elements throughout the description of each figure.

Terms such as first, second, A, and B may be used to describe various components, but the components should not be limited by the terms. These terms are only used for the purpose of distinguishing one component from another. For example, a first element may be termed a second element, and similarly, a second element may be termed a first element, without departing from the scope of the present invention. The term and/or includes a combination of a plurality of related items or any one of a plurality of related items.

Terms used in this application are only used to describe specific embodiments, and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this application, the terms "include" or "have" are intended to designate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, but one or more other features It should be understood that the presence or addition of numbers, steps, operations, components, parts, or combinations thereof is not precluded.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related art, and unless explicitly defined in the present application, they should not be interpreted in an ideal or excessively formal meaning. don't

Throughout the specification and claims, when a part includes a certain component, it means that it may further include other components, not excluding other components unless otherwise stated.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

2 is a block diagram for explaining a schematic configuration of an apparatus for predicting power demand on a special day according to an embodiment of the present invention.

Referring to FIG. 2 , the apparatus 100 for predicting power demand on a special day includes an input unit 110 , an output unit 120 , a control unit 130 and a memory 140 . The apparatus 100 for predicting power demand on a special day may further include a user input unit 110 and an output unit 120 . However, the apparatus 100 for predicting power demand on a special day may be implemented with more components than those shown in FIG. 1 or fewer components than shown.

The user input unit 110 means a means through which a user inputs data for controlling the power demand forecasting device 100 on a special day. For example, the user input unit 110 includes a key pad, a dome switch, and a touch pad (contact capacitance method, pressure resistive film method, infrared sensing method, surface ultrasonic conduction method, integral type tension measuring method, piezo effect method, etc.), a jog wheel, a jog switch, and the like, but are not limited thereto.

According to some embodiments, the user input unit 110 may receive an input for specifying a special day to be predicted by the user.

The output unit 120 may output an audio signal, a video signal or a vibration signal. Although not shown in FIG. 1 , the output unit 120 may include at least one of a display unit, a sound output unit, and a vibration motor.

According to some embodiments, the output unit 120 may output a result of performing an operation according to a request for predicting power demand on a special day to be predicted by a user. For example, when a request for predicting power demand on a special day to be predicted by the user occurs, the controller 130 may perform a predetermined operation, and the result of the predetermined operation is output through the output unit 120. It can be.

The control unit 130 typically controls the overall operation of the device 100 for predicting power demand on a special day. For example, the controller 130 may generally control the user input unit 110, the output unit 120, the communication unit 140, and the like by executing programs stored in the memory 140.

The controller 130 may be configured to process commands of a computer program by performing basic arithmetic, logic, and input/output operations. The command may be provided to the control unit 130 from the memory 140 or may be received through the communication unit 140 and provided to the control unit 130 . For example, the controller 130 may be configured to execute commands according to program codes stored in a recording device such as a memory.

According to some embodiments, the control unit 130 determines the effect of temperature and a behind-the-meter (BTM) solar power generator based on past data and analysis models stored in the memory 140 to predict power demand on a special day. Generates electricity demand forecasting results for special days reflecting the variability of electricity demand due to A detailed description thereof will be described later with reference to other drawings.

The memory 140 may store programs for processing and control by the control unit 130, and may store data input to or output from the power demand prediction device 100 for special days. may be

In one embodiment, the memory 140 may store various types of information related to power demand data in the past, weather performance by region, weather forecast by region, and the like. Also, the memory 140 may store a fuzzy linear regression model.

As an example of the memory 140, a flash memory type, a harddisk type, a multimedia card micro type, a card type memory (eg SD or XD memory, etc.), RAM (Random Access Memory), SRAM (Static Random Access Memory), ROM (ROM, Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), PROM (Programmable Read-Only Memory), magnetic memory, A storage medium of at least one of a magnetic disk and an optical disk may be used, but is not limited thereto.

3 is a block diagram illustrating a configuration of a control unit of an apparatus for predicting power demand on a special day according to some embodiments.

Referring to FIG. 3 , the control unit 130 may include a power demand estimation unit 132, a temperature correction unit 134, and a BTM solar power generation amount reflection unit 136. The power demand forecasting unit 132, the temperature correction unit 134, and the BTM photovoltaic power generation amount reflection unit 136 are shown separately to explain their respective functions. Accordingly, the controller 130 may include at least one processor configured to perform functions of the power demand predictor 132, the temperature corrector 134, and the BTM solar power generation amount reflection unit 136, respectively.

The power demand prediction unit 132 predicts the power demand on a special day based on a fuzzy linear regression analysis model based on the power demand data on a past special day. In one embodiment, the power demand predictor 132 uses the same special day of the past three years stored in the memory 140 and the power demand of four weekdays prior to the special day to estimate the regression coefficient of the fuzzy linear regression analysis.

The electric power demand on a special day predicted by the electric power demand forecasting unit 132 is the electric power demand in which the effect of temperature and the variability of electric power demand by the BTM photovoltaic generator are not reflected.

The temperature correction unit 134 determines whether to correct the temperature sensitivity based on the predicted temperature of the predicted day, and then reflects the effect of temperature on the electricity demand primarily predicted by the power demand prediction unit 132 to predict the special day. corrects the power demand of Here, the temperature sensitivity means the amount of change in electric power demand due to a change in unit temperature, and is used to correct the electric power demand on a special day by the difference between the past temperature and the predicted temperature of the predicted day.

In one embodiment, the temperature correction unit 134 computes the temperature sensitivity by calculating a ratio of the difference between the average temperature difference between the past special day and the average temperature of the 4 weekdays prior to the special day and the average power demand of the 4 weekdays prior to the special day and the special day. yield

The temperature correction unit 134 determines whether the predicted temperature of the forecast day affects the electricity demand of the forecast day, calculates the temperature sensitivity, and corrects the electricity demand of the special day based on the calculated temperature sensitivity.

The BTM solar power generation amount reflecting unit 136 predicts and reflects the BTM solar power generation amount to the corrected power demand of the special day.

4 is a flowchart illustrating a method for predicting power demand on a special day according to an embodiment of the present invention, and FIG. 5 is a flowchart illustrating a method for predicting power demand on a special day based on a fuzzy linear regression analysis model according to an embodiment of the present invention. FIG. 6 is a flowchart illustrating temperature correction according to an embodiment of the present invention.

Referring to FIG. 4 , in step S110 , electricity demand on a special day is predicted based on a fuzzy linear regression analysis model based on data on power demand on a special day in the past.

Referring to FIG. 5 , step S110 will be described in detail. In step S112, in one embodiment, a regression coefficient of a fuzzy linear regression analysis model is estimated based on the power demand for a special day in the past stored in the memory 140. In one embodiment, the special day in the past may be the power demand on the same special day in the past and the power demand on weekdays four days prior to the special day.

In step S114, the maximum and minimum power demand of the special day is predicted by fuzzy linear regression analysis using the estimated regression coefficient.

In step S116, the maximum and minimum normalized 24-hour power demand patterns of the same special day of the past three years are predicted.

In step S118, the electric power demand of the 24-hour special day is predicted based on the predicted maximum and minimum power demand and the maximum and minimum normalized 24-hour electric power demand pattern.

Referring back to FIG. 3 , in step S120, it is determined whether or not the temperature sensitivity is corrected based on the predicted temperature of the predicted day.

In one embodiment, whether or not to correct the temperature is determined according to whether or not the predicted temperature of a special day to be predicted (hereinafter referred to as a predicted day) belongs to a preset temperature range before reflecting the effect of the temperature. The preset temperature range is a reference temperature range for cooling and heating, and is 18°C to 26°C.

When the predicted temperature of the predicted day falls within the range of the reference temperature for cooling and heating, it is determined that the effect of the temperature on the power demand is relatively small, and the temperature correction is not performed, and the process proceeds to step S140.

In step S120, when the predicted temperature of the predicted day does not fall within the range of the cooling/heating reference temperature, the process proceeds to step S130 to analyze the past temperature and power demand to calculate the temperature sensitivity.

In step S131, the basic power demand of the special day is calculated to calculate the temperature sensitivity of the special day. Here, the basic electricity demand refers to electricity demand in which the amount of variation in electricity demand based on GDP is reflected, and the variation in electricity demand caused by temperature and other factors is excluded. The power demand is normalized using the basic power demand calculated in step S132. In the next step S133, temperature sensitivity is calculated through regression analysis between normalized power demand and temperature. In order to calculate the temperature sensitivity of a special day, the temperature sensitivity of each time interval is calculated. The calculated temperature sensitivity is shown in Equation 1.

[Equation 1]

는 p구간의 특수일 기온민감도,

는 기본전력수요로 정규화 된 p구간의 특수일 전력수요,

는 기본전력수요로 정규화 된 p구간의 특수일 이전 평일 4일의 전력수요 평균,

는 p구간의 특수일 기온,

는 p구간의 특수일 이전 평일 4일의 기온 평균을 의미한다.Here, p means 8 sections of 3 time units,

is the special day temperature sensitivity of the p section,

is the power demand for a special day in section p normalized to the basic power demand,

is the average power demand for the 4 weekdays prior to the special day in section p normalized to the basic power demand,

is the temperature on a special day in section p,

means the average temperature of the 4 weekdays prior to the special day in section p.

In step S134. Based on the calculated temperature sensitivity of the special day, the predicted power demand is corrected. The special day power demand correction formula is shown in Equation 2.

[Equation 2]

는 기온 차이만큼 보정된 특수일 t시의 전력수요예측 값, L_t는 특수일 t시의 전력수요예측 값,

는 특수일 p구간의 기온민감도,

은 m월 p구간의 기본 전력수요,

는 특수일 t시의 예보기온,

는 특수일 이전 평일 4일의 t시 기온 평균값,

는 과거 3개년의 동일 특수일 t시의 기온 평균값,

는 과거 3개년의 동일 특수일 이전 평일 4일의 t시 기온 평균값을 의미한다.Here, p denotes a section (e.g., 8) divided by a plurality of 24 hours, and 1, 2, 3, 4, 5, 6, 7, and 8 of p are respectively 1 to 3, 4 to 6, 7 to 9 o'clock, 10 to 12 o'clock, 13 to 15 o'clock, 16 to 18 o'clock, 19 to 21 o'clock, 22 to 24 o'clock. m is the month in which the special day is located, t is the hour,

is the predicted power demand at special day t corrected by the temperature difference, L _t is the predicted power demand at special day t,

is the temperature sensitivity of the special day p section,

is the basic electricity demand for the period p in month m,

is the predicted temperature at time t on a special day,

is the average temperature at time t of the 4 weekdays prior to the special day,

is the average temperature at time t on the same special day in the past three years,

means the average temperature at time t of 4 weekdays prior to the same special day in the past 3 years.

Returning to FIG. 3 again, in step S140, the amount of BTM photovoltaic power generation is predicted and reflected to the corrected electricity demand of the special day.

In order to reflect the influence of the BTM solar power generator, the amount of BTM solar power generation on the forecast day is estimated and reflected in the electricity demand forecast on the special day.

On the other hand, domestic photovoltaic generators can be divided into photovoltaic generators participating in the electricity market that trade produced power through the power market, KEPCO photovoltaic generators that trade produced power with Korea Electric Power Corporation, and other photovoltaic generators. can Photovoltaic generators participating in the power market are acquiring power generation in real time, and KEPCO solar power generators and other solar power generators do not acquire power generation in real time. Classify. Since some of the BTM photovoltaic generators do not have information on location, efficiency, installation year, etc., in order to estimate the hourly power generation of BTM photovoltaic generators, the generation efficiency of BTM solar photovoltaic generators is acquired in both facility capacity and real-time generation. Participation in the electricity market It is assumed that the efficiency of a solar generator is similar.

The hourly power generation of the BTM photovoltaic generator is estimated using a proportional expression between the installed capacity of the photovoltaic generator and the power generation based on Equation 3.

[Equation 3]

는 d일 t시의 BTM 태양광 발전량 추정치,

는 d일 t시의 전력시장 참여 태양광 발전기의 발전량,

는 d일의 전력시장 참여 태양광 발전기의 설비용량,

는 d일의 BTM 태양광 발전기의 설비용량 추정치를 의미한다. 수학식 3과 같이 BTM 태양광 발전량을 추정하기 위해서는 BTM 태양광 발전기의 설비용량이 필요하며, BTM 태양광 발전기의 설비용량은 수학식 4와 같이 나타낼 수 있다. where d is the date, t is the time,

is the estimated BTM solar power generation at time t on day d,

is the amount of power generated by solar power generators participating in the electricity market at time t on day d,

is the installed capacity of solar power generators participating in the electricity market on day d,

is the estimated installed capacity of the BTM photovoltaic generator on day d. As shown in Equation 3, in order to estimate the amount of BTM solar power generation, the installed capacity of the BTM solar power generator is required, and the installed capacity of the BTM solar power generator can be expressed as shown in Equation 4.

[Equation 4]

는 d일의 BTM 태양광 발전기 설비용량 추정치,

는 d일의 한국전력 태양광 발전기 설비용량,

는 d일의 기타 태양광 발전기 설비용량 추정치를 의미한다. BTM 태양광 발전기의 설비용량 중 한국전력 태양광 발전기의 설비용량은 취득되고 있지만, 기타 태양광 발전기의 정확한 설비용량을 알 수 없기 때문에 기타 태양광 발전기의 설비용량에 대한 추정이 필요하다. 기타 태양광 발전기의 설비용량을 추정하기 위해 기온이 전력수요에 미치는 영향이 다른 월에 비해 상대적으로 작은 4월과 5월, 9월과 10월 중 평일 낮 시간의 전력수요 차이는 BTM 태양광 발전량에 의해 발생한다고 가정한다. 따라서 기타 태양광 발전기의 설비용량을 추정하기 위해 4월과 5월 또는 9월과 10월의 평일 데이터가 사용된다.here,

is the estimated installed capacity of the BTM solar generator on day d,

is the installed capacity of the KEPCO solar power generator on day d,

is the estimated installed capacity of other solar PV generators on day d. Among the installed capacities of BTM photovoltaic generators, the installed capacity of KEPCO's photovoltaic generators is acquired, but since the exact installed capacity of other photovoltaic generators is unknown, it is necessary to estimate the installed capacities of other photovoltaic generators. In order to estimate the installed capacity of other solar power generators, the difference in power demand during the daytime on weekdays during April and May, and September and October, when the effect of temperature on power demand is relatively small compared to other months, is the BTM solar power generation amount is assumed to be caused by Therefore, weekday data from April and May or September and October are used to estimate the installed capacity of other photovoltaic generators.

Since the power demand of the power generation stage used for the operation of the power system is currently measured at the output terminal of the generator, the BTM photovoltaic power generation amount, which is not measured in real time, reduces the power demand. Therefore, among the data used, the daytime power demand on a sunny day with a lot of solar power generation is relatively less than that on a cloudy day, and the variability of power demand by BTM solar power generators is obtained by adding the BTM solar power generation amount to the power demand. , the daytime power demand deviation between sunny days and cloudy days decreases. Therefore, the installed capacity of other photovoltaic generators is changed, and the installed capacity of other photovoltaic generators is searched for at a point where the daytime power demand deviation between a sunny day and a cloudy day is minimized. Figure 3 is a pseudocode showing how to estimate the installed capacity of BTM solar power generators in April and May, and the installed capacity of BTM solar power generators in September and October is also estimated in the same way.

Here, sunny days and cloudy days are classified according to the area occupied by clouds. In one embodiment, sunny may mean a sky state in which cloudiness is less than 20% of the total sky, and cloudy may mean a sky state in which cloudiness is 80% or more of the total sky. This information can be obtained from the Korea Meteorological Administration server.

The following pseudo code shows how to estimate the installed capacity of a BTM solar power generator.

Here, Equations 5, 6, and 7 shown are as follows,

[Equation 5]

[Equation 6]

[Equation 7]

,

는 각각 d일의 기타 태양광 발전기의 설비용량 추정치와 한국전력 태양광 발전기의 설비용량,

는 d일 t시의 전력수요 Load_d,t에 d일 t시의 추정된 BTM 태양광 발전량

이 더해진 전력수요를 의미한다. n 은 4월과 5월 중 평일의 개수를 의미하며, 맑은 날과 흐린 날은 각각 한 개 이상씩 존재해야 한다. In addition, ratio means the ratio of the installed capacity of other photovoltaic generators to the installed capacity of KEPCO's photovoltaic generator, and has a value ranging from r _min (0%) to r _max (100%).

,

is the estimated installed capacity of other photovoltaic generators on day d and the installed capacity of KEPCO's photovoltaic generator, respectively,

is the estimated BTM photovoltaic power generation at d day t at load _d,t

This is the added power demand. n means the number of weekdays in April and May, and there must be at least one sunny day and one cloudy day each.

는 n개의 BTM 태양광 발전량이 더해진 전력수요 중 13시의 평균값, Std₁₃는 n개의 BTM 태양광 발전량이 더해진 전력수요 중 13시의 표준편차를 의미하며, 기타 태양광 설비용량 추정은 태양광 발전량이 최대인 13시를 기준으로 한다. 즉, BTM 태양광 발전기의 설비용량을 추정하기 위해 기타 태양광 발전기의 설비용량을 한국전력 태양광 발전기의 설비용량 대비 비율로 계산하고, 이 비율을 변경해가며 BTM 태양광 발전량이 더해진 전력수요의 13시 편차가 최소가 되는 지점의 비율을 탐색한다. 분석 결과 기타 태양광 발전기의 설비용량이 한국전력 태양광 발전기 설비용량의 50%일 때 낮 시간의 전력수요 편차가 최소가 되었다. 따라서 한국전력 태양광 발전기 설비용량의 50%를 기타 태양광 발전기의 설비용량으로 한다.

is the average value at 13:00 of the power demand with n BTM solar power generation added, Std ₁₃ means the standard deviation at 13:00 among the power demand with n BTM solar power added This is based on the maximum of 13 o'clock. In other words, to estimate the installed capacity of the BTM photovoltaic generator, the installed capacity of other photovoltaic generators is calculated as a ratio of the installed capacity of KEPCO's photovoltaic generator, and this ratio is changed to obtain 13 The ratio of the points where the time deviation is the minimum is searched. As a result of the analysis, when the installed capacity of other photovoltaic generators was 50% of the installed capacity of KEPCO's photovoltaic generator, the variation in power demand during daytime was minimized. Therefore, 50% of KEPCO's photovoltaic generator installed capacity is used as the installed capacity of other photovoltaic generators.

In order to reflect the influence of BTM solar power generation on the forecast of electricity demand on special days, the average value of the estimated BTM solar power generation amount on the forecast date and the estimated BTM solar power generation amount on the same special day for the past 3 years is calculated as shown in Equation 8, and the two values are Reduce the electricity demand on the forecast day by the difference. At this time, in order to consider the annual increase in BTM facility capacity, the hourly power generation of BTM is normalized to facility capacity before use.

[Equation 8]

는 BTM 태양광 발전기의 영향이 반영된 특수일 t시간의 전력수요예측 값,

는 기온에 의한 영향이 반영된 특수일 t시간의 전력수요예측 값,

는 특수일의 BTM 태양광 설비용량 추정치,

는 설비용량으로 정규화 된 특수일 t시의 BTM 태양광 발전량 추정치, PPV_avr,t는 설비용량으로 정규화된 과거 3개년의 동일 특수일 t시의 BTM 태양광 발전량 추정치평균값을 의미한다. 만약 예측일이 맑은 날인 경우 예측일의 BTM 태양광 발전량이 과거의 BTM 태양광 발전량보다 많으므로 전력수요가 하향 조정되며, 예측일이 흐린 날인 경우 예측일의 BTM 태양광 발전량이 과거의 BTM 태양광 발전량보다 적으므로 전력수요가 상향 조정된다.where t is time,

is the predicted power demand at time t on a special day reflecting the influence of the BTM photovoltaic generator,

is the predicted power demand at time t on a special day reflecting the influence of temperature,

is the estimated BTM solar installed capacity on a special day,

is the estimated BTM solar power generation amount normalized to the installed capacity at special day t, and PPV _avr,t is the average value of the estimated BTM solar power generation amount normalized to the installed capacity at the same special day t for the past 3 years. If the forecast date is a sunny day, the amount of BTM solar power generation on the forecast day is greater than the amount of BTM solar power generation in the past, so the power demand is adjusted downward. Since it is less than the generation amount, the power demand is adjusted upward.

According to an embodiment of the present invention, the average absolute percentage error of the electric power demand forecast on a special day reflecting the influence of the BTM photovoltaic generator and the influence of the temperature at each hour is 2.38%, and the 2019 special The average absolute percentage error of daily electricity demand forecasting was 3.55%, indicating that the accuracy was improved.

In this way, by improving the accuracy of electricity demand forecasting on special days with a high error in year-round electricity demand forecasting, it is possible to achieve the effect of stably and efficiently operating the power system and the electricity market.

Exemplary methods of the present invention are presented as a series of operations for clarity of explanation, but this is not intended to limit the order in which steps are performed, and each step may be performed concurrently or in a different order, if desired. In order to implement the method according to the present disclosure, other steps may be included in addition to the exemplified steps, other steps may be included except for some steps, or additional other steps may be included except for some steps.

On the other hand, the above-described embodiments of the present invention can be written as a program that can be executed on a computer, and can be implemented in a general-purpose digital computer that operates the program using a computer-readable recording medium.

The computer-readable recording medium includes a magnetic storage medium (eg, ROM, floppy disk, hard disk, etc.) and an optical reading medium (eg, CD-ROM, DVD, etc.).

So far, the present invention has been looked at with respect to its preferred embodiments. Those skilled in the art to which the present invention pertains will be able to understand that the present invention can be implemented in a modified form without departing from the essential characteristics of the present invention. Therefore, the disclosed embodiments should be considered from an illustrative rather than a limiting point of view. The scope of the present invention is shown in the claims rather than the foregoing description, and all differences within the equivalent scope will be construed as being included in the present invention.

132: power demand forecasting unit
134: temperature correction unit
136: BTM solar power generation amount reflection part

Claims (10)

As a power demand prediction method of a power demand prediction device on a special day,
predicting power demand on a special day to be predicted by applying the power demand data of a past special day to a fuzzy linear regression analysis model;
determining whether temperature sensitivity is corrected based on the predicted temperature of the special day to be predicted;
Calculating the temperature sensitivity of the special day according to the determination result and correcting the electricity demand of the predicted special day; and
The step of predicting and reflecting the amount of BTM solar power generation to the power demand of the corrected special day
A method for predicting power demand on special days that includes.
According to claim 1,
In the step of determining whether to correct the temperature sensitivity,
A method for predicting power demand on a special day, characterized in that it is determined that temperature sensitivity correction is necessary when the predicted temperature of the special day to be predicted does not fall within the cooling and heating reference temperature of 18 ° C to 26 ° C.
According to claim 1,
Calculating the temperature sensitivity of the special day and correcting the electricity demand of the predicted special day,
Calculating basic electricity demand for special days;
normalizing power demand using the basic power demand;
Calculating temperature sensitivity for each time interval through regression analysis of normalized electricity demand and temperature; and
Correcting electricity demand on special days based on the calculated temperature sensitivity for each time interval
A method for predicting power demand on a special day, comprising:
According to claim 3,
In the step of calculating the temperature sensitivity for each time section,
Forecasting power demand on special days, characterized in that the temperature sensitivity is calculated by calculating the ratio of the difference between the average temperature difference between the past special day and the average temperature of 4 weekdays before the special day and the average power demand of 4 weekdays before the special day Way.
According to claim 3,
The step of correcting the power demand on the special day
According to Equation 1 below, a method for predicting power demand on a special day, characterized in that for correcting the power demand on a special day.
[Equation 1]

(In Equation 1, p means a time interval, m is the month where the special day is located, t is the time,

is the predicted power demand at special day t corrected by the temperature difference, L _t is the predicted power demand at special day t,

is the temperature sensitivity of the special day p section,

is the basic electricity demand for the period p in month m,

is the predicted temperature at time t on a special day,

is the average temperature at time t of the 4 weekdays prior to the special day,

is the average temperature at time t on the same special day in the past three years,

means the average temperature at time t of 4 weekdays prior to the same special day in the past 3 years.)
According to claim 1,
The step of predicting and reflecting the BTM solar power generation amount to the power demand of the corrected special day
According to Equation 2 below, the BTM solar power generation amount is reflected in the corrected power demand on the special day.
[Equation 2]

(In Equation 2, t is time,

is the predicted power demand at time t on a special day reflecting the influence of the BTM photovoltaic generator,

is the predicted power demand at time t on a special day reflecting the influence of temperature,

is the estimated BTM solar installed capacity on a special day,

is the estimated BTM solar power generation capacity at special day t, normalized to installed capacity,

represents the estimated average value of BTM solar power generation at time t on the same special day for the past three years, normalized to installed capacity.)
a memory for storing power demand data and a fuzzy linear regression analysis model of a past special day;
a power demand prediction unit for predicting power demand on a special day to be predicted by applying the power demand data of the past special day to a fuzzy linear regression analysis model;
a temperature correction unit that determines whether or not to correct temperature sensitivity based on the predicted temperature of the special day to be predicted, calculates the temperature sensitivity of the special day according to the determination result, and corrects electricity demand for the predicted special day; and
BTM solar power generation amount reflection unit that predicts and reflects the BTM solar power generation amount to the corrected power demand on special days
Power demand forecasting device for a special day comprising a.
According to claim 7,
The temperature correction unit,
An apparatus for predicting power demand on a special day, characterized in that it determines that temperature sensitivity correction is necessary when the forecast temperature of the special day to be predicted does not fall within the cooling and heating reference temperature of 18 ° C to 26 ° C.
According to claim 7,
The temperature correction unit calculates the basic power demand of a special day, normalizes the power demand using the calculated basic power demand, calculates the temperature sensitivity for each time section through regression analysis of the normalized power demand and temperature, and calculates the temperature sensitivity. An apparatus for predicting power demand on a special day, characterized in that it corrects the power demand on a special day based on the temperature sensitivity for each time section.
According to claim 9,
The temperature correction unit,
Electric power demand forecasting device, characterized in that for calculating the temperature sensitivity by calculating the ratio of the difference between the average temperature difference between the special day and the average temperature of the four weekdays before the special day and the average temperature of the four weekdays before the special day in the past.

Images