JP4531638B2 - Transformer load assumption method, transformer load assumption device - Google Patents

Description

  The present invention relates to a distribution load assumption method, and more particularly to a method and apparatus for assuming a load of a transformer in a distribution system.

  A power distribution system is a power facility composed of electrical lines and SW devices (hereinafter referred to as switches) between substations, power plants or transmission lines and power demand facilities, or between demand facilities. The future power consumption (load) is assumed along with the distribution facility plan. That is, when a future load becomes necessary in accordance with the above plan, a future load in the distribution system is assumed. And among the future loads that will be required in connection with the distribution system facility plan, the future daily load for each switch (section), feeder, and transformer unit will be accurately determined based on past performance data. It is a distribution load assumption problem.

  In this power distribution system, since many transformers are installed, in order to reduce the cost of the entire power distribution system, the transformer has an optimal capacity according to the daily load of the customer who is fed through the transformer. It is necessary to install.

  And the following patent document 1 describes the distribution system load estimation apparatus which calculates | requires the daily load of the transformer itself by assuming the daily load of the consumer connected to the transformer. In the load assumption method in the apparatus of Patent Document 1, with respect to a measurement consumer measuring a daily load, consumers having a strong correlation with a graph shape indicating a transition of the daily load are grouped, and the consumers belonging to the group The daily load transition graph that averages the daily load is set as the daily load of the group. In addition, representative attributes (contract type, floor area of residence, etc.) of measurement consumers belonging to each group are specified. And the daily load of the non-measurement consumer who has not measured the daily load is classified into any group based on the attribute of the non-measurement consumer.

In this way, if all the consumers are classified into any group, the daily load of all the customers fed through the transformer is synthesized by the daily load of each group to which each consumer belongs, The daily load of the transformer is being sought.
JP 2003-224931 A

  In the conventional daily load assumption method, the daily load of all customers is classified into any group. The daily load of the group is average, and there is always an error between the actual daily load and the average daily load of non-measurement consumers. And if the daily load of the customer connected to the transformer is synthesized, the error is accumulated, which is far from the actual daily load, and lacks accuracy. In order to increase the amount of data by installing a large number of measuring instruments or the like in order to increase accuracy, the measurement and installation costs a great deal.

The present invention has been created in view of the above problems. An object of the present invention is to provide a transformer load assumption method and a transformer load assumption device for assuming a load for each transformer at low cost and with high accuracy.
The basis of the present invention is specified by the following items (1) to (3).
(1) A transformer load assumption method for predicting a load for each transformer in a distribution system in which a measurement consumer having daily load measurement data and a non-measurement consumer having no daily load are mixed.
(2) Day of measurement customer assumed to be similar to daily load of forecasted day for each non-measurement consumer based on annual fluctuation of past monthly power consumption of both non-measurement consumer and measurement consumer Identify load measurement data.
(3) Demand for daily load based on the specified daily load measurement data for non-measurement consumers, and daily load based on daily load measurement data for measurement consumers, and demand for power supply for each transformer Total daily load on the house to determine the daily load per transformer.

In addition to the above items (1) to (3), the present invention includes any of the following items (11) to (16) or any combination of (11) to (16) as appropriate. It is good also as a distribution load assumption method provided with.
(11) Identify the maximum load in each transformer based on the daily load of the transformer.
(12) When specifying daily load measurement data of a measured consumer that is assumed to be similar for a non-measured consumer, the pattern shape of the annual fluctuation of the past monthly power consumption and the absolute power consumption In addition to obtaining the correlation, the daily load measurement data of the measurement consumer that is assumed to be similar to the daily load of the forecast target day for each non-measurement consumer is specified based on the correlation between the pattern shape and the absolute power consumption. .
(13) The daily load of the non-measurement consumer is specified based on the daily load measurement data of the measurement consumer having the same consumer type.
(14) When specifying daily load measurement data of a measured consumer that is assumed to be similar for a non-measurement consumer, the shape of the annual transition graph of the monthly power consumption for the daily load measurement data of the measurement consumer The shapes are associated in the order of strong correlation with each other, and the ranks are associated in order of strong correlation of the absolute power consumption of the monthly power consumption, and are associated with the daily load measurement data of each measurement consumer. And daily load measurement data with the lowest total sum of ranks for absolute power consumption and absolute power consumption.
(15) The daily load for a non-measurement consumer is the ratio of the monthly power consumption of both the non-measurement consumer and the measurement consumer assumed to be similar to this non-measurement consumer in the month to which the forecast date belongs. To be corrected accordingly.
(16) The daily load should be corrected according to the temperature data of the prediction target day.

The present invention extends to a device that assumes a distribution load based on the above transformer load assumption method, and includes the following items (21) to (24).
(21) A transformer load assumption device that predicts a load for each transformer in a wiring system in which a measurement customer having daily load measurement data and a non-measurement customer having no daily load are mixed.
(22) Provide a means for accessing a database for storing and managing information such as personal information, daily load measurement data, and monthly power consumption for each consumer.
(23) One measured demand that is assumed to be the most similar as the daily load of the forecasted day for each non-measured consumer based on the annual fluctuations in past monthly power consumption of both the non-measured consumer and the measured consumer Provide a means to identify the daily load measurement data of the house.
(24) Demand for daily load based on the specified daily load measurement data for non-measured consumers, and daily load based on the daily load measurement data for measured consumers, as well as demand to be the power supply destination for each transformer Provide a means to determine the daily load of each transformer by integrating the daily load of the house.

In addition to the above items (21) to (24), the power distribution load assumption device of the present invention appropriately applies any of the following items (31) to (36) or (31) to (36): You may provide the matter combined with.
(31) Provide a means for specifying the maximum load in each transformer based on the daily load of the transformer.
(32) The means for specifying the daily load measurement data obtains a correlation for each of the pattern shape of the annual fluctuation of the past monthly power consumption and the absolute power consumption, and predicts the target date for each non-measurement consumer. The daily load measurement data of the measurement consumer assumed to be similar to the daily load is specified based on the correlation between the pattern shape and the absolute power consumption.
(33) The means for specifying the daily load measurement data is to specify the daily load of each non-measurement consumer from the same type of measurement consumer as the type of each non-measurement consumer.
(34) The means for identifying the daily load measurement data associates a ranking with the daily load measurement data of the measurement consumer in order of increasing correlation of the pattern shape of the annual fluctuation of the monthly power consumption, and the monthly consumption Associating ranks in descending order of the correlation between the absolute power consumption and the power consumption, the total value of the ranks for each of the pattern shape and the absolute power consumption associated with the daily load measurement data of each measurement consumer is The lowest daily load measurement data should be specified.
(35) Percentage of daily power consumption of a non-measurement consumer for both the non-measurement consumer and the measurement consumer assumed to be similar to this non-measurement consumer in the month to which the forecast target day belongs Provide a means to correct according to.
(36) Provided with means for acquiring the temperature data of the daily load prediction target day, and means for correcting the daily load according to the temperature data of the prediction target day.

  According to the transformer load estimation method of the present invention, the future load of each transformer can be accurately estimated based on the monthly power consumption. Therefore, it is possible to reduce the cost for load assumption without additionally installing a device for measuring the daily load transition to a large number of consumers. In addition, since the load can be accurately assumed, the capacity of the transformer to be installed can be set appropriately, and an excessive investment by installing a transformer with a capacity larger than the actual or an accident due to a lack of capacity can be prevented.

=== Distribution system ===
FIG. 1 shows an example of a distribution system to which the transformer load assumption method of the present invention is applied. The power distribution system in this example is connected via a distribution line breaker (CB) 1 in a substation. A plurality of switches 2 are installed in the power distribution system, and a section 5 is defined between the switches 2. A large number of customers (facilities that consume power) 4 belong to each section 5. High voltage power from the substation outlet is stepped down by the transformer 3 in each section 5 and distributed to each main consumer (house 4a, store 4b, office 4c, etc.). In this way, a distribution network (feeder) 10 that supplies power from the CB 1 to each consumer 4 is configured.

  In CB 1, actual data obtained by measuring the load of the entire feeder 10 in increments of time is recorded, but in the switch 2 and the transformer 3, actual load data is not recorded. Further, the consumer 4 includes a measurement consumer having a load survey measuring instrument for measuring daily load and a non-measurement consumer having no load. As is well known, the load survey meter records a time transition of power consumption (load) in one day.

  The example of the daily load measurement data about various measurement consumers was shown in Drawing 2 (A)-(C). This figure shows the daily load measurement data of a general house (A), a consumer (B) who consumes midnight power, and a commercial facility (C). In addition, the electricity meter is installed in the non-measurement consumer, and the electric power company records the monthly power consumption in the non-measurement consumer by the meter reader reading the electricity meter every month. Fig. 3 shows an example of annual fluctuation of monthly power consumption.

=== Outline of Transformer Load Assumption Method ===
According to the transformer load assumption method of the present invention, the daily load data of all the consumers 4 are integrated for each transformer 3, and the daily load data of each transformer 3 is obtained. Further, the maximum load (peak load) in the transformer 3 is extracted based on the daily load data of the transformer 3. However, since there are measurement consumers and non-measurement consumers in the consumer 4, it is necessary to assume the daily load of the non-measurement consumer with reference to the daily load measurement data of the measurement consumer.

  The present inventors have empirically found that there is a high possibility that both daily loads are similar between consumers of the same type. The type is a distinction between facilities such as commercial facilities and houses, a distinction between industries such as chemical factories and garment factories, and received voltages such as 100 V and 6.6 kV. In addition, it is also found that the tendency of daily load (graph shape in FIG. 2) is more similar between consumers who have similar pattern shapes (graph shape in FIG. 3) of annual fluctuation of monthly power consumption. ing. However, for example, when looking at the annual trend of monthly power consumption between customers of the same industry, “Chemical Factory” as the same type, the absolute amount of power consumption differs between large factories and town factories. The daily load measurement data of the same industry cannot be adopted as the daily load of non-measurement consumers.

  In the present embodiment, the fluctuation pattern of the monthly power consumption for a certain non-measurement consumer for one year and the absolute value of the monthly power consumption are respectively represented by the power consumption of all measurement consumers of the same type as this non-measurement consumer. Compare the transition and absolute value. Based on the correlation between the fluctuation pattern of the monthly power consumption and the absolute value, the daily load measurement data of the measurement consumer that seems to be the most similar / approximate is identified, and the daily load measurement data is determined as non-measurement demand. It is adopted as the most suitable origin for assuming the daily load of the house. As a result, the monthly power consumption of the same type of consumers is compared, and based on the correlation between the annual fluctuation pattern shape of the power consumption and the absolute value of the monthly power consumption, Since the daily load measurement data that is the origin of the load is specified, it is possible to narrow down the comparison target and suppress the load and time required for the specifying process and extract more similar daily load measurement data. In this example, for the measured consumer, the daily load of the expected date is obtained based on the daily load measurement data of the same month / week / day of the week as the expected date of the previous year. The daily load of the assumed date is obtained based on the daily load measurement data of the same month / week / day of the past year of similar customers.

  When the daily load of the assumed date is obtained for all the consumers 4 in this way, the daily load of all the consumers 4 to which the user belongs belongs for each assumed date. As a result, the daily load for each of the transformers 3 and 5 is obtained. FIG. 4 shows an outline of daily load for each transformer. When the loads 20a to 20c of the consumers a to c to which electric power is supplied via a certain transformer 3 are summed, the daily load 30 of the transformer 3 is obtained. Thus, by assuming the daily load of the transformer 3 over a predetermined period such as a year, the peak load and the date and time can be acquired.

=== Load Assumption Device by Transformer ===
Here, a daily load of each transformer 3 on the distribution system is assumed based on the method of the present invention, and a device that assumes a peak load in each transformer 3 based on the assumed result (hereinafter, load assumed device) This is given as an embodiment of the present invention. FIG. 5 shows a functional block configuration of the load assumption device. The transformer load assumption device 40 is mainly composed of a computer with a database 42, and the database 42 includes personal information (name, address, telephone number, contract number, customer) about all consumers 4 belonging to the power distribution system. Type, belonging telephone pole number, presence / absence of daily load data, etc.), past monthly power consumption, and daily load measurement data, etc., are stored and managed if measured consumers. In this embodiment, the remote monitoring control device 60 is a computer different from the transformer load assumption device 40, and monitors the state of the distribution system such as the open / close state of the switch 2 and the power flow by an appropriate communication means. Or remotely controlling the opening and closing operation of the switch.

  The load assumption calculation control unit (hereinafter, calculation control unit) 43 is realized by executing a program installed in the transformer load assumption device 40, and is for assuming a daily load for each transformer 3 according to the method of the present invention. Perform various calculations. The main processing units included in the calculation control unit 43 include a customer type determination determination unit (hereinafter referred to as a type determination unit) 44, a correlation processing unit 45, a similar customer extraction unit 46, a load correction unit 47, and a processing result output. Part 48 is present. Then, the overall control unit 41 inputs information from the input device 49 attached to the transformer load assumption device 40 such as a keyboard, inputs information output from the remote monitoring control unit 60, or obtains stored information in the database 42. The input / acquisition information is provided to the calculation control unit 43, the processing result from the calculation control unit 43 is output to the output device 50 such as a display and a printer, the remote monitoring control device 60, or the database 42. An input / output interface function for storing data is provided.

=== Transformer daily load assumption processing ===
FIG. 6 shows the flow of the transformer peak load extraction process for obtaining the peak load in the assumed transformer and the date and time of the load by the transformer load assumption device 40 of this embodiment. In this example, a sample for extracting a peak load is acquired from a daily load on an assumed date included in a predetermined period in the future. For this purpose, first, a period of an assumed date to be a sample such as one year of the next year is designated (s1). Next, a daily load for all consumers connected to the assumed target transformer on an assumed date during the assumed period is assumed (s2), and belongs to the transformer extracted by this day assumed process (s2). The daily load of all customers is integrated (s3). As a result, the daily load (transformer daily load) on an expected date of the transformer is obtained. Then, the peak load and the time in the daily load on the assumed date are extracted (s4).

  In this way, the transformer daily load on each day of the assumed period is sequentially obtained (s5 → s6, s2 to s4). When the peak load and the time for all days in the assumed period are obtained, the maximum value among the peak loads for all days is set as the peak load of the transformer (s5 → s7).

  In the flow of the above transformer peak load extraction process, the daily load assumption process (s2) is the main point of the present invention. FIG. 7 shows the flow of the daily load assumption process (s2). Here, the number of non-measurement consumers connected to the assumed transformer is N, the measurement consumer in the entire distribution system is L, and a non-measurement consumer (assumed consumer) who is going to assume daily load data. ) Is i (i = 1, 2,..., N), the number of measurement consumers of the same type as i is M, and measurement consumers of the same type as i is k (k = 1, 2,...). , M). Therefore, M is a number that varies according to i. Further, a certain measurement consumer is assumed to be j (j = 1, 2,..., L).

  The overall control unit 41 reads the monthly power consumption data x (i) of the non-measurement consumer and the daily load measurement data y (j) of the measurement consumer from the database 42 (s11, s12). Furthermore, in this embodiment, since the daily load varies depending on the temperature, the temperature in the area where the assumed customer is located on the expected date, which is input via the input device, is acquired (s13). Data on temperature can be obtained from the Japan Meteorological Agency in time increments by region. Of course, the transformer load assumption device 40 may be configured to be able to communicate with an external computer that provides temperature data in the Japan Meteorological Agency or the like, and the temperature data may be obtained from this external computer.

  Once the necessary temperature data has been input, the daily load measurement data is integrated for each measured consumer by month to calculate the monthly power consumption (s14). In addition, since the measurement consumer also records the monthly power consumption by meter reading, the monthly power consumption acquired by the meter reading may be read from the database 42 as it is, as with non-measurement consumers. Then, the overall control unit 41 gives various data related to the acquired daily load assumption to the calculation control unit 43. The calculation control unit 43 extracts a non-measurement consumer (assumed customer) i that is a target of daily load (s15), and assumes a daily load for the non-measurement consumer by various processing units attached to the i. To do.

A specific daily load assumption process will be described below. First, the type determination unit 44 searches the database for a measured customer k of the same type as the assumed consumer (s16), and the correlation processing unit 45 determines the monthly power consumption of the assumed customer i and the measured customer k for the year. The shape correlation coefficient r1 of the graph indicating the trend of transition and the absolute amount correlation coefficient r2 of the monthly electric energy are calculated by the following equations (Equation 1) and (Equation 2), respectively (s17, s18). In Equations 1 and 2, S _i and S _k are standard deviations regarding monthly power consumption of each of the non-measurement customer i and the measurement customer k, and m is a month. F _im and f _iave are respectively the monthly power consumption and monthly average power consumption of non-measurement customer i in m months, and f _km and f _kave are the monthly consumptions of measurement customer k in m months, respectively. Power consumption and average monthly power consumption. The standard deviation S is obtained from (Equation 3).

  When the above two correlation coefficients r1 and r2 are calculated for all measured consumers (total number M) of the same type as i for a certain assumed customer i, the similar customer extraction unit 46 causes the daily load of this i And the most similar daily load measurement data and the measurement customer (similar customer) having the measurement data are specified (s19 → s21). In the present embodiment, a method called a “ranking method” is employed for identifying similar customers. In the ranking method, each of r1 and r2 is listed by assigning ranks to the corresponding measurement consumers in descending order, and the customer with the smallest total rank is set as a similar consumer. The daily load measurement data of the similar consumer is adopted as the calculation origin of the daily load (expected daily load) of the assumed consumer (s22 → s23). 8A and 8B show an outline of the ranking method. For an assumed consumer i, ranks are assigned to the measurement consumers a to e according to the magnitudes of the values r1 and r2 (FIG. 8A). And the sum total of the order | rank with respect to the value of r1 and r2 is calculated | required for every measurement consumer (FIG. 8B). In the example illustrated in FIG. 8, it is determined that the measured customer b is a similar customer, and an expected daily load is obtained based on the daily load measurement data of the similar customer b. In this way, by determining the correlation based on the total value of the two ranks, it is possible to treat the two correlations that should be equally important as well as different entities, and more appropriately measure the daily load. Data can be extracted. In this embodiment, when the values of r1 and r2 do not exceed a preset threshold value, the daily load measurement data of the measurement consumer of the same type as i is assumed as the daily load of the assumed consumer i. And the average daily load is adopted as the origin for obtaining the assumed daily load (s22 → s24).

  Since the similar consumer and the assumed consumer i have different power consumption, the daily load measurement data of the similar consumer cannot be directly adopted as the daily load of the assumed consumer i. Therefore, the load correction unit 47 corrects the daily load measurement data of the similar consumer based on the monthly power consumption of the assumed consumer i and the monthly power consumption of the similar consumer. Specifically, for example, for the same month as the month of the past expected date, obtain the monthly power consumption of the assumed customer and similar customers, find the ratio of both power consumption, and calculate this ratio of similar customers First, the daily load data is multiplied to perform first correction (s25). With this first correction, it is possible to accurately determine the power consumption of the assumed consumer in time increments.

  Furthermore, the load correction unit 47 corrects the daily load measurement data of similar customers based on the average temperature on the assumed day (s26). In the present embodiment, the past daily average temperature data is stored in the database 42, and the load correction unit 47 acquires the average temperature data for the past year and the daily load measurement data for the past year for the measurement consumer. To do. In addition, the measurement consumer in this case may be the same type of consumer as the assumed consumer, or may be a similar consumer specified earlier. Then, an average load is calculated for each day in the past year, and a correlation between the temperature and the average load is obtained. FIG. 9 shows the correlation. A function of a regression curve 71 indicating the correlation between the temperature and the average load is derived from the scatter diagram 70 in which the actual values of the average load with respect to the average temperature are plotted. The forecast average temperature is calculated by substituting the forecast average temperature on the expected function into the derived function. Then, a ratio between the average load with respect to the air temperature and the average daily load obtained by the first correction is obtained, and this ratio is further multiplied by the daily load obtained by the first correction. As a result, it is possible to finally obtain the daily load of the assumed customer with higher accuracy considering the temperature of the expected date.

  If the daily load for all non-measurement consumers to be assumed is obtained in this way, the calculation control unit 43 proceeds to the process (s3) for integrating the daily load for the consumer by transformer. . The processing result output unit outputs various information required in the process of the above-described maximum load extraction process for each transformer and daily load assumption process to an output device 50 such as a display or a printer, Or output as stored data. The output information and the presentation format of the information include, for example, a graph that displays the daily load on the expected date of the entire feeder, or a user input that selects a consumer or a transformer. For example, an expected day load graph for a house or a transformer, a peak value of the load, and the date and time can be output as appropriate.

It is the schematic of the distribution system used as the assumption object of the transformer load based on the transformer load assumption method in the Example of this invention. It is the schematic about the daily load of the various consumers who belong to the said power distribution system. It is a graph which shows annual transition of monthly power consumption in a certain consumer. It is a figure explaining the outline about the transformer load assumption method in a present Example. It is a functional block diagram of the transformer load assumption apparatus which estimates the daily load of each transformer on a distribution system based on the method of a present Example. It is a flowchart of the transformer peak load extraction process in the said load assumption apparatus. It is a flowchart of the daily load assumption process included in the said transformer peak load extraction process. It is explanatory drawing of the similar consumer extraction process included in the said load assumption process. It is a correlation diagram of temperature and average load.

Explanation of symbols

1 Circuit breaker for distribution line (CB)
DESCRIPTION OF SYMBOLS 2 Switch 3 Transformer 4, 4a-4c Consumer 10 Feeder 40 Transformer load assumption apparatus 41 Overall control part 42 Database 43 Load assumption calculation control part 60 Remote monitoring control apparatus

Claims (14)

In a distribution system where measurement customers with daily load measurement data and non-measurement customers without it,
Based on annual fluctuations in past monthly power consumption of both non-measurement consumers and measurement consumers, identify daily load measurement data of measurement consumers that are assumed to be similar for each non-measurement consumer,
For non-measurement consumers, the daily load based on the specified daily load measurement data is obtained, and for the measurement consumer, the daily load based on the daily load measurement data is obtained, and for each consumer who is a power supply destination for each transformer. Total daily load to determine the daily load per transformer,
Transformer load assumption method characterized by this.   The transformer load assumption method according to claim 1, wherein a maximum load in each transformer is specified based on the daily load of the transformer.   In Claim 1 or 2, when specifying daily load measurement data of a measurement consumer assumed to be similar for a certain non-measurement consumer, the pattern shape and absolute power consumption of the annual fluctuation of the past monthly power consumption The daily load measurement data of the measured consumer that is assumed to be similar to the daily load of the prediction target day for each non-measured consumer is obtained based on the correlation between the pattern shape and the absolute power consumption. Transformer load assumption method characterized by   4. The transformer load assumption method according to claim 1, wherein the daily load of the non-measurement consumer is specified based on the daily load measurement data of the measurement consumer having the same consumer type. In any one of Claims 1-4, when specifying the daily load measurement data of one measurement consumer assumed to be the most similar about a certain non-measurement consumer,
Corresponding ranks in order of increasing correlation of the pattern shape of the annual fluctuation of the monthly power consumption to the daily load measurement data of the measurement consumer, and ranking in order of strong correlation of the absolute power consumption of the monthly power consumption , And
The daily load measurement data that is associated with the daily load measurement data of each measurement consumer and has the lowest total value of the ranks for each of the pattern shape and the absolute power consumption is targeted for identification.
Transformer load assumption method characterized by this.   In any one of Claims 1-5, the daily load about a certain non-measurement consumer is both the said non-measurement consumer in the month to which a prediction object day belongs, and the measurement consumer assumed to be similar to this non-measurement consumer. A method for assuming a transformer load, which is corrected according to the ratio of monthly power consumption.   The transformer load assumption method according to claim 1, wherein the daily load is corrected according to temperature data on a prediction target day. Transformer load assumption device that predicts the load for each transformer in a wiring system in which measurement customers with daily load measurement data and non-measurement consumers who do not have
Means for accessing a database for storing and managing information such as personal information, daily load measurement data, and monthly power consumption for each consumer;
Daily load measurement data of measured consumers that are assumed to be similar to the daily load of the forecasted day for each non-measured consumer, based on the annual fluctuations in past monthly power consumption of both the non-measured consumer and the measured consumer A means of identifying
For non-measurement consumers, the daily load based on the specified daily load measurement data is obtained, and for the measurement consumer, the daily load based on the daily load measurement data is obtained, and for each consumer who is a power supply destination for each transformer. A means to determine the daily load by transformer by integrating the daily load;
A transformer load assumption device characterized by comprising:   9. The transformer load assumption device according to claim 8, further comprising means for specifying a maximum load in each transformer based on the daily load of the transformer.   In Claim 8 or 9, the means for specifying the daily load measurement data obtains a correlation for each of the pattern shape of the annual fluctuation of the past monthly power consumption and the absolute power consumption, and for each non-measurement consumer. Power supply for each transformer, characterized by specifying daily load measurement data of a measurement consumer that is assumed to be similar to the daily load of the forecast target day based on the correlation between the pattern shape and the absolute power consumption Transformer load assumption device that calculates the daily load for each transformer by integrating the daily load for the previous customers.   In any one of Claims 8-10, the means to specify the said daily load measurement data specifies the daily load of each non-measurement consumer from the measurement consumer of the same classification as the classification of each non-measurement consumer. A characteristic transformer load assumption device.   The means for specifying daily load measurement data according to any one of claims 8 to 11 corresponds to the daily load measurement data of a measurement consumer in order of decreasing correlation of the pattern shape of the annual fluctuation of the monthly power consumption. In addition, each of the pattern shape and the absolute power consumption associated with the daily load measurement data of each measurement consumer is associated with the rank in order of strong correlation of the absolute power consumption of the monthly power consumption. A transformer load assumption device, characterized in that the daily load measurement data having the lowest total ranking value is specified.   In any one of Claims 8-12, both the measurement consumer assumed that the daily load about a certain non-measurement consumer is similar to the said non-measurement consumer in the month to which a prediction object day belongs, and this non-measurement consumer. The transformer load assumption apparatus provided with the means to correct | amend according to the ratio of monthly power consumption. The method according to any one of claims 8 to 13, comprising means for acquiring temperature data of a daily load prediction target day, and means for correcting the daily load according to the temperature data of the prediction target day. Transformer load assumption device.

Images