Power estimation device and power estimation method

Abstract

This invention provides a method for accurately estimating the power supply at each receiving point in a power distribution system in real time or near real time. [Solution] The power estimation device (10) comprises a historical data creation unit (11), a reference power calculation unit (12), and an estimation unit (13). The historical data creation unit creates historical data by linking monitor section power values, which represent the power of a predetermined monitor section within the power distribution system, with transformer unit power values, which represent the power under each transformer within the monitor section. The reference power calculation unit calculates the reference power, which represents the power of the monitor section at the power estimation time. The estimation unit identifies the monitor section power value with the highest similarity to the reference power in the historical data, and outputs the respective transformer unit power values ​​corresponding to the identified monitor section power value as estimated transformer unit power values ​​for each transformer at the power estimation time.

Description

Technical Field

[0001] The present invention relates to an apparatus and a method for estimating power for each transformer in a power distribution system.

Background Art

[0002] In general, electric power utilities manage the voltage at the power receiving points of the power distribution system within a certain range in order to maintain power quality. For this purpose, an apparatus for estimating the voltage of a high-voltage distribution line by performing a power flow calculation using the measurement values of a measuring device that measures the power amount / voltage of the high-voltage distribution line and a smart meter that measures the power amount of each customer has been proposed (for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, it is not easy to collect in real time the measurement values of a large number of smart meters provided for a large number of customers. Therefore, it is difficult for conventional techniques to accurately estimate the power at each power receiving point of the power distribution system in real time or almost in real time.

[0005] An object related to one aspect of the present invention is to provide a method for accurately estimating the power at each power receiving point of a power distribution system in real time or almost in real time.

Means for Solving the Problems

[0006] A power estimation device according to one aspect of the present invention estimates the power under a transformer in a power distribution system that includes a high-voltage distribution line and a transformer connected to the high-voltage distribution line. This power estimation device includes: a past data creation unit that creates past data in which a plurality of monitor section power values ​​representing the power of a predetermined monitor section in the power distribution system acquired at a plurality of different times in a first cycle and a plurality of transformer unit power values ​​representing the power under each of the plurality of transformers in the monitor section acquired at a plurality of different times in a second cycle, are associated with each other in a predetermined monitoring cycle; a reference power calculation unit that calculates a reference power representing the power of the monitor section at the power estimation time; and an estimation unit that identifies the monitor section power value with the highest similarity to the reference power in the past data and outputs each transformer unit power value corresponding to the identified monitor section power value as an estimated value of the transformer unit power value for each of the plurality of transformers at the power estimation time. [Effects of the Invention]

[0007] According to the above-described embodiment, the power at each receiving point in the power distribution system can be estimated accurately in real time or near real time. [Brief explanation of the drawing]

[0008] [Figure 1] This figure shows an example of a power distribution system according to an embodiment of the present invention. [Figure 2] This figure shows an example of the functional configuration of a power estimation device. [Figure 3] This flowchart shows an overview of the power estimation device's processing. [Figure 4] This figure shows an example of a sensor measurement database and a smart meter measurement database. [Figure 5] This flowchart shows an example of the process for creating historical data. [Figure 6] This figure shows an example of past data. [Figure 7] This flowchart shows an example of the process for calculating the reference power. [Figure 8]This flowchart shows an example of a process for estimating the power under each transformer. [Figure 9] This figure shows an example of historical data extracted for a reference power value. [Figure 10] This figure shows an example of historical data related to the first variation. [Figure 11] This figure shows the processing flow for power estimation related to the first variation. [Figure 12] This figure shows an example of historical data related to the second variation. [Figure 13] This diagram shows the processing flow for power estimation related to the second variation. [Figure 14] This diagram shows variations in the composition of past data. [Figure 15] This diagram illustrates a method for estimating the voltage distribution of a power distribution system. [Figure 16] This figure shows an example of the hardware configuration of a power estimation device. [Modes for carrying out the invention]

[0009] General power transmission and distribution operators are required by law to maintain the voltage at each receiving point in the distribution system within a certain range in order to maintain power quality. However, distribution systems may be connected not only to loads but also to distributed power sources such as solar power generators. Therefore, voltage maintenance is becoming increasingly stringent. Thus, voltage monitoring is becoming crucial for properly maintaining voltage.

[0010] To solve this problem, it is effective to grasp the power of each customer in a state close to real time for each pole-mounted transformer and perform voltage calculation (power flow calculation). In addition, by sequentially grasping the power of each customer for each pole-mounted transformer, it is possible to accurately determine the presence or absence of overload when performing system switching related to a failure, and an improvement in system operation can be expected. Regarding the power of each pole-mounted transformer, it is possible to calculate the power consumed and generated within the target section and then perform an approximate apportionment calculation using the contract capacity, power usage record, etc. to obtain it.

[0011] In recent years, smart meters have been introduced as electricity meters with communication functions. Smart meters measure the electricity consumption of each customer and convert the value into average power. Then, by aggregating the measurement data from each smart meter for each pole-mounted transformer, the power of each pole-mounted transformer is estimated. Furthermore, by performing a power flow calculation using the power of each pole-mounted transformer, the voltage at each power receiving node (for example, each pole-mounted transformer) of the high-voltage distribution line can be estimated.

[0012] FIG. 1 shows an example of a distribution system according to an embodiment of the present invention. In FIG. 1, the high-voltage distribution line is represented in a simulated single-phase manner.

[0013] In the example shown in FIG. 1, the distribution system includes a high-voltage distribution line 2 starting from a transformer with a load tap changer (LRT) 1. The transformer with a load tap changer 1 is provided, for example, in a distribution substation.

[0014] The high-voltage distribution line 2 is provided with a plurality of switches 3 (3A, 3B). The switch 3 has a function of interrupting or switching the distribution line as needed. In addition, the switch 3 includes a sensor 4. The sensor 4 can measure the voltage, current, and phase (i.e., power factor) in the high-voltage distribution line 2. In the following description, the area surrounded by two switches 3 may be referred to as an "interval". Also, the interval between the switch 3A and the switch 3B shown in FIG. 1 may be referred to as "interval AB".

[0015] Transformers 5 (5x, 5y) are connected to the high-voltage distribution line 2. Transformers 5 have the function of converting high-voltage power to a predetermined low-voltage power. The power stepped down by transformers 5 is then supplied to each consumer 7 via the low-voltage distribution line 6. In the example shown in Figure 1, power is supplied from transformer 5x to consumers 7a to 7b, and from transformer 5y to consumers 7c to 7d. Transformers 5 are, for example, pole-mounted transformers fixed to utility poles in overhead wiring lines. However, transformers 5 are not limited to pole-mounted transformers and may, for example, be installed underground. Although omitted in Figure 1, high-voltage consumers may be connected to the high-voltage distribution line 2 of the distribution system via transformers. Furthermore, termination circuits may be provided at the ends of the high-voltage distribution line 2.

[0016] The power distribution system is equipped with smart meters (SM) 8 that measure the power consumption of each customer 7. In the example shown in Figure 1, smart meters 8a to 8d are provided for each of the customers 7a to 7d. In this case, smart meters 8a to 8d can each measure the power consumption of customers 7a to 7d.

[0017] The power distribution system may be connected to distributed power sources such as solar power generators. Furthermore, the distributed power sources may be located under the transformer 5. In the example shown in Figure 1, customer 7a is equipped with a solar power generator (SP) 9. In this case, the amount of electricity measured by the smart meter 8a corresponds to the difference between the amount of electricity consumed by customer 7a and the amount of electricity generated by the solar power generator 9.

[0018] The power estimation device 10 estimates the power under each transformer 5. Here, the power estimation device 10 may estimate the active power and reactive power under each transformer 5, respectively. Furthermore, if a distributed power source (for example, a solar power generator) is connected under the transformer 5, the power estimation device 10 may estimate the power considering the amount of power generated by the distributed power source. In addition, the power estimation device 10 may determine the voltage distribution of the high-voltage distribution line 2 by performing power flow calculations using the power at each transformer 5.

[0019] Figure 2 shows an example of the functional configuration of a power estimation device 10 according to an embodiment of the present invention. As shown in Figure 2, the power estimation device 10 comprises a historical data creation unit 11, a reference power calculation unit 12, and an estimation unit 13. The power estimation device 10 may further include other functions not shown in Figure 2. The power estimation device 10 can also refer to an equipment information database 21, a sensor measurement value database 22, and a smart meter measurement value database 23. These databases 21 to 23 may be provided outside the power estimation device 10 or implemented inside the power estimation device 10.

[0020] The equipment information database 21 stores equipment information that represents the configuration of the power distribution system. The equipment information includes information indicating the location where each switch 3 is installed, information identifying each section, and information indicating the location where each transformer 5 is installed (i.e., information indicating which section each transformer 5 is installed in). The information identifying each section includes, for example, information identifying the switch 3 located at both ends of that section. The equipment information also includes information indicating the customers 7 and smart meters 8 connected under each transformer 5, and information indicating the location and capacity of the solar power generators 9.

[0021] The sensor measurement database 22 stores the measurement values ​​of the sensors 4 equipped with each switch 3 (3A, 3B) in a time series. The sensors 4 measure the power, current, and phase of the high-voltage distribution line 2 at predetermined intervals (e.g., every minute). The sensor measurement database 22 then stores the sensor values ​​collected at one-minute intervals over a predetermined period (e.g., one year) in a time series.

[0022] The smart meter measurement database 23 stores the measurement values ​​from each smart meter 8 (8a to 8d) installed under each transformer 5 in chronological order. The smart meter 8 measures the amount of electricity consumed by the corresponding customer at predetermined intervals (for example, every 30 minutes). The smart meter measurement database 23 then stores the amount of electricity consumed by each customer, collected at 30 intervals over a predetermined period (for example, one year), in chronological order.

[0023] The power estimation device 10 estimates the power under each transformer 5 (transformer unit power) in a distribution system equipped with a high-voltage distribution line 2 and transformers 5 (5x, 5y) connected to the high-voltage distribution line 2. The power under each transformer 5 represents the total power of the equipment connected under that transformer 5. The total power of the equipment connected under each transformer 5 corresponds to the sum of the power consumption of each customer 7 under that transformer 5. Furthermore, when a distributed power source is connected under each transformer 5, the total power of the equipment connected under each transformer 5 corresponds to the sum of the power consumption of each customer 7 under that transformer 5 minus the power generated by the distributed power source.

[0024] Figure 3 is a flowchart illustrating the processing overview of the power estimation device 10. The processing in this flowchart is performed by the historical data creation unit 11, the reference power calculation unit 12, and the estimation unit 13. The equipment information database 21 stores equipment information representing the configuration of the power distribution system. The sensor measurement value database 22 stores measurement values ​​obtained from the sensors 4 of each switch 3. The smart meter measurement value database 23 stores measurement values ​​obtained from each smart meter 8.

[0025] In S1, the historical data creation unit 11 creates historical data using the measurement values ​​stored in the sensor measurement value database 22 and the smart meter measurement value database 23. At this time, the historical data creation unit 11 creates historical data by referring to the equipment information stored in the equipment information database 21.

[0026] In the historical data, multiple monitor section power values ​​representing the power of a predetermined monitor section within the power distribution system, acquired at multiple different times in the first cycle, and multiple transformer unit power values ​​representing the power under each of the multiple transformers 5 within the monitor section, acquired at multiple different times in the second cycle, are associated with each other in a predetermined monitoring cycle. The first cycle is the measurement cycle of the sensors 4 provided by each switch 3 (3A, 3B), for example, "1 minute". The monitor section power values ​​are calculated based on the measurements taken by each sensor 4. The second cycle is the measurement cycle of each smart meter 8 (8a~8d), for example, "30 minutes". The transformer unit power values ​​are calculated for each transformer 5 based on the measurements taken by each smart meter 8. For example, when section AB shown in Figure 1 is a monitor section, the monitor section power values ​​are calculated based on the measurements taken by the sensors 4 provided by the switches 3A and 3B located at both ends of section AB. Furthermore, the transformer unit power values ​​are calculated for transformers 5x and 5y connected to high-voltage distribution line 2 in section AB. The monitoring period is not particularly limited, but for example, it is the same as the measurement period of each smart meter 8.

[0027] In S2, the reference power calculation unit 12 calculates a reference power representing the power of the monitoring interval at the power estimation time. Therefore, the reference power corresponds to the monitoring interval power value calculated based on the measurement value from the corresponding sensor 4 at the power estimation time.

[0028] In S3, the estimation unit 13 identifies the monitor section power value that has the highest similarity to the reference power in the past data. The estimation unit 13 then outputs the unit power value of each transformer corresponding to the identified monitor section power value in the past data as an estimated value of the unit power value of each transformer 5 within the monitor section.

[0029] Thus, the power estimation device 10 estimates the power under each transformer 5 by searching past data using the monitor section power calculated based on the measurement values ​​from the sensor 4 provided in the switch 3. However, since the measurement cycle of the smart meter 8 is long (for example, 30 minutes), it is difficult to obtain the power under each transformer 5 in real time using the measurement values ​​from the smart meter 8. In contrast, since the measurement cycle of the sensor 4 (for example, 1 minute) is short compared to the measurement cycle of the smart meter 8, according to the embodiment of the present invention, the power under each transformer 5 can be estimated in real time with high accuracy.

[0030] <Examples> In the following embodiment, in the power distribution system shown in Figure 1, section AB is assumed to be the monitoring section. Therefore, the power estimation device 10 estimates the power under each transformer 5x and 5y. Furthermore, the equipment information database 21 is assumed to have been created in advance. The sensor measurement value database 22 and the smart meter measurement value database 23 are as follows.

[0031] Figure 4A shows an example of a sensor measurement value database 22. In this embodiment, the sensor measurement value database 22 stores measurement values ​​(voltage, current, phase) acquired from sensors 4 installed in each switch 3 (3A, 3B) at a 1-minute interval over a one-year period. In the notation for each measurement value, the first characters "V", "I", and "θ" represent voltage, current, and phase, respectively. The second character identifies the switch 3 on which the measurement was taken, with "A" and "B" representing switch 3A and switch 3B, respectively. The following numbers represent the date and time the measurement was taken.

[0032] Figure 4B shows an example of a smart meter measurement database 23. In this embodiment, the smart meter measurement database 23 stores measurement values ​​acquired from each smart meter 8 (8a to 8d) at 30-minute intervals over a one-year period. In the notation for each measurement value, the first letter "P" represents the amount of electricity. The second letter identifies the smart meter 8 on which the measurement was taken, and "a" to "d" represent smart meters 8a to 8d, respectively. The following numbers represent the period during which the measurement was taken. For example, "01 / 01 00:30" represents the period from 0:00 to 0:30 on January 1st, and "01 / 01 01:00" represents the period from 0:30 to 1:00 on January 1st.

[0033] Figure 5 is a flowchart showing an example of the process for creating historical data. This flowchart corresponds to S1 shown in Figure 3 and is executed by the historical data creation unit 11 in response to instructions from the user or administrator of the power estimation device 10.

[0034] In S11, the historical data creation unit 11 calculates the power of the monitoring section based on the equipment information stored in the equipment information database 21 and the sensor measurement values ​​stored in the sensor measurement value database 22 shown in Figure 4A. At this time, the historical data creation unit 11 calculates the active power P and reactive power Q of the monitoring section.

[0035] Active power P and reactive power Q are calculated using the following formulas, respectively. P = V·I·cos θ Q = V·I·sin θ

[0036] Here, the active power P and reactive power Q detected by switch 3A at time t are denoted by PA(t) and QA(t), respectively. Also, the active power P and reactive power Q detected by switch 3B at time t are denoted by PB(t) and QB(t), respectively. In this case, the active power P_AB(t) and reactive power Q_AB(t) of the monitoring section AB at time t are approximated by the following equations. P_AB(t)=PA(t)-PB(t) Q_AB(t)=QA(t)-QB(t) The historical data creation unit 11 calculates the active power P_AB(t) and reactive power Q_AB(t) for each sampling time (from 0:01 on January 1st to 24:00 on December 31st) of the sensor measurement value database 22 shown in Figure 4A, for each monitoring interval AB.

[0037] In S12, the historical data creation unit 11 calculates the time average of the power in the monitoring section. Specifically, the power (active power and reactive power) in the monitoring section is averaged over the measurement cycle of the smart meter 8. Here, the power in the monitoring section is calculated at a 1-minute cycle. The measurement cycle of the smart meter 8 is 30 minutes, as shown in Figure 4B. In this case, the historical data creation unit 11 calculates the average of 30 consecutive power values. For example, by averaging the 30 active power values ​​and 30 reactive power values ​​obtained from 0:01 to 0:30 on January 1st, the average active power and average reactive power for the monitoring section during this time period can be obtained.

[0038] In S13, the historical data creation unit 11 converts the measured values ​​(electricity) of each smart meter 8 into average power. At this time, the amount of electricity measured by the smart meter 8 is converted into average power according to the measurement cycle of the smart meter 8. In this embodiment, since the measurement cycle of the smart meter 8 is 30 minutes, the average power is calculated by dividing the amount of electricity measured by the smart meter 8 by "30".

[0039] In S14, the historical data creation unit 11 aggregates the average power corresponding to the amount of electricity measured by each smart meter 8 for each transformer 5. For example, smart meters 8a and 8b are connected to transformer 5x. Here, the average power corresponding to smart meters 8a and 8b at time t is represented by pa(t) and pb(t), respectively. In this case, the average power ps_x(t) of transformer 5x at time t is expressed by the following formula. ps_x(t)=pa(t)+pb(t)

[0040] Similarly, if the average power corresponding to smart meters 8c and 8d at time t is denoted by pc(t) and pd(t), respectively, then the average power ps_y(t) of transformer 5y at time t is expressed by the following formula. ps_y(t)=pc(t)+pd(t)

[0041] In steps S15-S16, the historical data creation unit 11 links the average power of the monitoring section calculated in steps S11-S12 with the average power of each transformer 5 calculated in steps S13-S14. At this time, the average power of the monitoring section and the average power of each transformer 5 are linked for each time (or time period) in which averaging was performed. That is, for each time (or time period) in which averaging was performed, the average active power and average reactive power of the monitoring section, the average power of transformer 5x, and the average power of transformer 5y are linked to each other. In addition, a timestamp representing the time (or time period) in which averaging was performed is assigned to each linked dataset. This creates historical data.

[0042] Figure 6 shows an example of historical data created by the procedure shown in Figure 5. In this embodiment, the average active power and average reactive power of the monitoring section, the average power of transformer 5x, and the average power of transformer 5y are linked to each other for each measurement time period. Each measurement time period may be identified by a timestamp representing the start or end time of the reference monitoring time period. "P_AB" and "Q_AB" represent the average active power and average reactive power in section AB, respectively, calculated in S11 to S12 of Figure 5. "ps_x" and "ps_y" represent the average power of transformer 5x and transformer 5y, respectively, calculated in S13 to S14 of Figure 5. The value on the left in parentheses attached to each power value represents the date. The value on the right in parentheses attached to each power value (t1 to t48) identifies each time period when a day is divided into 48 time periods of 30 minutes each.

[0043] For example, the first record represents the average active power and average reactive power of the monitoring period, the average power of transformer 5x, and the average power of transformer 5y from 0:00 to 0:30 on January 1st. The second record represents the average active power and average reactive power of the monitoring period, the average power of transformer 5x, and the average power of transformer 5y from 0:30 to 1:00 on January 1st. The last record represents the average active power and average reactive power of the monitoring period, the average power of transformer 5x, and the average power of transformer 5y from 23:30 to 24:00 on December 31st.

[0044] Figure 7 is a flowchart showing an example of the process for calculating the reference power. This flowchart corresponds to S2 shown in Figure 3 and is executed by the reference power calculation unit 12 when the power estimation time arrives. The power estimation time is, for example, the current time and corresponds to the timing of acquiring the measurement value from the sensor 4 installed on each switch 3. In this embodiment, the sensor 4 outputs the measurement value at one-minute intervals. In this case, the power estimation time arrives every minute. That is, the process shown in the flowchart in Figure 7 is repeatedly executed at one-minute intervals.

[0045] In S21, the reference power calculation unit 12 calculates the power of the monitoring section based on the equipment information stored in the equipment information database 21 and the newly measured values ​​(voltage, current, phase) from each sensor 4. At this time, the reference power calculation unit 12 calculates the active power P and reactive power Q of the monitoring section. The method for calculating the active power P and reactive power Q is as described with reference to Figure 5. That is, the power (active power, reactive power) of the monitoring section is calculated by calculating the difference between the power values ​​(PA(t), QA(t)) obtained based on the measured values ​​at switch 3A and the power values ​​(PB(t), QB(t)) obtained based on the measured values ​​at switch 3B.

[0046] In S22, the reference power calculation unit 12 calculates the time average of the power in the monitoring section. Here, the flowchart shown in Figure 7 is executed repeatedly at a predetermined cycle (for example, 1 minute). The power values ​​of the monitoring section calculated at the predetermined cycle are stored in time series in a memory (not shown). Therefore, the reference power calculation unit 12 can calculate the time average of the power in the monitoring section by referring to this memory. As an example, averaging is performed at the measurement cycle of the smart meter 8. In this case, the average of the latest 30 sets of power values ​​(active power, reactive power), including the power value obtained based on the new measurement value (i.e., the measurement value at the time of arrival of power estimation), is calculated.

[0047] The reference power calculation unit 12 stores the power values ​​(active power and reactive power) calculated in S22 as "reference power". Note that the averaging time in S22 is not limited to the measurement cycle of the smart meter 8.

[0048] The reference power calculation unit 12 does not need to perform averaging in S22. However, if averaging in S22 is not performed, the reference power will contain "noise" if the power fluctuates instantaneously within the monitoring interval. Therefore, it is preferable for the reference power calculation unit 12 to perform averaging in S22. In the following description, "monitoring interval power (i.e., reference power) representing the power of the monitoring interval at the power estimation time" includes both the case in which averaging in S22 is performed and the case in which averaging in S22 is not performed.

[0049] Figure 8 is a flowchart showing an example of the process for estimating the power under each transformer 5. This flowchart corresponds to S3 shown in Figure 3 and is executed by the estimation unit 13 when the reference power calculation unit 12 calculates the reference power at the power estimation time.

[0050] In S31, the estimation unit 13 extracts records from past data for time periods corresponding to the power estimation time. Specifically, records for time periods containing the power estimation time are extracted from past data. For example, if the past data shown in Figure 6 has been created and the power estimation time is 0:15, then 365 records with a measurement time period of "0:00 to 0:30" are extracted, as shown in Figure 9.

[0051] In S32, the estimation unit 13 identifies the monitor interval power value that has the highest similarity to the reference power among multiple records extracted from past data. That is, the estimation unit 13 identifies the monitor interval power value that has the highest similarity to the reference power among multiple records corresponding to the time period including the power estimation time. When 365 records as shown in Figure 9 have been extracted, the estimation unit 13 compares the reference power with 365 sets of monitor interval power values.

[0052] For example, let's assume that the power estimation time (i.e., the current time) is time t1, and that multiple records shown in Figure 9 have been extracted. Also, let's assume that the active power and reactive power at the power estimation time are denoted as "P_AB(t1)" and "Q_AB(t1)", respectively. The estimation unit 13 then calculates the error e for each record using the following equation (1).

number

[0053] The error e(d) corresponds to the sum of the squares of the errors for active power and reactive power, as shown in equation (1). "d" identifies the date and represents a range from "1 (i.e., January 1st)" to "365 (i.e., December 31st)". The estimation unit 13 then identifies the record (i.e., date) in which the error e is smallest. This identifies the monitor interval power value that is most similar to the reference power.

[0054] In S33, the estimation unit 13 outputs the transformer unit power value corresponding to the monitor section power value identified in S32 as the estimated transformer unit power at the power estimation time. For example, suppose the monitor section power value with the highest similarity to the reference power (i.e., the monitor section power value that minimizes the error e calculated by equation (1)) is the power value for "January 2nd" in the past data. In this case, the estimation unit 13 outputs the transformer unit power for "January 2nd" in the past data (i.e., ps_x(2, t1) and ps_y(2, t1)) as the estimated transformer unit power at the power estimation time. That is, at the power estimation time, it is estimated that the power under transformer 5x is "ps_x(2, t1)" and the power under transformer 5y is "ps_y(2, t1)".

[0055] Thus, in this embodiment of the present invention, if the current monitoring section power is similar to the monitoring section power at a specific past time, the current transformer unit power is estimated by assuming that the current transformer unit power is also similar to the transformer unit power at that specific time. Here, the measurement cycle of the sensor 4 implemented on each switch 3 is short, and the monitoring section power value is obtained in near real time. In this case, even if the transformer unit power cannot be measured in real time, the transformer unit power can be estimated in near real time by using the monitoring section power value that can be measured in near real time.

[0056] The flowcharts shown in Figures 7 and 8 are executed, for example, each time a measurement value is acquired from the sensor 4 mounted on the switch 3. Here, the measurement cycle of the sensor 4 is, for example, 1 minute. The power estimation device 10 then calculates the power of the monitoring section each time a measurement value is acquired from the sensor 4 and estimates the power of the transformer unit. Therefore, the power under each transformer 5 can be estimated in near real time.

[0057] <First variation> As shown in Figure 1, a solar power generator (SP) 9 may be connected to the power distribution system. In this case, the power within the section will consist of load power resulting from the power consumed by consumers 7 and solar power generated by the solar power generator. Here, solar power fluctuates greatly depending on the weather, season, time of day, etc. Therefore, in the first variation, the power under each transformer 5 is estimated with load power and solar power separated.

[0058] The equipment information database 21 stores information representing the capacity of the solar power generators installed in each section. Furthermore, the efficiency of the solar power generators is assumed to be known; that is, the relationship between solar radiation and power generation is assumed to be known. Additionally, the power factor of the solar power generators during power generation is assumed to be known.

[0059] Past data may include solar radiation data, as shown in Figure 10A. Solar radiation is measured, for example, by solar radiation sensors installed within each section. Solar radiation sensors may be installed in each switch 3. In cases where multiple solar radiation sensors are installed within each section, the average of the measurements taken by the multiple solar radiation sensors may be calculated for each section.

[0060] Furthermore, as mentioned above, the efficiency of the solar power generator and the power factor during power generation are known. Therefore, by measuring the amount of solar radiation within the monitoring section, the active power and reactive power related to solar power generation within the monitoring section can be calculated, respectively. The power estimation device 10 can then separate the monitoring section power calculated based on the measurements of the sensor 4 into load power (monitoring section load power) caused by the power consumed by the consumer 7 and solar power generated by the solar power generator (monitoring section power generation). In this case, past data may be linked to the unit power values ​​of each transformer for the active power and reactive power related to the load, as well as the active power and reactive power related to solar power generation, as shown in Figure 10B. In Figure 10B, "P1" represents the active power related to the load, "Q1" represents the reactive power related to the load, "P2" represents the active power related to solar power generation, and "Q2" represents the reactive power related to solar power generation.

[0061] Figure 11 shows the processing flow for power estimation related to the first variation. Here, it is assumed that the historical data shown in Figure 10B has been created.

[0062] When the power estimation time (i.e., the current time) arrives, the power estimation device 10 calculates the power for the monitoring section (i.e., reference power) based on the measurement values ​​from the sensor 4. At this time, the power estimation device 10 calculates the active power and reactive power related to solar power generation based on the amount of solar radiation within the monitoring section. Then, the power estimation device 10 separates the power for the monitoring section (active power and reactive power) into load power (active power and reactive power) consumed by the load and generated power (active power and reactive power) generated by solar power generation.

[0063] Next, the power estimation device 10 compares the reference power at the power estimation time with the corresponding power data for each time period in the past data and calculates the similarity between them. Specifically, it identifies the monitor section power data for which the sum of the squared difference in active power related to the load, the squared difference in reactive power related to the load, the squared difference in active power related to solar power generation, and the squared difference in reactive power related to solar power generation is minimized. Then, the power estimation device 10 outputs the transformer unit power value corresponding to the identified monitor section power data as the estimated power value under each transformer 5 at the power estimation time.

[0064] Thus, in the first variation, even when solar power generators are connected to the power distribution system, the power under each transformer 5 can be estimated in near real time.

[0065] <Second variation> In the first variation, the power under each transformer 5 is estimated from the power of the monitoring section, taking solar power generation into consideration. In the second variation, the power related to the load and the power related to solar power generation are estimated for each transformer 5.

[0066] In the second variation, when creating historical data, the power related to solar power generation in the monitoring section is calculated based on the amount of solar radiation. Furthermore, the ratio of the capacity of the solar power generators connected under each transformer 5 is identified by referring to the equipment information. This allows the amount of solar power generated under each transformer 5 to be calculated. Then, as shown in Figure 12, historical data is created by linking the load power and solar power generation power within the monitoring section with the load power and solar power generation power of each transformer 5. Thus, in the historical data, each monitoring section power value includes the monitoring section load power value related to the load within the monitoring section and the monitoring section power generation value generated within the monitoring section, while the transformer unit power value includes the transformer unit load power value related to the load under the transformer and the transformer unit power generation value generated under the transformer. Note that in the example shown in Figure 12, although omitted for clarity, active power and reactive power are also linked as historical data in the second variation.

[0067] Figure 13 shows the processing flow for power estimation related to the second variation. Here, it is assumed that the historical data shown in Figure 12 has been created.

[0068] The process of comparing the reference power at the power estimation time (i.e., the current time) with the corresponding power data for each time period in the historical data and calculating their similarity is substantially the same in both the first and second variations. That is, in the historical data, the monitor section power data with the highest similarity to the monitor section power (i.e., reference power) at the power estimation time is identified. The power estimation device 10 outputs the transformer unit power value corresponding to the identified monitor section power data as the estimated power under each transformer 5 at the power estimation time. However, in the historical data of the second variation, as explained with reference to Figure 12, the load power and solar power within the monitor section are linked to the load power and solar power of each transformer 5. Therefore, in the second variation, the load power and solar power of each transformer 5 can be estimated individually in near real time in a power distribution system to which solar power generators are connected.

[0069] <Third variation> In the embodiment described above, the power for each transformer 5 is estimated using historical data from the same time period as the power estimation time. For example, in S31 of Figure 8, the estimation unit 13 extracts records from the historical data for the time period corresponding to the power estimation time and compares the reference power with the extracted historical data. This is because, even if the days are different, the power consumption trends of each consumer 7 are considered to be similar if they are in the same time period.

[0070] However, the electricity consumption trends of each customer 7 may shift forward or backward in time. Therefore, in the third variation, past data from a predetermined time range that includes the same time as the electricity estimation time is not referenced.

[0071] For example, let's assume that the historical data shown in Figure 6 has been created. Also, let's assume that the power estimation time is 0:45. In this case, in the example shown in Figure 8, historical data corresponding to "0:30 to 1:00" is extracted and compared with the reference power. In contrast, in the third variation, the time period to be compared with the reference power is expanded, and historical data corresponding to "0:00 to 0:30", "0:30 to 1:00", and "1:00 to 1:30" are extracted and compared with the reference power.

[0072] Thus, according to this third variation, the time range for searching for similar monitoring interval power values ​​is extended. Therefore, the power for each transformer 5 can be estimated based on historical data that is closer to the power state at the time of power estimation.

[0073] The power estimation device 10 may compare the reference power with past data for all time periods. However, considering that the power under each transformer 5 can change independently, it is preferable to perform power estimation on the premise that "even on different days, the power consumption trends of each consumer 7 are similar for the same time period." In other words, it is preferable for the power estimation device 10 to perform power estimation by referring to past data for the same time period as the power estimation time or for a predetermined time range that includes the same time as the power estimation time.

[0074] <Fourth variation> In the embodiment described above, the smart meter 8 measures the integrated value of active power. However, a smart meter installed for a high-voltage consumer may measure the integrated values ​​of active power and reactive power separately. Depending on the type of smart meter or data acquisition system, it may also measure voltage. In these cases, the reactive power / voltage ratio may be calculated for each transformer 5 (or for each interconnection point with the distribution system for high-voltage consumers) and used together with the active power for power estimation.

[0075] <Fifth variation> In the embodiment described above, power estimation is performed for the section enclosed by two switches (switches 3A and 3B in Figure 1), but power estimation may also be performed for the entire distribution line (high-voltage distribution line 2 in Figure 1). For example, the power of each transformer connected to the distribution line between the distribution point (LRT1 in Figure 1) and the termination circuit of the distribution line may be estimated by comparing the current power value with past power values.

[0076] <Sixth variation> In the above-described embodiment, historical data is created according to the measurement cycle of the smart meter 8. Specifically, the monitoring section power, calculated from the measurement values ​​of the sensor 4 obtained at short cycles (e.g., 1 minute), is averaged over the measurement cycle of the smart meter 8 (e.g., 30 minutes), thereby linking the average value of the monitoring section power with the transformer unit power value. In other words, the monitoring cycle (time resolution) of the historical data matches the measurement cycle of the smart meter 8. However, in the embodiment of the present invention, the monitoring cycle of the historical data does not need to match the measurement cycle of the smart meter 8.

[0077] In this embodiment, historical data is created according to the measurement cycle of the sensor 4. For example, in the case shown in Figure 14, the monitoring section power is calculated at 1-minute intervals from 0:00 to 0:30, and the transformer unit power is calculated at 30-minute intervals. In this case, the transformer unit power values ​​(transformer unit power interpolation values) for the period from 0:01 to 0:29 are generated at 1-minute intervals by interpolation based on the monitoring section power values ​​for the period from 0:01 to 0:29. At this time, it is preferable that the transformer unit power values ​​for the period from 0:01 to 0:29 are generated such that the ratio of the amount of energy under transformer 5x to the amount of energy under transformer 5y changes continuously from 0:00 to 0:30.

[0078] The power estimation device 10 estimates the transformer unit power by referring to the historical data created as described above. In other words, the transformer unit power is estimated by referring to historical data with fine time resolution. Therefore, it is expected that the accuracy of the transformer unit power estimation will be high.

[0079] <Seventh variation> The electricity consumption trends of customer 7 are thought to depend on the calendar. For example, electricity consumption trends on holidays (Saturdays, Sundays, public holidays, etc.) are thought to differ from those on weekdays. Therefore, it is preferable for the power estimation device 10 to perform power estimation using calendar information. That is, the estimation unit 13 extracts past data that has the same calendar information as the power estimation time from the past data created by the past data creation unit 11. Then, the estimation unit 13 estimates the power under each transformer 5 at the power estimation time by identifying the monitor section power value that has the highest similarity to the reference power in the extracted past data. For example, the power estimation device 10 may extract corresponding past data depending on whether the current time is a weekday or a holiday, and estimate the power under each transformer 5 by comparing the reference power with the extracted past data.

[0080] <8th variation> In accordance with laws and regulations, general power transmission and distribution operators manage the voltage at the receiving points of the distribution system within a certain range in order to maintain power quality. Therefore, general power transmission and distribution operators are required to monitor the voltage at the receiving points of the distribution system in near real time. To address this requirement, the power estimation device 10 is equipped with a function to calculate the voltage at each node of the distribution system using power flow calculation. Power flow calculation is a known technique that calculates the power flow through the transmission lines and the voltage and phase at each node when the power output of the generators and the power consumption of the loads in the distribution system are given.

[0081] In S41, as shown in Figure 15A, the power estimation device 10 estimates the power under each transformer 5 (i.e., the transformer unit power). The transformer unit power is estimated by the method described with reference to Figures 2 to 14. Then, in S42, the power estimation device 10 creates the voltage distribution of the distribution system by providing the power values ​​of each node (in this case, each transformer 5 connected to the high-voltage distribution line 2 of the distribution system) to the power flow calculation model.

[0082] Figure 15B schematically shows the voltage distribution of the power distribution system. In Figure 15B, RX represents the impedance of the high-voltage distribution line 2 between the distribution system's transmission point and transformer 5, or between adjacent transformers 5, and these are assumed to be known. Then, the power under each transformer 5 (i.e., the power per transformer) is estimated in near real time, and these values ​​are fed into the power flow calculation model to estimate the voltage at each node.

[0083] <Hardware Configuration> Figure 16 shows an example of the hardware configuration of power estimation device 0. Power estimation device 10 is implemented by a computer 100 which includes a processor 101, memory 102, storage device 103, input / output device 104, recording medium reader 105, and communication interface 106.

[0084] The processor 101 can execute various programs stored in the storage device 103. When the processor 101 executes the power estimation program according to an embodiment of the present invention, the functions of the historical data creation unit 11, the reference power calculation unit 12, and the estimation unit 13 shown in Figure 2 are provided. The power estimation program according to an embodiment of the present invention includes, for example, program code describing the procedure shown in Figure 3 (or Figures 5, 7, 8, and 15A). Memory 102 is used as a workspace for the processor 101. The storage device 103 stores the power estimation program and other programs. Furthermore, the equipment information database 21, the sensor measurement value database 22, and the smart meter measurement value database 23 may be built within the storage device 103. However, these databases 21-23 may be built in an external storage device accessible by the computer 100. In addition, historical data is stored, for example, in the storage device 103.

[0085] The input / output device 104 may include input devices such as a keyboard, mouse, touch panel, and microphone. The input / output device 104 may also include output devices such as a display device and speaker. The recording medium reader 105 can acquire data and information recorded on the recording medium 110. The recording medium 110 is a removable recording medium that can be attached to and detached from the computer 100. The recording medium 110 can be implemented, for example, by semiconductor memory, a medium from which signals can be read by optical action, or a medium from which signals can be read by magnetic action. The power estimation program may be provided to the computer 100 from the recording medium 110. The communication interface 106 provides the function of connecting to a network. When the power estimation program is stored on the program server 120, the computer 100 may acquire the power estimation program from the program server 120. [Explanation of symbols]

[0086] 2. High-voltage distribution lines 3(3A, 3B) Switch 4 sensors 5(5x, 5y) transformer 6. Low-voltage distribution lines 7(7a~7d) Consumer 8 (8a~8d) Smart meter 9. Solar power generator 10 Power estimation device 11. Historical Data Creation Department 12 Reference power calculation section 13 Estimation part

Claims

1. In a power distribution system comprising a high-voltage distribution line and a transformer connected to the high-voltage distribution line, a power estimation device for estimating the power under the transformer, A past data creation unit creates past data in which multiple monitor section power values ​​representing the power of a predetermined monitor section within the power distribution system, acquired at multiple different times in a first cycle, and multiple transformer unit power values ​​representing the power under each of the multiple transformers within the monitor section, acquired at multiple different times in a second cycle, are associated with each other over a predetermined monitoring cycle. A reference power calculation unit calculates a reference power representing the power of the monitoring interval at the power estimation time, An estimation unit identifies the monitor section power value with the highest similarity to the reference power in the aforementioned past data, and outputs the unit power value of each transformer corresponding to the identified monitor section power value as an estimated value of the unit power value of each of the multiple transformers at the power estimation time. A power estimation device equipped with the following features.

2. The second period is N (where N is an integer of 2 or more) times the first period. The aforementioned monitoring period is the same as the second period. The aforementioned historical data creation unit, Averaging is performed for each of the N consecutive monitor interval power values ​​to generate multiple average monitor interval power values. The historical data is created by correlating the multiple average monitoring interval power values ​​with the transformer unit power values ​​for each of the multiple transformers. The power estimation device according to feature 1.

3. The second period is longer than the first period. The monitoring period is the same as the first period. The aforementioned historical data creation unit, To ensure that the time resolution of the transformer unit power value matches the first period, a predetermined number of interpolated transformer unit power values ​​are generated between adjacent transformer unit power values ​​by interpolation using the plurality of monitor interval power values. The historical data is created by correlating the power values ​​for the monitoring interval with the transformer unit power values ​​and the transformer unit power interpolation values ​​for each of the multiple transformers. The power estimation device according to feature 1.

4. Each time a sensor measurement value is generated for calculating the power value of the monitoring section, The reference power calculation unit calculates the reference power, The estimation unit estimates the transformer unit power for each of the plurality of transformers according to the reference power. The power estimation device according to feature 1.

5. In the aforementioned historical data, each monitoring section power value includes a monitoring section load power value representing the power related to the load within the monitoring section and a monitoring section generated power value representing the power generated within the monitoring section. The reference power calculation unit calculates, as the reference power, a reference load power representing the power related to the load within the monitoring interval at the power estimation time and a reference generated power representing the power generated within the monitoring interval. The estimation unit estimates the transformer unit power for each of the multiple transformers by identifying the monitor section load power value and monitor section power power value that have the highest similarity to the reference load power and reference power generation power in the past data. The power estimation device according to feature 1.

6. In the aforementioned past data, each monitoring section power value includes a monitoring section load power value representing the power related to the load within the monitoring section and a monitoring section generated power value representing the power generated within the monitoring section, and the transformer unit power value includes a transformer unit load power value representing the power related to the load under the transformer and a transformer unit generated power value representing the power generated under the transformer. The reference power calculation unit calculates, as the reference power, a reference load power representing the power related to the load within the monitoring interval at the power estimation time and a reference generated power representing the power generated within the monitoring interval. The estimation unit estimates the transformer unit load power and transformer unit power for each of the multiple transformers by identifying the monitor section load power value and monitor section power power value that have the highest similarity to the reference load power and reference power generation power in the past data. The power estimation device according to feature 1.

7. The estimation unit, From the historical data created by the historical data creation unit, historical data within a predetermined range including the same time as the power estimation time is extracted. The system identifies the monitor section power value with the highest similarity to the reference power in the extracted historical data, and outputs the unit power value of each transformer corresponding to the identified monitor section power value as an estimated value of the unit power value of each of the multiple transformers at the power estimation time. The power estimation device according to feature 1.

8. The estimation unit, From the historical data created by the historical data creation unit, historical data having the same calendar information as the power estimation time is extracted. The system identifies the monitor section power value with the highest similarity to the reference power in the extracted historical data, and outputs the unit power value of each transformer corresponding to the identified monitor section power value as an estimated value of the unit power value of each of the multiple transformers at the power estimation time. The power estimation device according to feature 1.

9. A power estimation method for estimating the power under a transformer in a power distribution system that includes a high-voltage distribution line and a transformer connected to the high-voltage distribution line, The computer creates historical data in which multiple monitor section power values ​​representing the power of a predetermined monitor section within the power distribution system, acquired at multiple different times in a first cycle, and multiple transformer unit power values ​​representing the power under each of the multiple transformers within the monitor section, acquired at multiple different times in a second cycle, are associated with each other over a predetermined monitoring cycle. The computer calculates a reference power representing the power of the monitoring interval at the power estimation time, The computer identifies the monitor section power value in the past data that has the highest similarity to the reference power, and outputs the unit power value of each transformer corresponding to the identified monitor section power value as an estimated value of the unit power value of each of the multiple transformers at the power estimation time. A power estimation method characterized by the above.

Images