JP2022084398A - Solar power generation output estimation device, solar power generation output estimation method, and solar power generation output estimation program - Google Patents

Abstract

To provide a solar power generation output estimation device capable of estimating the solar power generation output of a solar power generation facility without newly installing many solar radiation meters or measuring instruments having sufficiently high time resolution.SOLUTION: In a solar power generation output estimation system 10, a solar power generation output estimation device 100 includes: an estimation unit 150 configured to estimate a second solar power generation output that is a power generation output of a second solar power generation facility installed within a predetermined distance from a first solar power generation facility, using a first solar power generation output that is a power generation output of the first solar power generation facility and a residual demand that is apparent power consumption of a consumer who has installed the second solar power generation facility; and an estimation accuracy verification unit 140 configured to determine whether or not estimation accuracy of the second solar power generation output satisfies a specified accuracy using the power generation output of a third solar power generation facility installed within a predetermined distance from the first solar power generation facility and a residual demand that is apparent power consumption of a consumer who has installed the third solar power generation facility.SELECTED DRAWING: Figure 2

Description

The present disclosure relates to a photovoltaic power generation output estimation device for estimating the power output of a photovoltaic power generation facility, a photovoltaic power generation output estimation method, and a photovoltaic power generation output estimation program.

In recent years, the importance of expanding the use of renewable energy has increased, and distributed power sources such as photovoltaic power generation facilities have been installed to supply power to power transmission systems or distribution systems (hereinafter referred to as power systems). The number of consumers is increasing. On the other hand, although electric power companies that operate electric power systems know the power generation output of some photovoltaic power generation facilities, they cannot grasp the power generation output of many photovoltaic power generation facilities. Therefore, a device for estimating the power generation output of the photovoltaic power generation equipment installed by each consumer has been proposed (see, for example, Patent Document 1). The above-mentioned photovoltaic power generation equipment that can grasp the power generation output is the photovoltaic power generation equipment that is subject to the contract of the total purchase system (hereinafter referred to as the total purchase contract), and the photovoltaic power generation equipment that cannot grasp the power generation output is , It is a photovoltaic power generation facility that is subject to the surplus power purchase system contract (hereinafter referred to as the surplus purchase contract). This is a smart meter that measures the amount of power generated by the photovoltaic power generation equipment and a smart that measures the power consumption of the load inside the customer for consumers who own the photovoltaic power generation equipment that is the subject of the total purchase contract. While the meters are installed individually, for consumers who own the photovoltaic power generation equipment that is the subject of the surplus purchase contract, the total value of the power generation amount of the solar power generation equipment and the power consumption of the load is calculated. This is because only a smart meter for measuring is installed.

Patent Document 1 is a photovoltaic power generation output estimation device that estimates a photovoltaic power generation output, which is a photovoltaic power generation output at a predetermined time point of a photovoltaic power generation facility that supplies power to a load connected to a power system, and is a photovoltaic power generation output estimation device at a predetermined time point. In the previous period, when the south-middle altitude of the sun is within a predetermined range from the south-middle altitude at a predetermined point, the photovoltaic power output is used by using the solar radiation intensity at a predetermined point and the active power of the power system. A photovoltaic power generation output estimation device is disclosed, which comprises a power generation output estimation unit for estimating. This photovoltaic power generation output estimation device uses the solar radiation intensity measured by a solar radiation meter installed around the substation and the active power measured by a measuring instrument installed in the substation, etc., to connect the sun to the power system. Estimate the power output of the photovoltaic power generation facility.

Japanese Unexamined Patent Publication No. 2016-139270

If a large number of photovoltaic power generation facilities are introduced in a situation where electric power companies and the like cannot grasp the power generation output of many photovoltaic power generation facilities, various problems will occur in the operation of the power system. At electric power companies, etc., at substations under their jurisdiction, the power consumption of the apparent power system load (hereinafter referred to as "apparent load") that takes into account the power output of the photovoltaic power generation equipment is measured with a measuring instrument or the like. Although it can be measured, the total power output of all photovoltaic power generation facilities connected to the power system is unknown, so the power consumption of the actual load (hereinafter referred to as "actual load") can be accurately grasped. Can not do it. For this reason, when a large number of photovoltaic power generation facilities are introduced, there is an increased risk in supply and demand control that the prediction error of the actual load power consumption, which is predicted using multiple regression analysis, becomes large. There is an increased risk in system control that the recovery operation after a system accident will be hindered.

Therefore, it is necessary to accurately estimate the power generation output of the photovoltaic power generation equipment. However, according to the photovoltaic power generation output estimation device of Patent Document 1, it is necessary to install many pyranometers in order to measure the solar radiation intensity, and the measurement and recording have a sufficiently high time resolution, that is, a sufficiently high sampling cycle. It becomes necessary to install a measuring instrument that can be used. Further, since the estimation is performed in units of substations and the like, there is also a problem that the estimation accuracy decreases as the installed amount of the photovoltaic power generation equipment connected to the target power system becomes smaller than the actual load. Therefore, in order to estimate the power generation output of the photovoltaic power generation equipment connected to the power system with high accuracy, there are technical and cost problems.

This disclosure is made in view of the above, and estimates the photovoltaic output of a photovoltaic power generation facility without newly installing many pyranometers and measuring instruments with sufficiently high time resolution. The purpose is to obtain a photovoltaic power generation output estimation device that can be used.

In order to solve the above-mentioned problems and achieve the object, the photovoltaic power generation output estimation device according to the present disclosure is a second photovoltaic power generation facility installed within a predetermined distance from the first photovoltaic power generation facility. The second photovoltaic power generation output, which is the power generation output, is the apparent power consumption of the consumer who has installed the first photovoltaic power generation output, which is the power generation output of the first photovoltaic power generation facility, and the second photovoltaic power generation facility. It is equipped with an estimation unit that estimates using the residual demand. The photovoltaic power generation output estimation device is further used for the power output of the third photovoltaic power generation facility installed within a predetermined distance from the first photovoltaic power generation facility and for the consumer in which the third photovoltaic power generation facility is installed. It is provided with an estimation accuracy verification unit for determining whether or not the estimation accuracy of the second photovoltaic power generation output satisfies the predetermined accuracy by using the residual demand which is the apparent power consumption.

According to the present disclosure, it is possible to estimate the photovoltaic power generation output of a photovoltaic power generation facility without newly installing many pyranometers and measuring instruments having sufficiently high time resolution.

The figure which shows the configuration example of the photovoltaic power generation output estimation system which concerns on Embodiment 1. Block diagram showing a functional configuration example of the photovoltaic power generation output estimation device of the first embodiment The figure which shows the configuration example of the computer system which realizes the photovoltaic power generation output estimation apparatus of Embodiment 1. A flowchart showing an example of an estimation processing procedure of the coefficient α by the photovoltaic power generation output estimation device of the first embodiment. The figure which shows typically the geographical position of the photovoltaic power generation facility of each consumer shown by the place information data of Embodiment 1. The figure which shows an example of the data in the calculated data stored in the storage part of Embodiment 1. The figure which shows an example of the coefficient α for each combination of the photovoltaic power generation facilities stored in the storage part of Embodiment 1. A flowchart showing an example of a procedure for calculating the first covariance by the first covariance calculation unit of the first embodiment. A flowchart showing an example of a procedure for calculating the second covariance by the second covariance calculation unit of the first embodiment. A flowchart showing an example of the calculation processing procedure of the coefficient α of the first embodiment. A flowchart showing an example of the procedure for estimating the photovoltaic power generation output of the first embodiment. The figure which shows an example of the 1st solar power generation facility and the 2nd solar power generation facility when the solar power generation facility of a plurality of consumers is estimated. A diagram showing an example of the coefficient α when the estimation target is the photovoltaic power generation equipment of multiple consumers. The figure which shows an example of the verification result of the power generation output estimation method of Embodiment 1. The figure which shows an example of the verification result of the estimation method of the power generation output of Embodiment 1 when the estimation accuracy is not verified. FIG. 14 is a diagram showing an example of the verification result on the day before the day when the verification result is shown. FIG. 15 is a diagram showing an example of the verification result on the day before the day when the verification result is shown. A diagram showing an example of a scatter diagram of | λ-1 | and the estimation error of the second photovoltaic power generation output. Schematic diagram showing an example of the order of estimation of power generation output A flowchart showing an example of the calculation processing procedure of the coefficient α of the second embodiment. A flowchart showing an example of an estimation processing procedure of the coefficient α by the photovoltaic power generation output estimation device of the third embodiment.

Hereinafter, the photovoltaic power generation output estimation device, the photovoltaic power generation output estimation method, and the photovoltaic power generation output estimation program according to the embodiment will be described in detail based on the drawings.

Embodiment 1.
FIG. 1 is a diagram showing a configuration example of the photovoltaic power generation output estimation system according to the first embodiment. The photovoltaic power generation output estimation system 10 of the present embodiment has a configuration in which the photovoltaic power generation output estimation device 100 and various facilities installed under a plurality of consumers are connected via a communication network 50. be. In FIG. 1, the solid line represents the flow of electric power, and the broken line represents the flow of information.

In FIG. 1, as a plurality of consumers, consumer 200-1, consumer 200-2, consumer 200-3, consumer 200-4, ..., consumer 200-m, ..., consumer 200 -N is illustrated. A load 201 and a photovoltaic power generation facility (abbreviated as PV (PhotoVoltaic) in the figure) 202 are connected to each of the consumers 200-1 to 200-n. In the example shown in FIG. 1, n is an integer of 6 or more and m is an integer of 5 or more and less than n, but the values of n and m are not limited to the example shown in FIG. Further, the load 201 of each consumer 200-1 to 200-n indicates one or more facilities that consume electric power, and in general, the specific equipment constituting the load 201 is all consumers 200-1. ~ 200-n is not the same. Similarly, the photovoltaic power generation facility 202 is not the same for all consumers 200-1 to 200-n.

The photovoltaic power generation facility 202 provided in the consumer 200-1 is a photovoltaic power generation facility subject to a purchase contract in its entirety. The photovoltaic power generation equipment 202 installed in the consumers 200-2 to 200-n is a photovoltaic power generation facility subject to a surplus purchase contract.

The photovoltaic power generation facility 202 is, for example, a photovoltaic power generation facility installed on the roof of a building owned by consumers 200-1 to 200-n, and is not only a small-scale solar power generation facility but also a so-called mega. Includes large-scale solar power plants such as solar. The photovoltaic power generation facility 202 is connected to the electric power system 20, and supplies the generated electric power to the electric power system 20. Further, as shown in FIG. 1, a load 201 of each consumer 200-1 to 200-n is connected to the power system 20. The load 201 receives power from the power system 20 and consumes power.

Consumers 200-1 to 200-n provided with the load 201 and the photovoltaic power generation facility 202 receive power from the power system 20 and supply the power generated by the photovoltaic power generation facility 202 to the power system 20. Will be done. Therefore, the apparent power consumption of each consumer 200-1 to 200-n as seen from the power system 20 depends on both the actual power consumption of the load 201 and the power generation output of the photovoltaic power generation facility 202. become. When the power generation output of the photovoltaic power generation facility 202 exceeds the power consumption of the load 201, the power generation output of the consumer is supplied to the power system 20 from the viewpoint of the power system 20. The power consumption becomes a negative value. In the following, the apparent power consumption of each consumer 200-1 to 200-n as seen from the power system 20 will be referred to as residual demand.

A smart meter (abbreviated as SM (Smart Meter) in the figure) 205 is installed in the consumers 200-2 to 200-n who have concluded a surplus purchase contract. Assuming that i is an arbitrary integer from 2 to n, the smart meter 205 of the consumer 200-i has a power output _PPVi (t) of the photovoltaic power generation facility 202 of the consumer 200-i when the time is t. And the power consumption _PLi (t) by the load 201 are added together to measure the total power. That is, the electric power measured by the smart meter 205 is the residual demand described above, which is the amount obtained by subtracting the power generation output of the photovoltaic power generation facility 202 from the power consumption of the load 201 for each consumer 200-2 to 200-n. .. The measurement data of the residual demand measured by the smart meter 205 is sent to the photovoltaic power generation output estimation device 100 via the communication network 25 and the communication network 50.

For example, in the consumer 200-2, the smart meter 205 measures and records the total value of the power generation output _PPV2 (t) of the photovoltaic power generation facility 202 and the power consumption _PL2 (t) of the load 201 as the residual demand. .. Then, the smart meter 205 transmits the measurement data of the residual demand of the consumer 200-2 to the photovoltaic power generation output estimation device 100 via the communication network 25 and the communication network 50. In this way, the photovoltaic power generation output estimation device 100 is the residual demand, which is the measured value of the smart meter 205 installed in the consumers 200-2 to 200-n who own the photovoltaic power generation equipment 202 subject to the surplus purchase contract. Acquire measurement data.

Further, the consumer 200-3 is provided with a measuring instrument 206 for measuring the power generation output _PPV3 (t) of the photovoltaic power generation facility 202 provided in the consumer 200-3. The measuring instrument 206 may be a part of a device such as a power conditioner for controlling the photovoltaic power generation facility 202, or may be an individually provided measuring instrument. The measuring instrument 206 sends the measured data of the measured power generation output P _PV3 (t) to the photovoltaic power generation output estimation device 100 via the communication network 25 and the communication network 50. Here, an example in which the measuring instrument 206 is provided in the consumer 200-3 will be described, but among the consumers 200-2 to 200-n, the number of consumers in which the measuring instrument 206 is provided may be 0. However, there may be more than one.

A smart meter 203 and a smart meter 204 are installed in the customer 200-1 who has concluded a purchase contract for all the products. The smart meter 203 measures and records the power consumption _PL1 (t) of the load 201 of the consumer 200-1. The smart meter 204 measures and records the power output P _PV1 (t) of the photovoltaic power generation facility 202 of the consumer 200-1. The measurement data of the power generation output of the photovoltaic power generation facility 202 of the consumer 200-1 measured by the smart meter 204 is sent to the photovoltaic power generation output estimation device 100 via the communication network 25 and the communication network 50. In this way, the photovoltaic power generation output estimation device 100 acquires the measurement data of the power generation output _PPV1 (t) of the photovoltaic power generation facility 202 of the consumer 200-1. The measurement data of the smart meter 203 is also sent to the photovoltaic power generation output estimation device 100 via the communication network 25 and the communication network 50. It should be noted that at least one of the smart meter 203 and the smart meter 204, or a device (not shown) instead of those devices, is apparently the sum of the measurement data of the smart meter 203 and the measurement data of the smart meter 204. The power consumption, that is, the residual demand may be measured (or calculated), and the measured (or calculated) residual demand may be transmitted to the photovoltaic power generation output estimation device 100.

As described above, among the consumers 200-1 to 200-n, for the consumers 200-1 and 200-3, the measurement data of the power generation output of the photovoltaic power generation facility 202 is obtained by the smart meter 204 or the measuring instrument 206. be able to. Since the smart meter 204 can also be considered to be a part of the measuring instrument, both the consumer 200-1 and the consumer 200-3 are consumers who can acquire the measurement data of the power generation output of the photovoltaic power generation facility 202. be. On the other hand, among the consumers 200-1 to 200-n, the consumers who do not have either the smart meter 204 or the measuring instrument 206 are measured for the residual demand combined with the power consumption, but the photovoltaic power generation is performed. Since the power generation output of the equipment 202 is not directly measured, it is difficult to accurately obtain the power generation output. In the present embodiment, the photovoltaic power generation output estimation device 100 uses at least one of the consumer 200-1 and the consumer 200-3, and the smart meter 204 and the measuring instrument 206 are not provided in the consumer. The method of accurately estimating the power generation output of the above will be described. Hereinafter, the photovoltaic power generation equipment of consumers 200-1, 200-3 used as a reference in the estimation of the power generation output is also referred to as the first solar power generation equipment, and the solar power generation equipment for which the power generation output is estimated is the second solar power generation equipment. Also called a photovoltaic power generation facility.

The photovoltaic power generation output estimation device 100 utilizes the first photovoltaic power generation output, which is the power generation output of the first photovoltaic power generation facility, at a predetermined time point of the second photovoltaic power generation facility (hereinafter, “estimated time point””. The second photovoltaic power generation output, which is the photovoltaic power generation output of (referred to as), is estimated. Estimated time points include not only current time points but also past or future time points. The photovoltaic power generation output estimation device 100 may be realized by a general-purpose computer system such as a personal computer executing a program, or may be realized by a dedicated computer system. The detailed configuration of the photovoltaic power generation output estimation device 100 will be described later. The distance between the first photovoltaic power generation facility and the second photovoltaic power generation facility is not more than a predetermined value. The predetermined value is, for example, several km, but is not limited to this.

Next, the function of the photovoltaic power generation output estimation device 100 will be described. The photovoltaic power generation output estimation device 100 uses the reference first photovoltaic power generation output and the residual demand of the consumer to be estimated, and is the second power output of the photovoltaic power generation facility 202 of the consumer to be estimated. Estimate the PV output. The first photovoltaic power generation output is the power generation output of the first photovoltaic power generation facility installed in the first consumer, and the second photovoltaic power generation output is the power generation of the second photovoltaic power generation facility installed in the second consumer. The output. The power output of the photovoltaic power generation facility 202 and the power consumption of the load 201 include active power and ineffective power. However, since general photovoltaic power generation equipment, especially household photovoltaic power generation equipment, is operated at a constant power factor, it is possible to easily estimate the ineffective power by estimating the active power. Therefore, the power generation output of the photovoltaic power generation facility 202 and the power consumption of the load 201 described below refer to the active power.

Here, the outline of the method of estimating the power generation output of the photovoltaic power generation facility 202 in the photovoltaic power generation output estimation device 100 will be described. Here, the photovoltaic power generation facility 202 of the consumer 200-2 will be described as an example of the estimation target of the power generation output, and the photovoltaic power generation facility of the consumer 200-1 will be used as the reference in the estimation. The power generation output of 202 will be described as an example. That is, in this example, the first photovoltaic power generation output is the photovoltaic power generation output of the photovoltaic power generation facility 202 of the consumer 200-1, and the second photovoltaic power generation output to be estimated is the photovoltaic power generation of the consumer 200-2. It is the power generation output of the equipment 202. In addition, when the estimation target is the power generation output of each of the solar power generation facilities 202 of the consumers 200-4 to 200-n, that is, the second solar power generation output is each of the consumers 200-4 to 200-n. Even in the case of the power output of the photovoltaic power generation facility 202, the power output of the photovoltaic power generation facility 202 can be similarly estimated by using a value corresponding to each consumer as the residual demand. In addition, as the first photovoltaic power generation output, which is the photovoltaic power generation output used as a reference, instead of the photovoltaic power generation output of the photovoltaic power generation facility 202 of the consumer 200-1, the photovoltaic power generation facility 202 of the consumer 200-3 Similarly, when the power generation output, that is, the power generation output measured by the measuring instrument 206 is used, the second photovoltaic power generation output can be estimated.

The residual demand of the consumer 200-2, which is the second consumer, is P ₂ (t), the power consumption of the load 201 of the consumer _200-2 is PL 2 (t), and the consumer 200, which is the output of the second photovoltaic power generation. Let the power generation output of the photovoltaic power generation facility 202 of -2 be P _PV2 (t). Further, the power generation output of the photovoltaic power generation facility 202 of the consumer 200-1, which is the first consumer, that is, the first photovoltaic power generation output is _PPV1 (t). At this time, the equation relating to P ₂ (t) holds as in the following equation (1), and further, the equation relating to P _PV2 (t) can be assumed as in the equation (2).

Here, α is a coefficient for converting the first photovoltaic power generation output into the second photovoltaic power generation output. ε ₁₂ (t) is a temporal disturbance. τ _s is the delay time, that is, the time during which the solar radiation fluctuation propagates between the installation point of the first photovoltaic power generation facility and the installation point of the second photovoltaic power generation facility. The time during which the solar radiation fluctuation propagates between the installation point of the first photovoltaic power generation facility and the installation point of the second photovoltaic power generation facility, that is, the solar radiation fluctuation between the first photovoltaic power generation facility and the second photovoltaic power generation facility. The time for propagation is also called the first delay time.

Then, assuming that the fluctuations of _PL 2 (t) and ε ₁₂ (t) do not correlate with the fluctuations of P _PV1 (t), respectively, in the above equations (1) and (2), temporally. Assuming that the steady state is established, the following equation (3) is derived from the above equations (1) and (2).

Here, Cov _t [] means an operator for calculating a covariance function of two quantities, and Cov _t [P _PV1 (t), P _PV1 (t + τ + τ _s )] in the above equation (3) is , Τ = -τ _s takes the maximum value. Therefore, the covariance function Cov _t [P _PV1 (t), P2 ₍ t + τ)] of the time series data of the first photovoltaic power generation output and the residual demand of the second consumer is the minimum value (the sign is “-”. And the value obtained by multiplying the time lag when taking (the absolute value is the maximum value) by -1 is the delay time τ _s .

Further, by substituting τ = 0 into the above equation (3), the following equation (4) is obtained.

Then, by transforming the above equation (4) into the following equation (5), an estimated value of the coefficient α can be obtained.

Cov _t [P _PV1 (t), P2 ₍ t)] is a covariance between the first photovoltaic power generation output P _PV1 (t) and the residual demand P2 ₍ t), and this covariance is the first covariance. Called variance. Cov _t [P _PV1 (t), P _PV1 (t + τ _s )] is the autocovariance of P _PV1 (t) and P _PV1 (t + τ _s ), which are the two first photovoltaic outputs with a delay time of τ _s . It is a variance, and this autocovariance is called a second covariance. Then, α is obtained by dividing the first covariance Cov _t [P _PV1 (t), P2 (t)] by the _second covariance Cov _t [P _PV1 (t), P _PV1 (t + τ _s )]. It is a coefficient obtained by multiplying by 1.

Further, assuming that ε ₁₂ (t + τ) is minute and can be ignored, the following equation (6), which is an approximate equation for the second photovoltaic power generation output, can be obtained from the above equation (2).

The photovoltaic power generation output estimation device 100 of the present embodiment estimates the second photovoltaic power generation output by using the estimation method described above. That is, the photovoltaic power generation output estimation device 100 calculates the first covariance and the second covariance described above, and estimates the coefficient α by the equation (5) using the calculated first covariance and second covariance. Find the value. The process for obtaining the coefficient α may be performed before the estimated time point, which is the time of the estimation target of the second photovoltaic power generation output. However, if the estimated time point is in the past, it may be performed from the estimated time point, which is the future of the estimated time point, to the present. As described above, since the time series data of each measurement data is used for the calculation of the coefficient α, the measurement data for a certain period is required. This period is set before the estimated time point and within a range not too far from the estimated time point. For example, this period is an arbitrary period from the estimation time at which the second photovoltaic power generation output is estimated to the period retroactive by about 1 or 2 weeks. For example, the process of obtaining the estimated value of the coefficient α is performed on the day before the day when the second photovoltaic power generation output is estimated, but the timing of performing the process of obtaining the estimated value of the coefficient α is not limited to this example. Then, the photovoltaic power generation output estimation device 100 multiplies the estimated value of the coefficient α by the first photovoltaic power generation output P _PV1 (t + τ _s ) at the time when the delay time τ _s deviates from the estimation time point t, so that the second sun The estimated value of the photovoltaic power generation output P _PV2 (t) can be calculated.

In the above calculation method, there is no correlation between PL2 (t) and _PPV1 (t), that is, _{Cov t} [P _PV1 (t), _PL2 (t)] is extremely small and can be approximated to 0. I made an assumption, but sometimes this assumption does not hold. For example, when the solar radiation intensity is weak and the fluctuation of the first photovoltaic power generation output is small, or when the fluctuation of the solar radiation intensity is small and the fluctuation of the first photovoltaic power generation output is small, Covert [P _PV1 ( _t ), P _PV1 Since (t + τ _s )] becomes smaller, the value of Cov _t [P _PV1 (t), _PL2 (t)] cannot be ignored by approximating it to 0, and the above process does not hold. In such a case, the estimation accuracy of the coefficient α is lowered, so that the estimation accuracy of the second photovoltaic power generation output _PPV2 (t) is also lowered. Therefore, it can be said that the period under the condition that the above assumption does not hold is a period that is not suitable for the estimation of the coefficient α, that is, a period in which the estimation accuracy is lowered.

In the present embodiment, the estimation value of α estimated by using the measurement data of the period unsuitable for the estimation of the coefficient α is estimated in order to avoid being used for the estimation of the second photovoltaic power generation output _PPV2 (t). It is determined whether or not the period is suitable for the value, and if the period is not suitable for estimating the coefficient α, the estimated value of α calculated using the measurement data of the period is not adopted. As will be described later, for example, when the initial value of the coefficient α is predetermined and a process for obtaining the coefficient α is performed, the value of the coefficient α estimated by the process is updated, but it is suitable for estimating the coefficient α. If it is determined that there is no period, the process for obtaining the coefficient α is not performed or the update with the value calculated by the process for obtaining the coefficient α is not performed.

Specifically, the photovoltaic power generation output estimation device 100 calculates the self-coefficient λ, which is a coefficient for estimating the first photovoltaic power generation output, by the same calculation method as the coefficient α shown in the above equation (5). do. The self-coefficient λ is a verification coefficient used for verification of estimation accuracy. Specifically, as the first covariance of the molecule of the formula (5), instead of the covariance of the first photovoltaic output P _PV1 (t) and the residual demand P2 (t) of the _second consumer, the first covariance is performed. ₁ Using the covariance of the photovoltaic output P _PV1 (t) and the residual demand P1 (t) of the first consumer, the following equation (7) excluding the delay time τ _s in the denominator of the equation (5) To calculate the self-factor λ. In the present embodiment, such a covariance between the first photovoltaic power generation output P _PV1 (t) and the residual demand P1 (t) of the _first consumer is also referred to as the first covariance. That is, the first covariance is the covariance between the output of the first photovoltaic power generation and the residual demand, and this residual demand may be the residual demand of consumers other than the first consumer, and the residual of the first consumer. It may be demand. Further, the denominator in the equation (5) includes the delay time τ _s , and the denominator in the equation (7) does not include the delay time τ _s , but if the delay time τ _s is set to 0, both are the same. Therefore, in the following, the denominator in the equation (7) is also referred to as the second covariance like the denominator in the equation (5).

As described above, the self-coefficient λ corresponds to the conversion coefficient for estimating the first photovoltaic power generation output itself from the measurement data of the first photovoltaic power generation output. Ideally λ is 1. Actually, λ does not become 1 due to various errors and the like, but it is considered that the closer λ is to 1, the smaller the error. Therefore, when | λ-1 |, which is the absolute value of the difference between λ and 1, is close to 0, the error of λ is considered to be small, and the coefficient α estimated by the same method as λ is also | λ-1. When | is close to 0, the estimation error is considered to be small. Therefore, in the present embodiment, | λ-1 | is used to determine whether or not the period is not suitable for estimating the coefficient α. As described above, in the present embodiment, it is determined whether or not the period is not suitable for the estimation of the coefficient α, and if it is determined that the period is not suitable for the estimation of the coefficient α, the measurement data of the period is used. Since the coefficient α is not adopted, it is possible to prevent a decrease in the estimation accuracy of the second photovoltaic power generation output _PPV2 (t). The details of the processing of the photovoltaic power generation output estimation device 100 will be described later.

Next, an example of the functional configuration of the photovoltaic power generation output estimation device 100 will be described. FIG. 2 is a block diagram showing a functional configuration example of the photovoltaic power generation output estimation device 100 of the present embodiment. The photovoltaic power generation output estimation device 100 is a device that estimates the second photovoltaic power generation output, which is the power generation output at the time of estimation of the photovoltaic power generation facility 202. As shown in FIG. 2, the photovoltaic power generation output estimation device 100 includes a data acquisition unit 110, a first covariance calculation unit 120, a second covariance calculation unit 130, an estimation accuracy verification unit 140, and an estimation unit 150. A display unit 160, an input reception unit 170, and a storage unit 180 are provided.

The data acquisition unit 110 receives measurement data from the smart meters 203 to 205 and the measuring instrument 206 via the communication network 25 and the communication network 50, and stores the received measurement data in the storage unit 180. Specifically, the measurement data received from the smart meter 204 and the measuring instrument 206 is stored in the storage unit 180 as the photovoltaic power generation output data 182, and the measurement data received from the smart meter 205 is stored in the storage unit 180 as the residual demand data 183. Will be done. These measurement data are associated with the time for each measured value. The time interval of each measured value of these measurement data stored in the storage unit 180 is, for example, 1 second, 10 seconds, 1 minute, 30 minutes, or 1 hour, but is not particularly limited thereto. , The user determines the appropriate number. Further, these measurement data are stored together with the identification information of the consumers 200-1 to 200-n or the identification information of the photovoltaic power generation facility 202. Further, when the data acquisition unit 110 receives the measurement data of the smart meter 203 of the consumer 200-1 and the measurement data of the smart meter 204, the data acquisition unit 110 uses these measurement data to obtain the residual demand of the consumer 200-1. It is calculated and stored in the storage unit 180 as residual demand data 183.

Here, an example is shown in which the photovoltaic power generation output estimation device 100 acquires measurement data from the smart meters 203 to 205 and the measuring instrument 206 via the communication network 25 and the communication network 50, but the smart meters 203 to 205 are shown. The measurement data of the above may be collected by a so-called aggregation device, a head-end system, or the like, and may be transmitted from the collected device to the photovoltaic power generation output estimation device 100. That is, there may be a device for collecting measurement data in the communication network 25 and the communication network 50, and the photovoltaic power generation output estimation device 100 may acquire the measurement data from the device. Further, the communication route between the photovoltaic power generation output estimation device 100 and the smart meters 203 to 205 and the measuring instrument 206 is not limited to the examples shown in FIGS. 1 and 2, and the photovoltaic power generation output estimation device 100 is the measuring instrument. The communication route for receiving 206 may be different from the communication route for the photovoltaic power generation output estimation device 100 to receive the measurement data of the smart meters 203 to 205. Further, the communication network between the photovoltaic power generation output estimation device 100 and the smart meters 203 to 205 and the measuring instrument 206 does not have to be separated into the communication network 25 and the communication network 50. Further, the photovoltaic power generation output estimation device 100 may receive the measurement data of the smart meters 203 to 205 through a plurality of communication routes.

The storage unit 180 stores the above-mentioned photovoltaic power generation output data 182 and residual demand data 183, and stores various data calculated by the photovoltaic power generation output estimation device 100 as calculated data 185. Further, the storage unit 180 stores, for example, the location information data 181 and the coefficient initial value data 184 input by the user via the input reception unit 170. The location information data 181 and the coefficient initial value data 184 may be transmitted from another device instead of being input via the input receiving unit 170, acquired by the data acquisition unit 110, and stored in the storage unit 180. .. The location information data 181 is data on location information such as the address or latitude / longitude of consumers 200-1 to 200-n who own the photovoltaic power generation facility 202.

The first covariance calculation unit 120 refers to the location information data 181 of the storage unit 180, and the distance from the second consumer, who is the consumer whose power generation output is estimated, is within a predetermined distance. The first consumer who has the photovoltaic power generation facility 202 as a reference at the time of estimation is selected. As described above, the first consumer is the consumer 200-1 having the photovoltaic power generation facility 202 subject to the total purchase contract, or the consumer whose power generation output of the photovoltaic power generation facility 202 is measured by the measuring instrument 206. It is 200-3. The first covariance calculation unit 120 calculates and calculates the first covariance, which is the covariance between the first photovoltaic power generation output, which is the power output of the photovoltaic power generation facility 202 of the first consumer, and the residual demand. 1 The covariance is stored in the storage unit 180 as calculated data 185. As described above, the first covariance may be the residual demand of a consumer other than the first consumer, that is, the consumer whose power generation output is estimated, or the residual demand of the first consumer. As described above, the consumer whose power generation output is estimated is, for example, the consumer 200-2, but may be the consumer 200-4 to 200-n. All of the consumers 200-2, 200-4 to 200-n may be sequentially estimated, and the estimation may be performed for each of the consumers 200-2, 200-4 to 200-n, and the total value is required. If this is the case, the total value may be estimated in a lump sum.

Specifically, the first covariance calculation unit 120 first extracts the first photovoltaic power generation output in the first period from the photovoltaic power generation output data 182 of the storage unit 180 in order to calculate the self-factor λ described above. Then, the residual demand of the first consumer is extracted from the residual demand data 183 of the storage unit 180 and read out. The first covariance calculation unit 120 calculates the first covariance, which is the covariance between the first photovoltaic power generation output and the residual demand of the first consumer, using the read data, and the calculated data is stored in the storage unit 180. Store as 185. The first period is, for example, a period of about 6 to 10 hours, preferably about 8 hours, within a period of about 1 to 2 weeks from the estimation time. The first period is preferably a period in which the solar radiation intensity is strong and the fluctuation of the solar radiation intensity is large. However, in the present embodiment, in order to determine whether or not the first period is a period suitable for calculating the coefficient α, even if the first period is selected regardless of the solar radiation intensity, the solar radiation intensity is weak. Even if a period in which the fluctuation of the solar radiation intensity is small is set as the first period, the coefficient α corresponding to the first period is not used, so that the deterioration of the estimation accuracy of the coefficient α can be suppressed. Therefore, it is not necessary to consider the solar radiation intensity when setting the first period.

Further, when the first covariance calculation unit 120 is instructed by the estimation accuracy verification unit 140 to calculate the first covariance for calculating the coefficient α, the first covariance calculation unit 120 stores the residual demand of the second consumer in the first period. It is extracted and read from the residual demand data 183 of 180. The first covariance calculation unit 120 uses the read-out first photovoltaic power generation output and the read-out residual demand of the second consumer to generate the first photovoltaic power generation output and the residual demand of the second consumer in the first period. The first covariance, which is the covariance with the demand, is calculated and stored in the storage unit 180 as the calculated data 185. Further, the first covariance calculation unit 120 calculates the delay time τ _s in the first period by using the read-out first photovoltaic power generation output and the read-out residual demand of the second consumer. It is stored as calculated data 185 in the storage unit 180. The delay time may not be calculated and may be approximated to 0.

In the calculation of the second covariance for calculating the self-covariance λ, the second covariance calculation unit 130 extracts the time series data of the first solar power generation output in the first period from the calculation data 185 of the storage unit 180. The variance of the time-series data of the first solar power generation output read out is calculated as the second covariance.

Further, in the calculation process of the second covariance for calculating the coefficient α, the second covariance calculation unit 130 extracts and reads the delay time τ _s from the calculation data 185 of the storage unit 180, and reads the delay time τ s from the calculation data 185 of the storage unit 180. 1 The photovoltaic power generation output and the first photovoltaic power generation output in the second period deviated by the delay time τ _s from the first period are extracted from the photovoltaic power generation output data 182 of the storage unit 180 and read out. Then, the second covariance calculation unit 130 calculates the second covariance, which is the autocovariance of the two first photovoltaic power generation outputs deviated by the delay time τ _s , and stores the calculated second covariance in the storage unit 180. It is stored as calculated data 185. Specifically, the second co-dispersion calculation unit 130 deviates from the photovoltaic power generation output data 182 of the storage unit 180 with the time-series data of the first photovoltaic power generation output in the first period and the delay time τ _s from the first period. The time-series data of the first photovoltaic power generation output in the second period is read out. Then, the second covariance calculation unit 130 is a second covariance which is an autocovariance between the time series data of the first solar power generation output in the first period and the time series data of the first solar power generation output in the second period. The variance is calculated, and the calculated second covariance is stored in the storage unit 180 as the calculated data 185. When the delay time τ _s is approximated to 0 in the process for calculating the coefficient α, the second covariance calculation unit is the same as the process for calculating the self-variance λ. Reference numeral 130 calculates the variance of the time-series data of the first solar power generation output in the first period as the second covariance.

The estimation accuracy verification unit 140 determines whether or not the estimation accuracy of the second photovoltaic power generation output satisfies a predetermined accuracy. Specifically, the estimation accuracy verification unit 140 calculates the self-coefficient λ described above by using the first covariance and the second covariance for calculating the self-coefficient λ stored in the storage unit 180, and self. Using the coefficient λ, it is determined whether or not the first period is a period suitable for calculating the coefficient α. When the estimation accuracy verification unit 140 determines that the first period is a period suitable for calculating the coefficient α, the estimation accuracy verification unit 140 instructs the first covariance calculation unit 120 to calculate the first covariance for calculating the coefficient α. ..

The estimation unit 150 calculates a coefficient α, which is an estimated value of the ratio between the first photovoltaic power generation output and the second photovoltaic power generation output, using the first co-dispersion and the second co-dispersion, and calculates the coefficient α from the time of estimation. The second photovoltaic power generation output is estimated by multiplying the first photovoltaic power generation output with a delay time deviation. Specifically, the estimation unit 150 reads out the first covariance and the second covariance for calculating the coefficient α stored as the calculation data 185 in the storage unit 180, and reads out the first covariance and the second covariance. Using the variance, the coefficient α for each consumer whose power output is estimated is estimated by the above equation (5). The estimation unit 150 updates the value of the coefficient α stored as the calculated data 185 with the estimated value. Further, the estimation unit 150 extracts and reads out the first photovoltaic power generation output at the time when the delay time τ _s deviates from the estimation time from the photovoltaic power generation output data 182 of the storage unit 180, and has a coefficient α for the first photovoltaic power generation output. By multiplying by, the second photovoltaic power generation output, which is the photovoltaic power generation output of the consumer to be estimated, is estimated, and the estimation result is stored in the storage unit 180 as calculation data 185. Although the coefficient α is unlikely to change in a short period such as 1 or 2 weeks, the delay time τ _s tends to change in a short period due to a change in wind direction or speed. Therefore, it is preferable that the delay time τ _s when estimating the second photovoltaic power generation output is changed in real time in consideration of meteorological data and the like.

The display unit 160 displays various data stored in the storage unit 180. The input receiving unit 170 receives input from the user.

Here, the hardware configuration of the photovoltaic power generation output estimation device 100 will be described. The photovoltaic power generation output estimation device 100 of the present embodiment is a computer system by executing a photovoltaic power generation output estimation program, which is a program in which the processing in the photovoltaic power generation output estimation device 100 is described, on the computer system. Functions as the photovoltaic power generation output estimation device 100. FIG. 3 is a diagram showing a configuration example of a computer system that realizes the photovoltaic power generation output estimation device 100 of the present embodiment. As shown in FIG. 3, this computer system includes a control unit 101, an input unit 102, a storage unit 103, a display unit 104, a communication unit 105, and an output unit 106, which are connected via a system bus 107. There is.

In FIG. 3, the control unit 101 is, for example, a processor such as a CPU (Central Processing Unit), and executes a photovoltaic power generation output estimation program in which processing in the photovoltaic power generation output estimation device 100 of the present embodiment is described. .. The input unit 102 is composed of, for example, a keyboard, a mouse, or the like, and is used by a user of a computer system to input various information. The storage unit 103 includes various memories such as RAM (Random Access Memory) and ROM (Read Only Memory) and a storage device such as a hard disk, and is a necessary program obtained in the process of a program to be executed by the control unit 101. Memorize data, etc. The storage unit 103 is also used as a temporary storage area for the program. The display unit 104 is composed of a display, an LCD (liquid crystal display panel), and the like, and displays various screens to a user of a computer system. The communication unit 105 is a receiver and a transmitter that performs communication processing. The output unit 106 is a printer or the like. Note that FIG. 3 is an example, and the configuration of the computer system is not limited to the example of FIG.

Here, an operation example of the computer system until the photovoltaic power generation output estimation program of the present embodiment becomes executable will be described. In a computer system having the above-described configuration, for example, solar power output estimation is performed from a CD-ROM or DVD-ROM set in a CD (Compact Disc) -ROM drive or DVD (Digital Versatile Disc) -ROM drive (not shown). The program is installed in the storage unit 103. Then, when the photovoltaic power generation output estimation program is executed, the photovoltaic power generation output estimation program read from the storage unit 103 is stored in the storage unit 103. In this state, the control unit 101 executes the process as the photovoltaic power generation output estimation device 100 of the present embodiment according to the program stored in the storage unit 103.

In the above description, a program describing the processing in the solar power generation output estimation device 100 using a CD-ROM or a DVD-ROM as a recording medium is provided, but the present invention is not limited to this, and the configuration of the computer system, Depending on the capacity of the program to be provided, for example, a program provided by a transmission medium such as the Internet via the communication unit 105 may be used.

The first covariance calculation unit 120, the second covariance calculation unit 130, the estimation accuracy verification unit 140, and the estimation unit 150 shown in FIG. 2 are photovoltaic power generation output estimation programs stored in the storage unit 103 shown in FIG. Is realized by being executed by the control unit 101 shown in FIG. The storage unit 180 shown in FIG. 2 is a part of the storage unit 103 shown in FIG. The data acquisition unit 110 shown in FIG. 2 is realized by the communication unit 105 and the control unit 101 shown in FIG. The display unit 160 shown in FIG. 2 is realized by the display unit 104 shown in FIG. 3, and the input reception unit 170 shown in FIG. 2 is realized by the input unit 102 shown in FIG. The photovoltaic power generation output estimation device 100 may be realized by a plurality of computer systems.

For example, the photovoltaic power generation output estimation program of the present embodiment calculates the second photovoltaic power generation output, which is the photovoltaic power generation output of the second photovoltaic power generation facility installed within a predetermined distance from the first photovoltaic power generation facility. , A computer that estimates the steps using the first photovoltaic power generation output, which is the power generation output of the first photovoltaic power generation facility, and the residual demand, which is the apparent power consumption of the consumer in which the second photovoltaic power generation facility is installed. To execute. Further, the photovoltaic power generation output estimation program of the present embodiment uses the power output of the first photovoltaic power generation facility and the residual demand which is the apparent power consumption of the consumer in which the first photovoltaic power generation facility is installed. The second step (determination step) of determining whether or not the estimation accuracy of the photovoltaic power generation output satisfies the predetermined accuracy is executed by the computer.

Next, the details of the operation of the photovoltaic power generation output estimation device 100 of the present embodiment will be described. FIG. 4 is a flowchart showing an example of an estimation processing procedure for the coefficient α by the photovoltaic power generation output estimation device 100 of the present embodiment. Hereinafter, the process of estimating the second photovoltaic power generation output by the photovoltaic power generation output estimation device 100 will be described along with the flow of FIG. First, the first covariance calculation unit 120 sets the photovoltaic power generation facility 202 to be estimated with the coefficient α, and selects the photovoltaic power generation facility using the power generation output (step S1). When there are a plurality of photovoltaic power generation facilities 202 for which the coefficient α is estimated, one photovoltaic power generation facility 202 is selected from the plurality of photovoltaic power generation facilities 202. In the second and subsequent steps S1, among the plurality of photovoltaic power generation facilities 202, the photovoltaic power generation facility 202 that has not been processed in steps S2 to S6 is selected as the photovoltaic power generation facility for which the coefficient α is estimated. Set as 202. Note that this step S1 may be performed by the estimation unit 150, and the result may be notified to the first covariance calculation unit 120.

In step S1, specifically, the first covariance calculation unit 120 refers to the location information data 181 of the storage unit 180, and the distance from the second consumer who is the consumer whose power generation output is estimated is determined. A first consumer who is within a predetermined distance and has a photovoltaic power generation facility 202 as a reference for estimation is selected.

FIG. 5 is a diagram schematically showing the geographical position of the photovoltaic power generation facility 202 of each consumer shown by the location information data 181 of the present embodiment. In FIG. 5, it is assumed that the geographical position of the photovoltaic power generation facility 202 of each consumer coincides with the geographical position of the consumer. In FIG. 5, the photovoltaic power generation equipment 202 of the consumer 200-i (i = 1, 2, ..., 8) is shown as PVi. For example, PV1 is the photovoltaic power generation facility 202 of the consumer 200-1, and PV2 is the photovoltaic power generation facility 202 of the consumer 200-2. PV1 and PV3 shown in the solid line frame are the photovoltaic power generation facilities 202 of the consumers 200-1 and 200-3, respectively, and the power generation output is measured. PV2 and the like shown by the broken line frame are photovoltaic power generation facilities 202 such as the consumer 200-2 whose power generation output is not measured.

The range 301 of FIG. 5 is a circle whose radius is the above-mentioned predetermined distance centered on PV1 which is the solar power generation facility 202 of the consumer 200-1, and the solar power generation facility of the consumer within the range 301. The power generation output of PV1 can be used to estimate the power generation output of 202. Similarly, the range 302 of FIG. 5 is a circle whose radius is the above-mentioned predetermined distance centered on PV3, which is the photovoltaic power generation facility 202 of the consumer 200-3, and the sun of the consumer within the range 302. The power generation output of PV3 can be used to estimate the power generation output of the photovoltaic power generation facility 202. Therefore, for example, when PV2, which is the photovoltaic power generation facility 202 of the consumer 200-2, is set as the estimation target of the power generation output, PV2 corresponds to PV1 because it is within the range 301 and within the range 302. Both the consumer 200-1 and the consumer 200-3 corresponding to PV3 are candidates for the first consumer. In this way, when there are a plurality of candidates for the first consumer, the consumer 200-2, which is the consumer closer to the PV2 to be estimated, is selected as the first consumer. Alternatively, the measurement data of a plurality of candidates for the first consumer may be combined and used, in which case the plurality of consumers are selected as the first consumer.

Returning to the description of FIG. After step S1, the first covariance calculation unit 120 sets the first period (step S2). Specifically, the first covariance calculation unit 120 is the first photovoltaic power generation output which is the power generation output of the photovoltaic power generation facility 202 of the first consumer in the photovoltaic power generation output data 182 stored in the storage unit 180. Refer to the time series data of, and search for the period when the solar intensity is strong and the fluctuation of the solar intensity is large. For example, the first co-dispersion calculation unit 120 has 6 to 10 hours on the day before the day when the second photovoltaic power generation output is estimated, in which the first photovoltaic power generation output is large and the fluctuation of the first photovoltaic power generation output is large. Explore a period of time. Then, the first covariance calculation unit 120 sets the searched period as the first period. The first covariance calculation unit 120 does not search for the first period from the above period (for example, the day before the day when the second photovoltaic power generation output is estimated), but searches for the above period itself. Then, the first period may be searched from the period. As described above, the first period is set by an arbitrary method without searching for a period of about 6 to 10 hours in which the first photovoltaic power generation output is large and the fluctuation of the first photovoltaic power generation output is large. You may.

Next, the first covariance calculation unit 120 calculates the first covariance for calculating the self-coefficient λ (step S3). The first covariance calculation unit 120 stores the calculated first covariance in the storage unit 180 as calculated data 185. The details of the calculation of the first covariance will be described later.

Next, the second covariance calculation unit 130 calculates the second covariance for calculating the self-coefficient λ (step S4). The second covariance calculation unit 130 stores the calculated second covariance in the storage unit 180 as calculated data 185. The details of the calculation of the second covariance will be described later.

Next, the estimation accuracy verification unit 140 calculates the self-coefficient λ (step S5). Specifically, the estimation accuracy verification unit 140 uses the first covariance and the second covariance for calculating the self-coefficient λ stored in the storage unit 180, and the self-coefficient according to the above equation (7). Calculate λ.

Next, the estimation accuracy verification unit 140 determines whether or not | λ-1 | is equal to or less than the threshold value u (step S6). When | λ-1 | is equal to or less than the threshold value u (step S6 Yes), the coefficient α is calculated (step S7), and it is determined whether or not to end the estimation of the coefficient α (step S8). The details of the calculation process of the coefficient α will be described later. The process of step S8 is performed by the first covariance calculation unit 120 or the estimation unit 150. The determination of whether or not to end the estimation of the coefficient α in step S8 is the determination of whether or not the processes of steps S2 to S6 have been performed for all the photovoltaic power generation facilities 202 to be estimated by the coefficient α. When the processes of steps S2 to S6 are performed for all the photovoltaic power generation facilities 202 for which the coefficient α is estimated, it is determined as Yes in step S8, and the coefficient α estimation process ends. If there is a photovoltaic power generation facility 202 that has not been processed in steps S2 to S6 among the photovoltaic power generation facilities 202 for which the coefficient α is estimated, it is determined as No in step S8, and the processing from step S1 is repeated. ..

On the other hand, when | λ-1 | exceeds the threshold value u (step S6 No), step S7 is not executed and the process proceeds to step S8. As described above, the self-coefficient λ is a value calculated by the same method as the coefficient α using the residual demand of the first consumer, and ideally, the first photovoltaic power generation output is multiplied by λ. And the first photovoltaic power generation output. Therefore, as the self-coefficient λ deviates from 1, the error of the self-coefficient λ increases. Since the self-coefficient λ is calculated by the same method as the coefficient α, it can be assumed that the error of the coefficient α is also large when the error of the self-coefficient λ is large. It is estimated that the period is not suitable for the calculation of. Therefore, in the present embodiment, when | λ-1 | exceeds the threshold value u, it is possible to suppress a decrease in the estimation accuracy of the coefficient α by not calculating the coefficient α. Here, when | λ-1 | exceeds the threshold value u, the coefficient α is not calculated, but after the coefficient α is calculated before step S6, it is determined as No in step S6. In that case, the coefficient α may not be updated.

In addition, when the estimation accuracy verification unit 140 does not estimate the coefficient α, that is, when No in step S6, when the estimation of the second consumer to be estimated has not been performed even once by that time, the storage unit 140 As the coefficient α in the calculated data 185 of 180, the value stored as the coefficient initial value data 184 is stored. The coefficient initial value data 184 is an initial value of the coefficient α for each second consumer. The initial value of the coefficient α is the solar power generation facility 202 of the second consumer and the photovoltaic power generation facility 202 of the first consumer when the first consumer corresponding to each second consumer is predetermined. It may be the ratio of the rated capacity to. If the first consumer corresponding to each second consumer is not predetermined, the second consumer's photovoltaic power generation facility 202 and the first consumer are used for all the customers who are candidates for the first consumer. The ratio of the rated capacity to the candidate photovoltaic power generation facility 202 may be stored as an initial value.

FIG. 6 is a diagram showing an example of data in the calculated data 185 stored in the storage unit 180 of the present embodiment. As shown in FIG. 6, the storage unit 180 stores the first covariance, the second covariance, the coefficient α, the delay time, and the estimated power generation output as the calculated data 185. The first covariance in the calculated data 185 includes the first covariance calculated in step S3 and the first covariance calculated in the process of estimating the coefficient α in step S7. The second covariance in the calculated data 185 includes the second covariance calculated in step S4 and the second covariance calculated in the process of estimating the coefficient α in step S7. Since the first covariance and the second covariance need only be temporarily stored, they may be deleted each time the estimation of the coefficient α corresponding to each consumer whose power generation output is estimated is completed. The coefficient α is updated with the estimated value when the estimation process of the coefficient α in step S7 is performed. Since the coefficient α and the delay time are different for each combination of the second consumer and the first consumer, that is, for each combination of the second photovoltaic power generation facility and the first photovoltaic power generation facility, the calculated data 185 for each combination. Is stored in the storage unit 180.

FIG. 7 is a diagram showing an example of the coefficient α for each combination of the photovoltaic power generation equipment 202 stored in the storage unit 180 of the present embodiment. As shown in FIG. 7, when the coefficient α in step S7 is estimated, the second photovoltaic power generation facility 202, which is the photovoltaic power generation facility 202 for which the power generation output is estimated, and the photovoltaic power generation facility 202 used for the estimation. The estimated value of the coefficient α is stored as the calculated data 185 of the storage unit 180 for each combination with the first photovoltaic power generation facility. For example, PV2, which is the photovoltaic power generation facility 202 of the consumer 200-2, is estimated to have a power generation output using the power generation output of PV1, which is the photovoltaic power generation facility 202 of the consumer 200-1, and is used for the estimation. The estimated value of the coefficient α is α ₁ . As described above, when the coefficient α has never been estimated and is determined to be No in step S6 described above, the coefficient α of the calculated data 185 corresponds to the coefficient initial value data 184. The initial value is stored.

The calculation process of the first covariance, which is the process of step S3 of FIG. 4, will be described. FIG. 8 is a flowchart showing an example of the calculation processing procedure of the first covariance by the first covariance calculation unit 120 of the present embodiment. Although steps S3 and S4 are processes for calculating the self-coefficient λ, the first covariance and the second covariance are also used for calculating the coefficient α as shown in the above equation (5). Be done. FIG. 8 shows a flowchart shared by the process of calculating the first covariance for calculating the self-coefficient λ and the process of calculating the first covariance for calculating the coefficient α. Similarly, FIG. 9, which will be described later, also shows a flowchart shared by the process of calculating the second covariance for calculating the self-coefficient λ and the process of calculating the second covariance for calculating the coefficient α. .. As shown in the above equations (5) and (7), the delay time is used to calculate the coefficient α, whereas the delay time is not used to calculate the self-coefficient λ. Therefore, in the processes shown in FIGS. 8 and 9, it is determined whether or not to consider the delay time.

As shown in FIG. 8, first, the first covariance calculation unit 120 acquires the time-series data of the first photovoltaic power generation output in the first period and the time-series data of the residual demand (step S11). Specifically, the first co-dispersion calculation unit 120 reads out the first photovoltaic power generation output in the first period from the photovoltaic power generation output data 182 of the storage unit 180, so that the first solar power in the first period can be read. Acquire time-series data of power generation output. Further, when the first covariance calculation unit 120 calculates the first covariance for calculating the self-covariance λ, the first covariance calculation unit 120 determines the residual demand of the first consumer in the first period of the residual demand data 183 of the storage unit 180. By reading, the time series data of the residual demand in the first period is acquired. Further, when calculating the first covariance for calculating the coefficient α, the residual demand in the first period is read out from the residual demand data 183 of the storage unit 180 by reading the residual demand of the second consumer in the first period. Get time series data of demand. Since the time-series data of the first photovoltaic power generation output in the first period is used in both the processing for calculating the self-factor λ and the processing for calculating the coefficient α, the first covariance calculation unit. 120 calculates the coefficient α by holding the time-series data of the first photovoltaic power generation output in the first period read for the calculation of the self-factor λ until the processing is performed for the calculation of the coefficient α. In step S11 for the above, the process of extracting and reading the first photovoltaic power generation power from the photovoltaic power generation output data 182 of the storage unit 180 may be omitted.

The photovoltaic power generation output data 182 is the measurement data measured by the smart meter 204 and the measuring instrument 206 as described above, but an estimated value may be used instead of these measurement data. For example, the photovoltaic power generation output estimation device 100 estimates the photovoltaic power generation output by using the method of the present embodiment or another method, and stores the estimated photovoltaic power generation output as the photovoltaic power generation output data 182. You may. Alternatively, the user may input an estimated value or an actually measured value of the photovoltaic power generation output, and the input estimated value or the actually measured value may be stored as the photovoltaic power generation output data 182.

The first covariance calculation unit 120 calculates the covariance between the time-series data of the first photovoltaic power generation output and the time-series data of the residual demand as the first covariance (step S12). The first covariance calculation unit 120 writes the calculated first covariance in the storage unit 180 as calculated data 185.

Further, the first covariance calculation unit 120 determines whether or not to calculate the delay time (step S13), and when it is determined to calculate the delay time (step S13 Yes), the read-out first solar power generation output. And the residual demand of the second consumer who has also been read out, the delay time τ _s in the first period is calculated (step S14), stored in the storage unit 180 as the calculated data 185, and the first covariance is obtained. The calculation process is terminated. In step S13, the first covariance calculation unit 120 determines that the delay time is not calculated in the case of the process for calculating the self-covariance λ, that is, the process in step S3 shown in FIG. 4, and calculates the coefficient α. In the case of the process for calculating the first covariance in step S7 shown in FIG. 4, it is determined that the delay time is calculated. Even in the case of processing for calculating the coefficient α, the delay time may not be calculated and may be approximated to 0.

In step S14, in detail, the first covariance calculation unit 120 has a delay time between the installation point of the second photovoltaic power generation facility, which is the estimation target of the power generation output, and the installation point of the first photovoltaic power generation facility. Is calculated and stored in the storage unit 180 as the calculated data 185. This delay time is the time for the solar radiation fluctuation to propagate between the installation point of the first photovoltaic power generation facility and the installation point of the second photovoltaic power generation facility as described above. The first covariance calculation unit 120 has, for example, a covariance function Cov _t [P _PV1 (t), P2 ₍ t + τ)] of time-series data of the first solar power output and the residual demand of the second consumer. The delay time τ _s is calculated by multiplying the time lag when taking the minimum value (the sign is “−” and the absolute value is the maximum value) by -1.

When it is determined in step S13 that the delay time is not calculated (step S13 No), the first covariance calculation unit 120 ends the calculation process of the first covariance.

The second covariance calculation process, which is the process of step S4 of FIG. 4, will be described. FIG. 9 is a flowchart showing an example of the calculation processing procedure of the second covariance by the second covariance calculation unit 130 of the present embodiment.

The second covariance calculation unit 130 determines whether or not to use the delay time (step S21), and if it determines that the delay time is not used (step S21 No), sets the delay time τ _s to 0 and steps S22. Is not performed, but the processes after step S23, which will be described later, are carried out. When it is determined to use the delay time (step S21 Yes), the second covariance calculation unit 130 acquires the delay time τ _s by reading the delay time τ _s in the calculated data 185 stored in the storage unit 180. (Step S22). In step S21, the second covariance calculation unit 130 determines that the delay time is not used in the case of the process for calculating the self-covariance λ, that is, the process in step S4 shown in FIG. 4, and determines that the delay time is not used, and the coefficient α In the case of the processing for calculation, that is, the calculation processing of the second covariance in step S7 shown in FIG. 4, it is determined that the delay time is used.

Next, the second covariance calculation unit 130 acquires time-series data of the two first photovoltaic power generation outputs deviated by the delay time τ _s (step S23). That is, the second covariance calculation unit 130 uses the time series data of the first photovoltaic power generation output in the first period and the time series of the first photovoltaic power generation output in the second period, which is delayed by τ _s from the first period. The data is extracted from the photovoltaic power generation output data 182 of the storage unit 180, read out, and acquired. When the delay time τ _s is 0, the two first photovoltaic power generation outputs deviated by the delay time τ _s are the same, so that the second covariance calculation unit 130 is based on the photovoltaic power generation output data 182 of the storage unit 180. The first photovoltaic power generation output of the first period may be read out. Since the time-series data of the first photovoltaic power generation output in the first period is used in both the processing for calculating the self-factor λ and the processing for calculating the coefficient α, the second covariance calculation unit. The 130 calculates the coefficient α by holding the time-series data of the first photovoltaic power generation output in the first period read for the calculation of the self-factor λ until the processing is performed for the calculation of the coefficient α. In step S23 for the above, the process of extracting and reading the first photovoltaic power generation power in the first period from the photovoltaic power generation output data 182 of the storage unit 180 may be omitted.

Next, the second covariance calculation unit 130 calculates the second covariance using the acquired data (step S24). Specifically, the second covariance calculation unit 130 calculates the second covariance, which is the autocovariance of the time-series data of the two first photovoltaic power generation outputs deviated by the delay time τ _s . Specifically, the second covariance calculation unit 130 performs autocovariance between the time-series data of the first photovoltaic power generation output in the first period and the time-series data of the first photovoltaic power generation output in the second period. Calculated as the second covariance. When the delay time τ _s is 0, the second covariance calculation unit 130 calculates the variance of the time series data of the first photovoltaic power generation output in the first period as the second covariance. The second covariance calculation unit 130 writes the calculated second covariance in the calculation data 185, and ends the calculation process of the second covariance.

The calculation process of the coefficient α, which is the process of step S7 in FIG. 4, will be described. FIG. 10 is a flowchart showing an example of the calculation processing procedure of the coefficient α of the present embodiment. First, the first covariance calculation unit 120 calculates the first covariance for calculating the coefficient α (step S31). Specifically, when the estimation accuracy verification unit 140 determines Yes in step S6 of FIG. 4, it instructs the first covariance calculation unit 120 to calculate the first covariance for calculating the coefficient α, and first. The covariance calculation unit 120 calculates the first covariance and the delay time τ s by the process shown in FIG. 8, and stores the first covariance and the delay time τ _s in the storage unit 180 as the calculated data 185. When calculating the first covariance for calculating the coefficient α, the first covariance calculation unit 120 in step S11 shown in FIG. 8 in the first period of the residual demand data 183 of the storage unit 180. By reading the residual demand of the second consumer, the time series data of the residual demand in the first period is acquired. As a result, the covariance between the time-series data of the first photovoltaic power generation output and the time-series data of the residual demand of the second consumer is calculated as the first covariance. As described above, the delay time τ _s may be approximated to 0 and the calculation of the delay time τ _s may be omitted.

Next, the second covariance calculation unit 130 calculates the second covariance for calculating the coefficient α (step S32). Specifically, the second covariance calculation unit 130 uses the procedure shown in FIG. 9 to obtain the time-series data of the first solar power generation output in the first period and the second period in which the delay time τ _s deviates from the first period. The autocovariance with the time-series data of the first solar power generation output in the above is calculated as the second covariance, and is stored in the storage unit 180 as the calculated data 185.

Next, the estimation unit 150 calculates and updates the coefficient α (step S33). Specifically, the estimation unit 150 reads out the first covariance and the second covariance for calculating the coefficient α in the calculated data 185 of the storage unit 180, and divides the read first covariance by the second covariance. Then multiply by -1 to calculate the coefficient α. Then, the estimation unit 150 updates the coefficient α in the calculated data 185 of the storage unit 180 with the calculated value. As a result, the estimation process of the coefficient α is completed.

The estimated value of the coefficient α is calculated by the process described with reference to FIGS. 4 and 8 to 10. Further, as shown in FIG. 4, when it is determined that the self-coefficient λ is not suitable for estimating the coefficient α, the coefficient α is not calculated. In the above-mentioned example, when it is judged that the self-coefficient λ is not suitable for the estimation of the coefficient α, the calculation of the coefficient α is not performed, but it is estimated together with the calculation of the self-coefficient λ. When the unit 150 also calculates and temporarily holds the coefficient α and determines that it is suitable for estimating the coefficient α using the self-coefficient λ, the coefficient α in the calculated data 185 of the storage unit 180 is determined. May be updated with the calculated value. In this case, the estimation accuracy verification unit 140 instructs the estimation unit 150 not to update the coefficient α when it is determined that the coefficient α is not suitable for estimation using the self-coefficient λ, and the estimation unit 150 gives this instruction. Upon receiving it, the coefficient α in the calculated data 185 of the storage unit 180 is not updated.

Next, the estimation process of the photovoltaic power generation output using the coefficient α will be described. FIG. 11 is a flowchart showing an example of the estimation processing procedure of the photovoltaic power generation output of the present embodiment. When the estimation unit 150 determines the second photovoltaic power generation facility, which is the photovoltaic power generation facility to be estimated, the estimation unit 150 performs the process shown in FIG. 11 for estimating the power generation output of the second photovoltaic power generation facility at the time of estimation. The estimated time point may be within the first period described above, or may be different from the first period. The first photovoltaic power generation facility corresponding to the second photovoltaic power generation facility is selected by the estimation process of the coefficient α, and the coefficient α is set for each combination of the second photovoltaic power generation facility and the first photovoltaic power generation facility. Since it has been calculated, the combination used in the calculation of the coefficient α is used as the combination of the second photovoltaic power generation facility and the first photovoltaic power generation facility also in the estimation process of the photovoltaic power generation output. For example, when the coefficient α is calculated by the combination as shown in FIG. 7, in the estimation of the power generation output of the photovoltaic power generation facility 202 of the consumer 200-2, the photovoltaic power generation facility 202 of the consumer 200-1 is the first. 1 Used as a solar power generation facility.

As shown in FIG. 11, the estimation unit 150 first calculates the delay time (step S41). Specifically, the estimation unit 150 calculates the delay time between the installation point of the first photovoltaic power generation facility and the installation point of the second photovoltaic power generation facility at the time of estimation. The estimation unit 150 may use the delay time stored as the calculated data 185 in the storage unit 180 as the delay time at the estimation time, or update the delay time by reflecting the weather data at the estimation time. You may. For example, the estimation unit 150 may update the delay time using a value calculated by calculating the time for the clouds to pass from the wind direction and wind velocity data included in the meteorological data, or may update the delay time using a satellite image or the like. It may be updated, or it may be updated using meteorological data by a method other than these. Further, the delay time may be approximated to 0.

Next, the estimation unit 150 acquires the photovoltaic power generation output measurement data at a time deviated from the estimation time by the delay time (step S42). Specifically, the estimation unit 150 acquires the data corresponding to the first photovoltaic power generation facility in the photovoltaic power generation output data 182 of the storage unit 180 at a time delayed from the estimation time. Although the photovoltaic power generation output data 182 will be described here as the measurement data, as described above, an estimated value may be used instead of the measurement data.

Next, the estimation unit 150 acquires the coefficient α (step S43). Specifically, the estimation unit 150 acquires the coefficient α by reading the coefficient α stored as the calculated data 185 in the storage unit 180.

Next, the estimation unit 150 estimates the power generation output of the photovoltaic power generation facility to be estimated using the coefficient α and the photovoltaic power generation output measurement data (step S44), and ends the process. Specifically, in step S44, the second photovoltaic power generation output is estimated by the above equation (6) using the photovoltaic power generation output measurement data acquired in step S42, which is the first photovoltaic power generation output, and the estimated result is used. A certain power generation output estimated value is stored as the calculated data 185 of the storage unit 180.

The photovoltaic power generation output estimation device 100 of the present embodiment can use the photovoltaic power generation output of the photovoltaic power generation facility if there is at least one photovoltaic power generation facility capable of acquiring the photovoltaic power generation output by the above processing. It is possible to estimate the PV output of other desired PV equipment individually without installing a solarometer and a measuring instrument with high time resolution. Further, in the present embodiment, as described above, by verifying the coefficient α using the self-coefficient λ, it is possible to prevent the estimation accuracy of the coefficient α from deteriorating, so that the second photovoltaic power generation output is highly accurate. Can be estimated to.

Although an example of estimating the power generation output of the second photovoltaic power generation facility of one second consumer using the power generation output of the first photovoltaic power generation facility of one first consumer has been described, a plurality of second ones have been described. The total amount (total value) of the power generation output of the photovoltaic power generation facility may be estimated using the power generation output of the first photovoltaic power generation facility of one first consumer. Further, the total amount (total value) of the power generation output of the plurality of second photovoltaic power generation facilities may be estimated using the total amount (total value) of the power generation output of the first photovoltaic power generation facility of the plurality of first consumers. .. For example, when it is necessary to estimate the total amount of power generation output of photovoltaic power generation equipment of multiple consumers, such as distribution section unit such as pillar transformer unit, distribution line unit, transmission line unit, etc., power generation output measurement The total amount of power generation output of consumers who have not done so can be estimated in the same way.

FIG. 12 is a diagram showing an example of a first solar power generation facility and a second solar power generation facility when the solar power generation facilities of a plurality of consumers are targeted for estimation. In the example shown in FIG. 12, the photovoltaic power generation facility 202 of the consumer 200-i (i = 1, 2, ..., 8) is shown as PVi, as in the example shown in FIG. PV1 and PV3 shown in the solid line frame have the power generation output measured, and PV2 and the like shown in the broken line frame are the photovoltaic power generation facilities 202 in which the power generation output has not been measured. In the example shown in FIG. 12, PV2, 4 to 8 are the second photovoltaic power generation facilities whose power generation output is estimated, and PV1 and PV3 are the first photovoltaic power generation facilities whose power generation output is measured. In this case, the total amount of power generation output of PV2, 4 to 8 is defined as the second photovoltaic power generation output to be estimated, and the total amount of power generation output of PV1 and PV1 is defined as the first photovoltaic power generation output. (2) The power generation output can be estimated in the same manner as in the example of estimating the power generation output of the photovoltaic power generation facility using the power generation output of the first photovoltaic power generation facility of one first consumer. In this case as well, if PV1 to PV8 are within the range 303 whose radius is a predetermined distance, the distance between the first photovoltaic power generation facility and the second photovoltaic power generation facility is equal to or less than the predetermined distance. Will be.

FIG. 13 is a diagram showing an example of the coefficient α when the solar power generation equipment of a plurality of consumers is targeted for estimation. As shown in FIG. 13, when the power generation output of the plurality of second photovoltaic power generation facilities is estimated by using the power generation output of the first photovoltaic power generation facility of the plurality of first consumers, the power generation output of the plurality of photovoltaic power generation facilities is estimated. The coefficient α is calculated for each combination. In such a case, the delay time is the time for the solar radiation fluctuation to propagate between the position of the center of gravity of the installation point of the first photovoltaic power generation facility and the position of the center of gravity of the installation point of the second photovoltaic power generation facility. Will be. The delay time may be set to 0. FIG. 13 shows an example in which a coefficient α is set for each distribution section and the power generation output is estimated for each distribution section, but the unit for integrating consumers is not limited to this.

Next, the verification result of the effect of the photovoltaic power generation output estimation device 100 will be described. FIG. 14 is a diagram showing an example of verification results of the power generation output estimation method of the present embodiment. In FIG. 14, the horizontal axis indicates the time, and the vertical axis indicates the second photovoltaic power generation output (active power). The broken line shows the estimated value of the second photovoltaic power generation output estimated by the estimation method described in the present embodiment, and the solid line shows the measured value. In the example shown in FIG. 14, the solar power generation equipment of the consumer who is the target of the total purchase contract of the plurality of houses is set as the first solar power generation equipment, and the solar power generation of the consumer who is the target of the surplus purchase contract of the plurality of houses. The result of estimating the total amount of the power generation output of the 2nd photovoltaic power generation facility is shown with the facility as the 2nd photovoltaic power generation facility. That is, in the example shown in FIG. 14, the measurement data by the smart meter 204 of the first photovoltaic power generation facility is used as the first photovoltaic power generation output, and the smart meter 205 of the second consumer having the second photovoltaic power generation facility Using the measured data as the residual demand, the total amount of power generation output of the second photovoltaic power generation facility is estimated. Further, in the example shown in FIG. 14, the coefficient α is obtained on the day before the day of the estimation time, with 16 points of 30 units from 8:00 to 16:00 on a certain day in the past as the estimation time, and the day of the estimation time. The second photovoltaic power generation output was estimated using the measurement data of the first photovoltaic power generation output of. In this verification, the measured value obtained in FIG. 14 is calculated by separately specially measuring the power generation output of the second photovoltaic power generation facility. As can be seen from FIG. 14, it was confirmed that the second photovoltaic power generation output can be estimated with high accuracy by the power generation output estimation method of the present embodiment.

FIG. 15 is a diagram showing an example of verification results of the power generation output estimation method of the present embodiment when the estimation accuracy is not verified. In FIG. 15, similarly to FIG. 14, the broken line indicates the estimated value of the second photovoltaic power generation output estimated by the estimation method described in the present embodiment, and the solid line indicates the actually measured value. In the example shown in FIG. 15, the same verification as in the example shown in FIG. 14 is performed on another day. However, in the example shown in FIG. 15, the estimation accuracy using the self-coefficient λ is not verified. As described above, if the estimation accuracy using the self-coefficient λ is not verified, the estimation accuracy of the second photovoltaic power generation output may decrease.

Here, the factors that reduce the estimation accuracy will be described. FIG. 16 is a diagram showing an example of the verification result on the day before the day when the verification result is shown in FIG. 14, and FIG. 17 shows an example of the verification result on the day before the day when the verification result is shown in FIG. It is a figure. In the verification showing the results in FIGS. 14 and 15, the coefficient α is estimated the day before as described above. As can be seen by comparing FIGS. 16 and 17, the day corresponding to FIG. 17 has a much weaker illuminance intensity than the day corresponding to FIG. Therefore, the first photovoltaic power generation output used for calculating the coefficient α used to obtain the estimated value shown in FIG. 15, that is, the first photovoltaic power generation output on the day corresponding to FIG. 17 is also very small. .. Therefore, since the Cov _t [P _PV1 (t), P _PV1 (t + τ _s )] becomes small, the value of the Cov _t [P _PV1 (t), _PL2 (t)] is relatively close to 0. It cannot be ignored, and the assumptions made when deriving the above-mentioned equation (5) do not hold, and the estimation accuracy of the coefficient α decreases. As a result, in the example shown in FIG. 15, the estimation accuracy of the second photovoltaic power generation output is lowered. In the present embodiment, the self-coefficient λ is used, and when | λ-1 | exceeds the threshold value u, the estimated value of the coefficient α is not adopted. Therefore, the estimation accuracy as shown in FIG. 15 is achieved. It is possible to prevent the decrease.

FIG. 18 is a diagram showing an example of a scatter diagram of | λ-1 | and the estimation error of the second photovoltaic power generation output. In FIG. 18, the horizontal axis indicates | λ-1 |, and the vertical axis indicates the root-mean squared error (RMSE) of the estimated value of the second photovoltaic power generation output. In the scatter plot shown in FIG. 18, the same verification as in the examples shown in FIGS. 14 and 15 was performed for a plurality of days, and RMSE and | λ-1 | for the measured values of the estimated values were calculated for each of the plurality of days. The results obtained are shown. On days when the error is large and the estimation accuracy is low, | λ-1 | also becomes large as shown in FIG. 18, and on days when the estimation accuracy is high, | λ-1 | becomes a value close to 0. Therefore, when | λ-1 | exceeds the threshold value u, it is possible to prevent a decrease in estimation accuracy by not calculating the coefficient α. In the example shown in FIG. 18, the threshold value u is set to 0.5. The threshold value u is determined so that the estimation accuracy of the second photovoltaic power generation output satisfies a desired accuracy, that is, a predetermined accuracy. As a result, when the determination in step S6 shown in FIG. 4 determines Yes, the estimation accuracy of the second photovoltaic power generation output satisfies the desired accuracy.

The method for determining the threshold value u is not particularly limited. For example, the estimation accuracy verification unit 140 stores the RMSE and | λ-1 | obtained by the verification in the storage unit 180, and the display unit 160 stores them. Is displayed as a scatter plot. That is, the display unit 16 displays the scatter plot illustrated in FIG. The RMSE at this time is obtained by updating the coefficient α and estimating the power generation output in all cases regardless of the value of the self-coefficient λ without verifying the estimation accuracy using | λ-1 |. The result is. Then, while the user confirms this scatter plot, the threshold value u is determined so that the group having a large RMSE and the group having a small estimation error can be appropriately separated, and the determined threshold value u is set via the input reception unit 170. input. For example, the user adjusts the threshold value u by moving the straight line of the threshold value u shown in FIG. 18 left and right with a mouse or the like, which is an example of the input receiving unit 170. Then, for example, a confirmation button is displayed on the scatter diagram illustrated in FIG. 18, and the user moves the straight line indicating the threshold value u to a desired position and presses the confirmation button so that the threshold value u is set. You may do it. The consumer used to determine the threshold value u may be the first consumer measuring the photovoltaic power generation output, or a measuring instrument may be temporarily installed in the second consumer to generate photovoltaic power generation. It may be possible to measure the output. Furthermore, the estimated value of the photovoltaic power generation output installed in the second consumer may be used. The method of setting the threshold value u by the user is not limited to this example. Further, a simulation result or the like may be used instead of the verification result.

The threshold value u may be calculated by the photovoltaic power generation output estimation device 100 or another device by calculation such as machine learning. For example, the estimation accuracy verification unit 140 may determine the threshold value u by machine learning using the RMSE obtained by the verification and | λ-1 |. That is, the estimation accuracy verification unit 140 acquires a plurality of errors (RMSE) of the estimated value of the second photovoltaic power generation output estimated by the estimation unit 150 with respect to the measured value of the second photovoltaic power generation output, and uses the error and the error. Even if the threshold is calculated by machine learning so that the error when | λ-1 | is less than the threshold u is equal to or less than the specified value using a plurality of data sets composed of the corresponding self-factors λ. good. As machine learning, the maximum entropy method, a decision tree, or the like may be used, or the threshold value u may be obtained by classifying the results by a clustering method, and any method may be used. The display unit 160 displays the above-mentioned scatter diagram, that is, a scatter diagram showing the relationship with the error of the estimated value of the second photovoltaic power generation output estimated by the estimation unit 150 with respect to the measured value of the second photovoltaic power generation output. May be good. Further, the display unit 160 may superimpose the threshold value u calculated by calculation on the scatter plot and display it as shown in FIG. 18, for example. This allows the user to confirm whether the threshold value u is set correctly. Further, the user may be able to change the threshold value u calculated by calculation. For example, the threshold value u may be changed by shifting the straight line indicating the threshold value u calculated by calculation to the left or right, as in the case of setting the threshold value u by the user.

In the above example, the estimation accuracy was verified using the self-coefficient λ, but the coefficient λ'corresponding to the coefficient α was calculated using the residual demand of the photovoltaic power generation facility 202 other than the first photovoltaic power generation facility 202. , The coefficient λ'may be used instead of the self-coefficient λ to determine whether or not it is suitable for estimating the coefficient α. The coefficient λ'is also a verification coefficient used for verification of estimation accuracy. For example, the solar power generation facility 202 of the consumer 200-1 shown in FIG. 1 is the first solar power generation facility, and the solar power generation facility 202 of the consumer 200-2 shown in FIG. 1 is the second solar power generation facility. It is assumed that it is a facility. At this time, the photovoltaic power generation facility 202 of the consumer 200-3 is also measured by the measuring instrument 206, and the power generation output of the photovoltaic power generation facility 202 of the consumer 200-3 is also stored in the storage unit 180, and the photovoltaic power generation output data 182. It is stored as. The photovoltaic power generation facility 202 of the consumer 200-3 is the third photovoltaic power generation facility, the consumer 200-3 is the third consumer, and the residual demand, which is the apparent power consumption of the third consumer, is P ₃ ( If t), the coefficient λ'can be calculated by the following equation (8).

Since the coefficient λ'is a coefficient for converting the first photovoltaic power generation output into the third photovoltaic power generation output, which is the power generation output of the third photovoltaic power generation facility, the coefficient λ'is added to the first photovoltaic power generation output. Multiplying gives an estimate of the third PV output. On the other hand, the third photovoltaic power generation output is measured by the measuring instrument 206 in the consumer 200-3, and ideally, the measured data and the estimated value match. Therefore, the coefficient λ'is ideally a value obtained by dividing P _PV3 (t), which is the third photovoltaic power generation output (measurement data), by the first photovoltaic power generation output P _PV1 (t). Therefore, in step S6 of FIG. 4, it is an absolute value of the result of subtracting the value obtained by subtracting the value obtained by dividing the third photovoltaic power generation output by the first photovoltaic power generation output from the verification coefficient | λ'-P _PV3 ( By determining whether or not t) / P _PV1 (t) | is equal to or less than the threshold value u, it is possible to prevent the adoption of the coefficient α when the estimation accuracy is low, as in the case of using the self-coefficient λ. be able to. Since a plurality of P _PV3 (t) / P _PV1 (t) are obtained in the first period, the average value may be used or the median value may be used. Further, the regression coefficients ρ of P _PV3 (t) and P _PV1 (t) that can be calculated by the following equation (9) may be used.

In this way, it is possible to determine whether or not the estimation accuracy of the second photovoltaic power generation output is high by using the third photovoltaic power generation output, which is the power generation output of the third photovoltaic power generation facility. The second photovoltaic power generation facility and the third photovoltaic power generation facility are installed within a predetermined distance from the first photovoltaic power generation facility.

That is, the estimation unit 150 of the present embodiment first sets the second photovoltaic power generation output, which is the power output of the second photovoltaic power generation facility installed within a predetermined distance from the first photovoltaic power generation facility. It is estimated using the first photovoltaic power generation output, which is the power generation output of the photovoltaic power generation facility, and the residual demand, which is the apparent power consumption of the consumer in which the second photovoltaic power generation facility is installed. Then, the estimation accuracy verification unit 140 sees the power output of the third photovoltaic power generation facility installed within a predetermined distance from the first photovoltaic power generation facility and the apparentness of the consumer in which the third photovoltaic power generation facility is installed. It is determined whether or not the estimation accuracy of the second photovoltaic power generation output satisfies the predetermined accuracy by using the residual demand which is the above power consumption.

Specifically, in the first covariance calculation unit 120, the time series data of the first photovoltaic power generation output and the third photovoltaic power generation facility in the first period are installed for the calculation of the coefficient λ'which is the verification coefficient. The third covariance, which is the covariance with the time-series data of the residual demand, which is the apparent power consumption of the consumer, is calculated. At this time, the delay time is also calculated in the same manner as when the coefficient α is calculated. This delay time is the time for the solar radiation fluctuation to propagate between the installation point of the first photovoltaic power generation facility and the installation point of the third photovoltaic power generation facility, and is also called the second delay time. Using the second delay time, the second covariance calculation unit 130 uses only the time-series data of the first solar power output in the first period and the second delay time from the first period, as in the case of calculating the coefficient α. The fourth covariance, which is the autocovariance with the time-series data of the first solar power output in the shifted third period, is calculated. The third covariance and the fourth covariance correspond to the first covariance and the second covariance in the calculation of the coefficient α when the third photovoltaic power generation facility is used as the second photovoltaic power generation facility, respectively. Therefore, the third covariance can be referred to as the first covariance, and the fourth covariance can be referred to as the second covariance. The estimation accuracy verification unit 140 calculates a coefficient λ'which is a verification coefficient corresponding to an estimated value of the ratio between the first photovoltaic power generation output and the third solar power generation output using the third co-dispersion and the fourth co-dispersion. Then, whether or not the estimation accuracy of the second photovoltaic power generation output satisfies the predetermined accuracy by using the coefficient λ'and the ratio of the third photovoltaic power generation output to the first photovoltaic power generation output or the regression coefficient. Is determined. Then, when the estimation accuracy verification unit 140 determines that the estimation accuracy of the second photovoltaic power generation output satisfies the determined accuracy, the estimation unit 150 estimates the second photovoltaic power generation output using the coefficient α. If it is determined by the estimation accuracy verification unit 140 that the estimation accuracy of the second photovoltaic power generation output does not satisfy the specified accuracy, the initial value of the coefficient or the coefficient α calculated in the past is used for the second photovoltaic power generation. Estimate the power output. The third photovoltaic power generation facility may be the first photovoltaic power generation facility or may be a photovoltaic power generation facility different from the first photovoltaic power generation facility.

The example using the self-coefficient λ described with reference to FIG. 4 is the case where the third photovoltaic power generation facility is the first photovoltaic power generation facility, and the self-coefficient λ is an example of the coefficient λ ′. Here, an example has been described in which the first photovoltaic power generation facility is the photovoltaic power generation facility 202 of the consumer 200-1 and the third photovoltaic power generation facility is the photovoltaic power generation facility 202 of the consumer 200-3. On the contrary, the first photovoltaic power generation facility may be the photovoltaic power generation facility 202 of the consumer 200-3, and the third photovoltaic power generation facility may be the photovoltaic power generation facility 202 of the consumer 200-1. When the third photovoltaic power generation facility and the first photovoltaic power generation facility are the same, the second delay time is 0. Further, when the third photovoltaic power generation facility is different from the first photovoltaic power generation facility, the second delay time may be approximated to 0. Further, when the first photovoltaic power generation facility is the photovoltaic power generation facility 202 of the consumer 200-1, the photovoltaic power generation subject to another total purchase contract within a predetermined distance from the first photovoltaic power generation facility. If there is equipment, the solar power generation equipment may be used as a third solar power generation equipment. Furthermore, when estimating the second photovoltaic power generation output using the total amount of the first photovoltaic power generation output of a plurality of consumers, the estimation accuracy is estimated using the third photovoltaic power generation output and the residual demand of one third consumer. Or, the estimation accuracy may be verified using the total amount of the third photovoltaic power generation output of the plurality of third consumers and the total amount of the residual demand.

As described above, the example described with reference to FIG. 4 is when the third photovoltaic power generation facility is the first photovoltaic power generation facility, that is, when the third photovoltaic power generation facility and the first photovoltaic power generation facility are the same. Therefore, the above-mentioned determination step that the photovoltaic power generation output estimation program of the present embodiment causes the computer to execute is apparently the power output of the third photovoltaic power generation facility and the consumer in which the third photovoltaic power generation facility is installed. This is an example of a step of determining whether or not the estimation accuracy of the second photovoltaic power generation output satisfies the predetermined accuracy by using the residual demand which is the power consumption.

In the above example, the second photovoltaic power generation output is estimated using the first photovoltaic power generation output, which is the measurement data. However, after the estimation of the second photovoltaic power generation output, the estimated second photovoltaic power generation is generated. The output may be used as the first photovoltaic power generation output to estimate the power generation output of yet another photovoltaic power generation facility. By repeating this, for example, even if there is no photovoltaic power generation facility 202 whose power generation output is measured within a predetermined distance from the photovoltaic power generation facility 202 to be estimated, the solar power to be estimated is to be estimated. The power output of the power generation facility 202 can be estimated.

FIG. 19 is a schematic diagram showing an example of the order of estimation of power generation output. In FIG. 19, the photovoltaic power generation equipment 202 of the consumer 200-i (i = 1, 2, ..., 10) is shown as PVi. PV1 and PV3 shown in the solid line frame are the photovoltaic power generation facilities 202 of the consumers 200-1 and 200-3, respectively, and the power generation output is measured. PV2 and the like shown by the broken line frame are photovoltaic power generation facilities 202 such as the consumer 200-2 whose power generation output is not measured. The ranges 301 and 302 are the same as the example shown in FIG. The black arrow shown in FIG. 19 indicates the estimation using the measurement data of the power generation output, and the white arrow indicates the estimation of the power generation output using the estimation result of the power generation output of the other photovoltaic power generation equipment 202. As shown in FIG. 5, PV2, PV5, and PV6 are estimated for power generation output using the measurement data of the power generation output of PV1, and PV4, PV7, and PV8 generate power using the measurement data of the power generation output of PV3. The output is estimated. After that, the estimated value of the power generation output of PV6 is used as the first photovoltaic power generation output, the power generation output of PV8, which is the second photovoltaic power generation facility, is estimated, and the estimated value of the power generation output of PV4 is used as the first photovoltaic power generation. The power generation output of PV10, which is the second photovoltaic power generation facility, is estimated by using it as the power generation output. Further, the estimated value of the power generation output of PV8 is used as the first solar power generation output to estimate the power generation output of PV9, which is the second solar power generation facility.

PV8, PV9, and PV10 are not located within a predetermined distance from PV1 and are not located within a predetermined distance from PV3. However, by sequentially estimating the power generation output using the estimated value, PV8 , PV9, PV10 can be estimated.

Further, the measurement data of the smart meters 203, 204, 205 may be an integrated value such as an integrated value for 30 minutes, or may be an instantaneous value. For example, in the calculation process of the coefficient α, the integrated value is used, and the instantaneous value of the first photovoltaic power generation output may be used when estimating the second photovoltaic power generation output using the coefficient α at the time of estimation.

As described above, the photovoltaic power generation output estimation device 100 of the first embodiment is the power output of the second photovoltaic power generation facility installed within a predetermined distance from the first photovoltaic power generation facility. 2 The estimation unit 150 that estimates the photovoltaic power generation output using the first photovoltaic power generation output, which is the power output of the first photovoltaic power generation facility, and the residual demand of the consumer in which the second photovoltaic power generation facility is installed. Be prepared. Therefore, according to the photovoltaic power generation output estimation device 100 of the first embodiment, if there is at least one photovoltaic power generation facility capable of acquiring the photovoltaic power generation output, the photovoltaic power generation facility is referred to as the first photovoltaic power generation facility. By doing so, it is possible to individually estimate the photovoltaic power output of other desired photovoltaic power generation equipment without installing a photovoltaic meter and a measuring instrument having high time resolution.

Further, the photovoltaic power generation output estimation device 100 includes the power output of the third photovoltaic power generation facility installed within a predetermined distance from the first photovoltaic power generation facility and the consumer in which the third photovoltaic power generation facility is installed. It is provided with an estimation accuracy verification unit 140 for determining whether or not the estimation accuracy of the second photovoltaic power generation output satisfies a predetermined accuracy by using the residual demand of. Therefore, if it is determined that the estimation accuracy does not satisfy the predetermined accuracy, countermeasures can be taken. An example of this measure is to use the initial value or the previously calculated coefficient α without adopting the calculated coefficient α, but the measure is not limited to this. For example, when it is determined that the estimation accuracy does not satisfy the predetermined accuracy, the second photovoltaic power generation output may be estimated by another estimation method. When it is determined that the estimation accuracy does not satisfy the predetermined accuracy, it is possible to prevent the estimation accuracy of the second photovoltaic power generation output from deteriorating by taking measures.

Embodiment 2.
Next, the photovoltaic power generation output estimation device 100 according to the second embodiment will be described. The functional configuration and hardware configuration of the photovoltaic power generation output estimation device 100 of the present embodiment are the same as those of the photovoltaic power generation output estimation device 100 of the first embodiment. However, the operations of the first covariance calculation unit 120 and the estimation accuracy verification unit 140 are partially different from those of the first embodiment. The components having the same functions as those of the first embodiment will be described with the same reference numerals as those of the first embodiment, and the description overlapping with the first embodiment will be omitted. Hereinafter, the differences from the first embodiment will be mainly described.

The photovoltaic power generation output estimation device 100 according to the second embodiment carries out the estimation process of the coefficient α by the procedure shown in FIG. 4 in the same manner as in the first embodiment, but in the second embodiment, the coefficient α in step S7. In the calculation process of, the determination of whether α is a positive value is added.

FIG. 20 is a flowchart showing an example of the calculation processing procedure of the coefficient α of the second embodiment. Steps S31 to S32 shown in FIG. 20 are the same as those in the first embodiment. In step S33a, the coefficient α is calculated in the same manner as in the first embodiment and step S33 (step S33a). At this point, the coefficient α is not updated. After calculating the coefficient α in step S33a, the estimation unit 150 determines whether or not the coefficient α is a positive value (step S34). Since the coefficient α is the ratio of the second photovoltaic power generation output to the first photovoltaic power generation output, it should be a positive value. If the coefficient α does not become a positive value, it is considered that the estimation accuracy of the coefficient α is lowered. Therefore, when the coefficient α is not a positive value (step S34 No.), the estimation unit 150 ends the process without updating the coefficient α. When the coefficient α is a positive value (step S34 Yes), the estimation unit 150 updates the coefficient α in the calculated data 185 of the storage unit 180 with the calculated coefficient α (step S35), and ends the process.

Further, in the above example, whether or not the coefficient α is larger than 0 is determined in step S34, but instead, in step S34, is the coefficient α larger than 0 and the rated capacity ratio × k or less? May be determined. The rated capacity ratio is the ratio of the rated capacity of the second photovoltaic power generation facility to the rated capacity of the first photovoltaic power generation facility. k can be larger than 1, for example, about 1.5, but is not limited to this value as long as it is appropriately determined by verification or the like. Since the coefficient α is a conversion coefficient for obtaining the second photovoltaic power generation output by multiplying the first photovoltaic power generation output, the accuracy of the coefficient α decreases when the coefficient α is significantly different from the rated ratio. It may be. Therefore, in step S34, it is determined whether the coefficient α is larger than 0 and the rated capacity ratio × k or less, and when the coefficient α is larger than 0 and the rated capacity ratio × k or less, step S35 is performed. Therefore, it is possible to prevent deterioration of the estimation accuracy of the second photovoltaic power generation output. Further, in step S34, it may be determined whether the coefficient α is the rated capacity ratio × k ₁ or more and the rated capacity ratio × k ₂ or less. k ₁ is less than 1 and k ₂ is greater than 1. For example, k ₂ is about 1.5 and k ₁ is about 0.5, but k ₁ and k ₂ are not limited to these values as long as they are appropriately determined by verification or the like.

That is, when the coefficient α is smaller than the value obtained by multiplying the rated ratio by the first value k ₁ , and when the coefficient α is larger than the value obtained by multiplying the rated ratio by the second value k ₂ . , The second solar power output may be estimated using the initial value of the coefficient or the coefficient α calculated in the past.

The operation of the present embodiment other than that described above is the same as that of the first embodiment. In the present embodiment, as in the first embodiment, the estimation accuracy is verified using the self-coefficient λ, and the estimation accuracy is also verified based on the value of the coefficient α. As a result, the same effect as that of the first embodiment can be obtained, and the estimation accuracy can be further improved.

Embodiment 3.
Next, the photovoltaic power generation output estimation device 100 according to the third embodiment will be described. The functional configuration and hardware configuration of the photovoltaic power generation output estimation device 100 of the present embodiment are the same as those of the photovoltaic power generation output estimation device 100 of the first embodiment. However, the operation of the estimation unit 150 is partially different from that of the first embodiment. The components having the same functions as those of the first embodiment will be described with the same reference numerals as those of the first embodiment, and the description overlapping with the first embodiment will be omitted. Hereinafter, the differences from the first embodiment will be mainly described.

FIG. 21 is a flowchart showing an example of an estimation processing procedure for the coefficient α by the photovoltaic power generation output estimation device 100 of the present embodiment. Step S1 is the same as step S1 shown in FIG. 4 of the first embodiment. After step S1, the first covariance calculation unit 120 sets a plurality of first periods (step S2a). Then, as in the first embodiment, steps S3 to S6 are performed for one of the plurality of first periods. In step S7a, the coefficient α is calculated as in the first embodiment, but the calculated coefficient α is temporarily stored in the estimation unit 150 or the storage unit 180, and the coefficient α in the calculated data of the storage unit 180 is stored. Is not updated at this point. After step S7a, the first covariance calculation unit 120 determines whether or not the coefficients α of all the first periods have been calculated (step S9), and the first period in which the coefficients α have not been calculated. If there is (step S9 No), one of the first periods in which the coefficient α is not calculated is selected. After that, the process from step S3 is repeated for the selected first period.

When all the coefficients α of the first period are calculated (step S9 Yes), the first covariance calculation unit 120 instructs the estimation unit 150 to calculate the representative value, and the estimation unit 150 is the representative of the coefficient α. A value is calculated, and the stored coefficient α is updated with the calculated value (step S10). Specifically, the estimation unit 150 calculates a representative value of the coefficient α by using the coefficient α for each first period that is temporarily held. The representative value of the coefficient α is preferably the median value of the plurality of calculated coefficients α, but may be a statistic such as an average value or a mode value. Step S8 is the same as that of the first embodiment.

In addition, before step S7a, in step S6, it is determined whether the first period is suitable for the calculation of the coefficient α, that is, whether the estimation accuracy satisfies the desired accuracy. If it is determined that the first period is not suitable for calculating the coefficient α, the coefficient α is not calculated. Therefore, in this case, the number of calculated coefficients α is smaller than the number of the first period set in step S2a. Become. For example, it is assumed that 10 first periods are set in step S2a, and it is determined that one of the periods is not suitable for calculating the coefficient α. In this case, since the estimation unit 150 holds the nine coefficients α in step S10, the estimation unit 150 determines the representative value using the nine coefficients α. In this way, a plurality of first periods are set, the coefficient α corresponding to the case where the first period is suitable for the calculation of the coefficient α is calculated, and the representative value is calculated and calculated using the calculated coefficient α. The representative value is adopted. Therefore, it is possible to reduce the decrease in the estimation accuracy of the second photovoltaic power generation output, which is caused by the influence of the coefficient α, which rarely occurs, that is, the deviation value of the coefficient α, that is, the estimation error is large.

In this way, a plurality of first periods may be set, and at least one may be set. That is, the first covariance calculation unit 120 determines the time series data of the first photovoltaic power generation output and the residual demand of the second consumer in at least one first period before the estimation time of the second photovoltaic power generation output. The first covariance, which is the covariance with the time series data, may be calculated. The second covariance calculation unit 130, the estimation unit 150, and the estimation accuracy verification unit 140 carry out the operation described in the first embodiment for each first period. When a plurality of first periods are set, the estimation unit 150 calculates a representative value of the coefficient α using the coefficient α corresponding to the first period in which | λ-1 | is determined to be less than the threshold value u. The second photovoltaic output is estimated using the representative value.

It should be noted that the second embodiment and the present embodiment may be combined so that the coefficient α is not updated when the coefficient α is not a positive value. In this case, the estimation unit 150 may hold the coefficient α only when the coefficient α is a positive value, and may obtain a representative value of the coefficient α held in step S10, or the coefficient α is a negative value. However, it may be temporarily held and the representative value may be obtained by using the coefficient α of the positive value in step S10. That is, the estimation unit 150 calculates a representative value of the coefficient α by using the coefficient α of a positive value among the coefficients α corresponding to the first period in which | λ-1 | is determined to be less than the threshold u. The second photovoltaic power generation output will be estimated using the representative value. Similarly, it may be combined with the determination of whether the coefficient α is larger than 0 and the rated capacity ratio × k or less, or the coefficient α is the rated capacity ratio × k ₁ or more and the rated capacity ratio × k ₂ or less. It may be combined with the determination. Further, in FIG. 21, an example of determining whether or not to adopt the coefficient α by using the self-coefficient λ is shown, but as described in the first embodiment, the coefficient α is adopted by using the coefficient λ'. It may be determined whether or not.

The configuration shown in the above embodiments is an example, and can be combined with another known technique, can be combined with each other, and does not deviate from the gist. It is also possible to omit or change a part of the configuration.

10 Photovoltaic power generation output estimation system, 100 Photovoltaic power generation output estimation device, 110 Data acquisition unit, 120 1st covariance calculation unit, 130 2nd covariance calculation unit, 140 estimation accuracy verification unit, 150 estimation unit, 160 display unit , 170 Input reception unit, 180 storage unit, 200-1 to 200-4, 200-m, 200-n consumer, 201 load, 202 photovoltaic power generation equipment, 203-205 smart meter, 206 measuring instrument.

Claims (18)

The second photovoltaic power generation output, which is the power generation output of the second photovoltaic power generation facility installed within a predetermined distance from the first photovoltaic power generation facility, is the power generation output of the first photovoltaic power generation facility. An estimation unit that estimates using the photovoltaic power generation output and the residual demand, which is the apparent power consumption of the consumer in which the second photovoltaic power generation facility is installed,
It is the power generation output of the third photovoltaic power generation facility installed within the predetermined distance from the first photovoltaic power generation facility and the apparent power consumption of the consumer in which the third photovoltaic power generation facility is installed. An estimation accuracy verification unit that determines whether or not the estimation accuracy of the second photovoltaic power generation output satisfies the specified accuracy using the residual demand, and
A photovoltaic power generation output estimation device characterized by being equipped with. Time-series data of the first photovoltaic power generation output and residual demand of the consumer in which the second photovoltaic power generation facility is installed in at least one first period before the estimation time of the second photovoltaic power generation output. The first covariance calculation unit that calculates the first covariance, which is the covariance with the time series data,
By self-covariance between the time-series data of the first photovoltaic power generation output in the first period and the time-series data of the first photovoltaic power generation output in the second period deviated by the first delay time from the first period. A second covariance calculation unit that calculates a certain second covariance,
Equipped with
The first delay time is the time for the solar radiation fluctuation to propagate between the first photovoltaic power generation facility and the second photovoltaic power generation facility.
The estimation unit calculates a coefficient which is an estimated value of the ratio between the first photovoltaic power generation output and the second photovoltaic power generation output by using the first co-dispersion and the second co-dispersion, and calculates the coefficient. The second photovoltaic power generation output is estimated by multiplying the first photovoltaic power generation output whose delay time deviates from the estimation time point.
The photovoltaic power generation according to claim 1, wherein the estimation accuracy verification unit determines whether or not the estimation accuracy of the second solar power generation output satisfies a predetermined accuracy for each first period. Output estimator. The first covariance calculation unit is a combination of the time-series data of the first photovoltaic power generation output and the time-series data of the residual demand of the consumer in which the third photovoltaic power generation facility is installed in the first period. Calculate the third covariance, which is the variance,
The second covariance calculation unit is the time-series data of the first photovoltaic power generation output in the first period and the first photovoltaic power generation output in the third period deviated from the first period by the second delay time. Calculate the fourth covariance, which is the autocovariance with the time series data,
The second delay time is the time for the solar radiation fluctuation to propagate between the first photovoltaic power generation facility and the third photovoltaic power generation facility.
The estimation accuracy verification unit uses the third co-dispersion and the fourth co-dispersion to compare the first photovoltaic power generation output with the third photovoltaic power generation output, which is the power generation output of the third photovoltaic power generation facility. The verification coefficient, which is an estimated value of, is calculated, and the ratio or regression coefficient of the third photovoltaic power generation output to the first photovoltaic power generation output is used to estimate the second photovoltaic power generation output. Judging whether the accuracy meets the specified accuracy,
When the estimation unit determines that the estimation accuracy of the second photovoltaic power generation output satisfies the predetermined accuracy by the estimation accuracy verification unit, the estimation unit estimates the second photovoltaic power generation output using the coefficient. If the estimation accuracy verification unit determines that the estimation accuracy of the second photovoltaic power generation output does not satisfy the determined accuracy, the first value of the coefficient or the coefficient calculated in the past is used. 2. The photovoltaic power generation output estimation device according to claim 2, wherein the photovoltaic power generation output is estimated. The coefficient is a value obtained by dividing the first covariance by the second covariance and multiplying by -1.
The photovoltaic power generation output estimation device according to claim 3, wherein the verification coefficient is a value obtained by dividing the third covariance by the fourth covariance and multiplying by -1. 3. The estimation unit is characterized in that, when the coefficient is not a positive value, the second photovoltaic power generation output is estimated using the initial value of the coefficient or the coefficient calculated in the past. Or the photovoltaic power generation output estimation device according to 4. The estimation unit determines the initial value of the coefficient or in the past when the coefficient is smaller than the value obtained by multiplying the rated ratio by the first value, and when the coefficient is larger than the value obtained by multiplying the rated ratio by the second value. Using the calculated coefficient, the second photovoltaic power generation output is estimated.
The rated ratio is the ratio of the rated capacity of the second photovoltaic power generation facility to the rated capacity of the first photovoltaic power generation facility, the first value is 0 or more and less than 1, and the second value is 1. The photovoltaic power generation output estimation device according to claim 3 or 4, characterized in that it is larger. The second estimation accuracy verification unit is the second when the absolute value of the result of subtracting the value obtained by subtracting the value obtained by dividing the third photovoltaic power generation output by the first photovoltaic power generation output from the verification coefficient is equal to or less than the threshold value. It is determined that the estimation accuracy of the photovoltaic power generation output satisfies the defined accuracy, and when the absolute value is equal to or less than the threshold value, it is determined that the estimation accuracy of the second photovoltaic power generation output does not satisfy the defined accuracy. The photovoltaic power generation output estimation device according to claim 4, wherein the photovoltaic power generation output is estimated. A plurality of the first periods are set,
The estimation unit calculates a representative value of the coefficient using the coefficient corresponding to the first period in which the absolute value is determined to be equal to or less than the threshold value, and uses the representative value to generate the second photovoltaic power generation. The photovoltaic power generation output estimation device according to claim 7, wherein the output is estimated. The estimation unit calculates a representative value of the coefficient by using the coefficient having a positive value among the coefficients corresponding to the first period in which the absolute value is determined to be equal to or less than the threshold value, and obtains the representative value. The photovoltaic power generation output estimation device according to claim 8, wherein the second photovoltaic power generation output is estimated by using the device. The estimation unit sets the second value to the rated ratio, which is larger than the value obtained by multiplying the rated ratio by the first value among the coefficients corresponding to the first period in which the absolute value is determined to be equal to or less than the threshold value. A representative value of the coefficient is calculated using the coefficient equal to or less than the multiplied value, and the second solar power output is estimated using the representative value.
The rated ratio is the ratio of the rated capacity of the second photovoltaic power generation facility to the rated capacity of the first photovoltaic power generation facility, the first value is 0 or more and less than 1, and the second value is 1. The photovoltaic power generation output estimation device according to claim 8, which is characterized by being larger. The estimation accuracy verification unit acquires a plurality of errors of the estimated value of the second photovoltaic power generation output estimated by the estimation unit with respect to the measured value of the second photovoltaic power generation output, and corresponds to the error and the error. It is characterized in that the threshold value is calculated by machine learning so that the error when the absolute value is equal to or less than the threshold value is equal to or less than a predetermined value by using a plurality of data sets composed of the verification coefficient. The photovoltaic power generation output estimation device according to any one of claims 7 to 10. A display unit that displays a scatter diagram showing the relationship between the verification coefficient and the error of the estimated value of the second photovoltaic power generation output estimated by the estimation unit with respect to the measured value of the second photovoltaic power generation output.
The photovoltaic power generation output estimation device according to any one of claims 7 to 11, wherein the solar power generation output estimation device is provided. The photovoltaic power generation output estimation device according to claim 12, wherein the display unit displays the threshold value superimposed on the scatter diagram. The photovoltaic power generation output estimation device according to any one of claims 3 to 13, wherein the one delay time and the second delay time are approximated to 0. The photovoltaic power generation output estimation device according to any one of claims 1 to 14, wherein the third photovoltaic power generation facility is the first photovoltaic power generation facility. The photovoltaic power generation output estimation device according to any one of claims 1 to 14, wherein the third photovoltaic power generation facility is a photovoltaic power generation facility different from the first photovoltaic power generation facility. The photovoltaic power generation output estimation device converts the second photovoltaic power generation output, which is the power generation output of the second photovoltaic power generation facility installed within a predetermined distance from the first photovoltaic power generation facility, into the first photovoltaic power generation. Estimated using the first photovoltaic power generation output, which is the power generation output of the facility, and the residual demand, which is the apparent power consumption of the consumer in which the second photovoltaic power generation facility is installed.
The demand that the photovoltaic power generation output estimation device installs the power output of the third photovoltaic power generation facility installed within the predetermined distance from the first photovoltaic power generation facility and the third photovoltaic power generation facility. A photovoltaic power generation output estimation method, characterized in that it is determined whether or not the estimation accuracy of the second photovoltaic power generation output satisfies a predetermined accuracy by using the residual demand which is the apparent power consumption of the house. The second photovoltaic power generation output, which is the power generation output of the second photovoltaic power generation facility installed within a predetermined distance from the first photovoltaic power generation facility, is the power generation output of the first photovoltaic power generation facility. A step of estimating using the photovoltaic power generation output and the residual demand, which is the apparent power consumption of the consumer in which the second photovoltaic power generation facility is installed, and
The power output of the third photovoltaic power generation facility installed within the predetermined distance from the first photovoltaic power generation facility and the residual power consumption that is the apparent power consumption of the consumer in which the third photovoltaic power generation facility is installed. A step of determining whether or not the estimation accuracy of the second photovoltaic power generation output satisfies the defined accuracy by using the demand and
A photovoltaic power generation output estimation program characterized by having a computer execute.

Images