JP7224640B2 - Abnormality determination method and apparatus for photovoltaic power generation device without using weather information - Google Patents

Description

The present invention relates to an abnormality detection technology for a photovoltaic power generation device.

For example, in Patent Document 1, even if the state of receiving solar energy changes, an abnormality detection method for reliably determining whether an energy conversion device such as a solar cell panel is abnormal can be performed. A surveillance system is disclosed. Specifically, the monitoring device of the abnormality monitoring system has a communication unit that acquires the energy output amount of each energy conversion device, and grouping information that groups each energy conversion device according to the solar energy receiving state. a storage unit for storing a database; based on the database, the output amount of each energy conversion device is classified into groups, and the output amounts of the energy conversion devices belonging to the same group are compared to determine the output amount of each energy conversion device A computing unit that determines whether or not each energy conversion device is abnormal based on the similarity, and determines that the energy conversion device determined to be abnormal is abnormal and a notification unit that notifies.

Further, Patent Literature 2 discloses a solar power generation device management system that manages the solar power generation device installed on the user's site. Specifically, the photovoltaic power generation device can supply the power generated in its operating state to the user in conjunction with the power wiring system and can reverse power flow to the power wiring system. Abnormal photovoltaic power generation devices are detected by collecting and comparing power generation data of a plurality of photovoltaic power generation devices installed in the same area.

Furthermore, Patent Literature 3 discloses detecting defects in any solar panel included in, for example, a mega-solar plant based on the Mahalanobis distance. However, it does not mean that special measures have been taken to calculate the Mahalanobis distance.

As described above, there are various techniques for determining whether or not there is an abnormality in a solar power generation device by using the power generation status of surrounding solar power generation devices, but there is a demand for higher detection accuracy. .

Japanese Patent Application Laid-Open No. 2004-138293 JP 2011-147340 A U.S. Patent Application Publication No. 2018/0366979A1

SUMMARY OF THE INVENTION Accordingly, one aspect of the present invention is to provide a novel technique for accurately detecting an abnormality in a photovoltaic power generation device.

The abnormality determination method according to the present invention includes: (A) based on the power generation status of another solar power generation device associated with a specific solar power generation device among a plurality of solar power generation devices based on geographical conditions, for a specific period of time; (B) calculating a first reference value based on the amount of power generated by the other photovoltaic power generation device for each day determined to be fine; (C) a step of calculating a second reference value by correcting the calculated first reference value based on the amount of power generated by a specific solar power generation device for days determined to be fine weather in a predetermined period; and (D) determining whether or not there is an abnormality in the specific solar power generation device based on the second reference value and the power generation amount of the specific solar power generation device.

According to one aspect, it becomes possible to accurately detect an abnormality in the photovoltaic power generation device.

FIG. 1 is a diagram showing an outline of a system according to an embodiment of the invention. FIG. 2 is a diagram showing a configuration example of an information processing apparatus according to this embodiment. FIG. 3 is a diagram showing a main processing flow according to this embodiment. FIG. 4 is a diagram showing a processing flow of preprocessing according to this embodiment. FIG. 5 is a diagram showing the processing flow of the fine weather day extraction processing according to the present embodiment. FIG. 6 is a diagram illustrating an example of extraction of fine weather days. FIG. 7 is a diagram for explaining neighboring houses. FIG. 8 is a diagram showing a processing flow of control power generation amount calculation processing according to the present embodiment. FIG. 9 is a diagram showing an example of extraction of fine weather days in an area including neighboring houses. FIG. 10 is a diagram for explaining correction of the control power generation amount. FIG. 11 is a diagram for explaining correction of the control power generation amount. FIG. 12 is a diagram for explaining correction of the control power generation amount. FIG. 13 is a diagram showing a processing flow of abnormality determination processing according to the present embodiment. FIG. 14 is a diagram for explaining abnormality determination. FIG. 15 is a block configuration diagram of a computer device.

FIG. 1 shows a system outline according to an embodiment of the present invention.

For example, a network 100 such as the Internet is connected to a terminal device in each house and an information processing device 200 that executes main processing according to the present embodiment. For example, the amount of power generated by the photovoltaic power generation device 311 installed in the house 310 is measured every hour, for example, and transmitted from the terminal device 312 to the information processing device 200 . The amount of power generated by the photovoltaic power generation device 321 installed in the house 320 is also measured every hour, for example, and transmitted from the terminal device 322 to the information processing device 200 . The terminal device 332 of the house 330 also measures the amount of power generated by the photovoltaic power generation device 331 every hour, for example, and transmits it to the information processing device 200 . It should be noted that the number of houses shown in FIG. 1 is an example, and the data of the amount of generated electric power is transmitted to the information processing apparatus 200 from a larger number of houses.

In this embodiment, solar power generation devices installed in a plurality of houses are targeted, but solar power generation devices installed in factories and other buildings as well as on the ground may be used. good too.

FIG. 2 shows a configuration example of the information processing apparatus 200. As shown in FIG. The information processing device 200 has a measurement data acquisition section 210 , a measurement data storage section 220 , an abnormality determination section 230 and a data storage section 240 .

The measurement data acquisition unit 210 receives the measurement result of the power generation amount from the terminal device of each house, and stores it in the measurement data storage unit 220 . Note that the measurement data storage unit 220 holds, for example, data on the capacity of the photovoltaic power generation device installed in each house, data on other houses associated with each house, and the like.

The abnormality determination unit 230 includes a preprocessing unit 231 , a fine weather extraction unit 232 , a reference calculation unit 233 , and a determination unit 234 , and stores data during processing in the data storage unit 240 .

The preprocessing unit 231 executes processing that is a prerequisite for processing performed by other processing units of the abnormality determination unit 230 . The fine weather extracting unit 232 executes a process of extracting a fine weather day for each house from the generated power amount obtained for that house. In this embodiment, weather data and other data other than the generated power amount are not used. The reference calculation unit 233 calculates a reference value (also referred to as a reference value or control power generation amount) for determining whether or not the power generation amount of a specific house indicates an abnormality. The determination unit 234 determines whether there is an abnormality based on the amount of power generated for a specific house and the reference value calculated for the specific house.

Next, details of the processing executed by the information processing apparatus 200 will be described with reference to FIGS. 3 to 14. FIG.

First, the preprocessing unit 231 of the abnormality determination unit 230 executes preprocessing ( FIG. 3 : step S1). Preprocessing will be described with reference to FIG.

From the measurement results stored in the measurement data storage unit 220, the preprocessing unit 231 calculates the amount of power generated per unit time [kWh] during peak hours (for example, from 11:00 to 12:00) for each house and each day. Identify (FIG. 4: step S21). Then, the preprocessing unit 231 converts the specified amount of generated power for each house and each day into the amount of generated power per unit capacity, and stores it in the data storage unit 240 (step S23). That is, an operation is performed to divide the amount of power generated per unit time by the capacity [kW] of the photovoltaic power generation device.

In the following processing, the generated power amount per unit capacity and unit time [kWh/kW] will be processed, but for the sake of simplification of explanation, the generated power amount will be simply referred to below. Also, the amount of electric power generated by house p on a certain day t is denoted as G _p,t [kWh/kW]. Also, t−1 represents the day before a certain day t, and t+1 represents the day after a certain day t.

Returning to the description of the processing in FIG. 3, next, the fine weather extraction unit 232 of the abnormality determination unit 230 executes the fine weather extraction processing (step S3). The fine weather day extraction process will be described with reference to FIG.

In general, the power generation amount G _p,t varies in the range of 0 to 1 depending on the weather conditions, the type of solar power generation device, and the installation angle and orientation. In addition, if a problem occurs in the solar power generation system, the amount of power generated will decrease, and the problem will be detected by that decrease. It is difficult to determine whether the output drop was due to cloud cover. Further, since a minor abnormality of the photovoltaic power generation device is easier to detect as the amount of power generated increases, it is preferable to make a judgment based on the amount of power generated on a sunny day. On the other hand, meteorological information such as global solar radiation does not include data such as solar radiation with such fine granularity, and it is not possible to detect cases such as local cloudiness.

Therefore, in the present embodiment, without using weather information such as total solar radiation, after specifying the fine weather day for each house based on the amount of power generated, the fine weather day specified for the neighboring houses is specified. shall identify a clear day for the area containing the residence and its neighboring residences.

The fine weather extraction process extracts the fine weather days of each house, and the process of FIG. 5 is executed for each house.

First, the fine weather day extraction unit 232 extracts the maximum power generation between the current day and the past n days for each day t in a predetermined period from the power generation amount G _p,t for each day stored in the data storage unit 240 A power amount G _max[tn,t],p is specified (step S31). That is, the maximum value is specified from [G _p,tn , G _p,t-n+1 , G _p,t-n+2 , . . . , G _p,t-1 , G _p,t ].

このｄ_cloudy,p,tは、晴天時に０となり、曇りや雨では１に近い値となる。 Then, the fine weather day extraction unit 232 calculates the degree of cloudiness d _cloudy,p,t for each day of the predetermined period (step S33). The degree of cloudiness d _cloudy,p,t is expressed as follows.

This d _cloudy,p,t becomes 0 when the weather is fine, and becomes a value close to 1 when it is cloudy or rainy.

Then, the fine weather day extracting unit 232 identifies days u on which the degree of cloudiness d _cloudy,p,t is less than the threshold (step S35). t-1 is the day before t, but u-1 is not the day before u, but the fine weather one day before u. Then the process returns to the calling process.

For example, assuming the weather of a certain house p in early May as shown in the upper part of FIG. 6, the lower part of FIG. That is, the days t ₀ (May 10), t ₀ -1, t ₀ -3, t ₀ -5, t ₀ -7, and t ₀ -9 are extracted. u ₀ , u ₀ −1, u ₀ −2, u ₀ −3, u ₀ −4, u ₀ −5.

By calculating the degree of cloudiness based on the maximum power generation amount in this way, the sunny days of each house can be extracted. Note that the threshold is set in advance.

Returning to the description of the processing in FIG. 3, the abnormality determination unit 230 identifies one unprocessed house from all houses (step S5). Then, the reference calculation unit 233 executes control power generation amount calculation processing (step S7). This processing will be described with reference to FIGS. 7 to 12. FIG.

In addition, from the geographical conditions such as latitude, longitude, and altitude where each house i is located, another house that is estimated to generate a similar amount of power is specified in advance, and stored in the measurement data storage unit 210, for example. store it. As schematically shown in FIG. 7, neighboring houses p ₀ to p ₅ included in the predetermined range X of the house i are extracted in advance and stored in the measurement data storage unit 220, for example. Note that the number of neighboring houses of 5 is merely an example, and a larger or smaller number may be extracted. Also, in this example, since the house _p6 is not included in the predetermined range X, it is treated as not being a neighboring house. Let U ^[i] denote the set of neighboring houses thus associated with house i, where U ^[i] does not include house i. Each house included in U ^[i] is represented as j.

In this embodiment, neighboring houses are extracted in advance, but they may be dynamically extracted based on geographical conditions or the like.

First, the reference calculation unit 233 extracts neighboring houses j of the house i specified in step S5 from, for example, the measurement data storage unit 220 ( FIG. 8 : step S41). Then, the reference calculation unit 233 calculates the degree of anomaly for each fine weather day for each neighboring house, and excludes sunny days for which the degree of anomaly is equal to or greater than the threshold (step S43).

Abnormal power generation is removed as much as possible before calculating a reference value for control power generation. In particular, this is because an abnormality may occur in a neighboring house.

Specifically, the Mahalanobis distance and the degree of anomaly (Hotelling T ² ) based thereon are calculated as in the following equations.

Note that the threshold is a preset value. Clear days of G _j,u thus extracted are excluded. In addition, in order to simplify the following description, fine weather days u' other than the fine weather days excluded by this step are again defined as u.

Thereafter, the reference calculation unit 233 determines whether or not the area (the area including the house i and the group of neighboring houses) is sunny based on the ratio (or number) of the neighboring houses that are sunny for each day of the predetermined period. to specify a fine day v (step S45).

A group of neighborhood houses that constitute U ^[i] and t is a clear day is denoted as U _t ^[i] . Then, U _t ^[i] ⊆ U ^[i] . Also, the number of houses forming U ^[i] is denoted as U ^[i] _count , and the number of houses forming U _t ^[i] is denoted as U ^[i] _count,t .

Then, step S45 is represented as follows.

This processing is schematically shown using FIG. As shown in the upper part of FIG. 9, assuming the weather in the early May of neighboring houses j ₀ to j ₃ , it can be seen that there are variations in the number of sunny days even in neighboring houses. Therefore, if U ^[i] _count,t is counted for each day and according to the formula (5), fine weather days in this area are as shown in the lower part of FIG. It does not match the sunny day of house i shown in the lower part of FIG. Therefore, let v be a clear day in the area. That is, the days t ₀ (May 10), t ₀ -1, t ₀ -3, t ₀ -4, t ₀ -7, and t ₀ -9 are extracted. v ₀ , v ₀ −1, v ₀ −2, v ₀ −3, v ₀ −4, v ₀ −5.

Then, the reference calculation unit 233 calculates the average amount of generated power for neighboring houses on sunny days v as the controlled amount of generated power (step S47).

That is, the following calculation is executed to calculate the controlled power generation amount _{G'control[i],v} .

FIG. 10 shows the time variation of the power generation amount of a house i (solid line a) and the time variation of the control power generation amount G′ _control[i],v obtained from neighboring houses of the house i (dotted line b). . For example, as shown in the circled area, a simple comparison between the amount of power generated by house i and the average amount of power generated by neighboring houses (= control power generation amount G' _control[i],v ) is too large. Even if the threshold value is devised, it is determined to be abnormal. However, it is not abnormal because it is a seasonal output fluctuation due to the direction and angle of the roof.

Therefore, the reference calculation unit 233 executes correction processing of the controlled power generation amount according to the power generation amount of the specified house i, and stores the processing result in the data storage unit 240 (step S49). Then the process returns to the calling process.

Specifically, the following calculations are executed.

In the expression (7), the sunny day v specified for the neighboring house is used as a reference instead of the sunny day u of the house i. This is because there is a possibility that the sunny day u of the house i is erroneously determined due to an abnormality.

In order to solve the problem described with reference to FIG. 10, the correction is made using equation (7). The second and third terms on the right side of equation (7) both represent moving averages, and the difference between these moving averages is used to correct the controlled power generation amount. Also, the period for calculating these moving averages is from vn to v-1. This is to avoid reflecting the influence of v, which may be out of order, in the correction.

If the moving average is calculated in the case of the example of FIG. 10, it will be as shown in FIG. In FIG. 11, the solid line c represents the moving average of the power generation amount of a house i, and the dotted line d represents the moving average of the control power generation amount G' _control[i],v . This moving average difference has a size indicated by an arrow in FIG. 11 . By subtracting this moving average difference from the control power generation G' _control[i],v , the control power generation excluding the effects of seasonal output fluctuations such as the roof orientation and angle can be calculated. G _control[i],v will be obtained.

In FIG. 12, the solid line e represents the power generation amount of a house i, and the dotted line f represents the control power generation amount G _control[i],v . In this way, if the effects of seasonal output fluctuations such as the direction and angle of the roof are removed, the control power generation amount G _control[i],v will fluctuate in the same way as the power generation amount of a certain house i. .

Returning to the description of the processing in FIG. 3, the determination unit 234 of the abnormality determination unit 230 executes abnormality determination processing (step S9). The abnormality determination process will be explained using FIG. 13 .

First, the determination unit 234 calculates the power generation abnormality degree of each fine day v (step S51). Here, the power generation anomaly degrees a'i _,v are expressed as follows.
a′ _i,v =G _i,v −G _control[i],v

Further, the determination unit 234 identifies an unprocessed sunny day v among all the sunny days v (step S53). Then, the determination unit 234 calculates the degree of abnormality based on the Mahalanobis distance for the identified fine day v from the degree of abnormality in power generation (step S55).

ここで、ａ'_{mean[v-n,v-1],i}は、ｖ－ｎからｖ－１までの期間についての異常度の平均を表す。式の形は、（３）と同様であるが、個数はｎ＋１ではなくｎとなる。また、ａ'_{stdev[v-n,v-1],i}は、ｖ－ｎからｖ－１までの期間についての異常度の標準偏差を表す。式の形は、（４）と同様であるが、個数はｎ＋１ではなくｎとなる。 The degree of anomaly based on the Mahalanobis distance is expressed as follows.

Here, a' _{mean[vn,v-1],i} represents the mean of the anomaly degrees for the period from vn to v-1. The form of the equation is similar to (3), but the number is n instead of n+1. Also, _{a'stdev[vn,v-1],i} represents the standard deviation of the degree of anomaly for the period from vn to v-1. The form of the equation is similar to (4), but the number is n instead of n+1.

Then, the determination unit 234 determines whether or not the degree of abnormality based on the Mahalanobis distance exceeds a threshold (step S57). The threshold is a preset value.

If the degree of abnormality based on the Mahalanobis distance does not exceed the threshold, the process proceeds to step S61. On the other hand, when the degree of abnormality based on the Mahalanobis distance exceeds the threshold, the fine day v identified in step S53 is set as abnormal (step S59).

Calculation of the degree of anomaly based on the Mahalanobis distance for the examples described in FIGS. 10 and 12 yields the results shown in FIG.

In the example of FIG. 14, the power generation amount of house i is represented by a fine dotted line, the controlled power generation amount is represented by a solid line, and the degree of abnormality based on the Mahalanobis distance is represented by a long dotted line. The degree of anomaly based on the Mahalanobis distance has a value represented by the vertical axis on the right side. In this example, at the beginning of May of year x+2, the degree of anomaly based on the Mahalanobis distance suddenly increases, and it is detected that an anomaly has occurred in house i at this point.

Then, the determination unit 234 determines whether or not there is an unprocessed fine day v (step S61). If there is an unprocessed fine day v, the process returns to step S53. On the other hand, when there is no more unprocessed sunny day v, the determination unit 234 outputs the determination result to an output device (for example, a display device or another computer) (step S63).

The determination result may include, for example, the data of the fine weather identified in step S59 or the degree of abnormality based on the Mahalanobis distance.

By doing so, it becomes possible to perform a highly accurate abnormality determination.

Returning to the process of FIG. 3, abnormality determination unit 230 determines whether or not there is an unprocessed house (step S11). If there is an unprocessed house, the process returns to step S5. On the other hand, if there is no unprocessed house, the process ends.

As described above, by correcting the control power generation amount with the difference between the moving average of the control power generation amount and the moving average of the power generation amount of house i, a more appropriate control power generation amount can be obtained, and the Mahalanobis distance By calculating the degree of anomaly based on this, more accurate anomaly detection can be performed.

Although the embodiment of the present invention has been described above, the present invention is not limited to this. For example, the processing flow is an example, and as long as the processing result does not change, the order of steps may be changed or multiple steps may be executed in parallel. The functional block configuration of FIG. 2 is an example, and may not match the module configuration of the program. Also, the information processing apparatus 200 may be implemented not by one computer but by a plurality of computers.

Also, in the correction processing of the control power generation amount, an example using a moving average has been shown, but a weighted moving average may also be used. Alternatively, a statistic may be defined to better reflect seasonal fluctuations, and correction may be made based on the statistic. Even in this case, the power generation amount for the predetermined period for the house i is used.

In the processing flow described above, it is assumed that abnormality determination is performed for a plurality of houses over a long period of time. , or may focus on a specific house on a specific day. In that case, it is sufficient to calculate only the data used in each case.

The information processing apparatus 200 described above is a computer apparatus, and as shown in FIG. A display control unit 2507 connected to 2509 , a drive device 2513 for a removable disk 2511 , an input device 2515 and a communication control unit 2517 for connecting to a network are connected via a bus 2519 . Note that the HDD may be a storage device such as a solid state drive (SSD). An operating system (OS) and an application program for performing processing in the embodiment of the present invention are stored in the HDD 2505 and read from the HDD 2505 to the memory 2501 when executed by the CPU 2503. . The CPU 2503 controls the display control unit 2507, the communication control unit 2517, and the drive device 2513 according to the processing content of the application program to perform predetermined operations. In addition, data in the middle of processing is mainly stored in the memory 2501, but may be stored in the HDD 2505 as well. In the embodiment of the present technology, an application program for performing the processing described above is stored and distributed in a computer-readable removable disk 2511 and installed from the drive device 2513 to the HDD 2505. It may be installed in the HDD 2505 via a network such as the Internet and the communication control unit 2517 . Such a computer device implements the various functions described above through the organic cooperation of hardware such as the CPU 2503 and memory 2501 described above and programs such as the OS and application programs. .

It should be noted that data used by executing the above-described processing is stored in a storage device such as the memory 2501 or the HDD 2505 regardless of whether it is in the middle of processing or processing results.

The embodiments described above are summarized as follows.

The abnormality determination method according to the present embodiment includes: (A) based on the power generation status of another solar power generation device associated with a specific solar power generation device among a plurality of solar power generation devices based on geographical conditions; (B) calculating a first reference value based on the amount of power generated by the other photovoltaic power generation device for each day determined to be sunny; and (C) calculating the second reference value by correcting the calculated first reference value based on the amount of power generated by the specific photovoltaic power generation device for the day determined to be fine in the predetermined period. and (D) determining whether or not there is an abnormality in the specific photovoltaic power generation device based on the second reference value and the amount of power generated by the specific photovoltaic power generation device.

As described above, by correcting the first reference value to calculate the second reference value and making a determination based on the second reference value, it is possible to accurately calculate the abnormality. Note that the amount of power generated is standardized.

In addition, in the step of calculating the second reference value described above, the first reference value is the average of the first reference values in the predetermined period and the specific sun for the day judged to be fine in the predetermined period The second reference value may be calculated by correcting the difference from the average amount of power generated by the photovoltaic power generation device. By doing so, it is possible to reflect seasonal fluctuations in the direction and angle of the roof in the first reference value. Note that the predetermined period is, for example, a certain period from the day before the day when the first reference value was calculated. As a result, it is possible to avoid reflecting the abnormality on the day when the first reference value was calculated in the correction.

Further, the above-described step of determining whether the weather is fine includes (a1) for each of the plurality of photovoltaic power generation devices, from the amount of power generated by the photovoltaic power generation device, and (a2) determining whether or not each day of the specific period is sunny based on the number or ratio of the solar power generation devices identified as sunny days among the other solar power generation devices. You may make it include a step. Clear days can be specified by removing cases where clouds are locally covered.

Further, the determination step includes (a21) specifying a fine weather day for each of the other photovoltaic power generation devices, from the fine weather days specified in the specifying step, based on a predetermined abnormality degree value related to the power generation amount. may include the step of This is because it is preferable to remove the influence of other photovoltaic power generation devices in which an abnormality has occurred.

In addition, in the step of calculating the first reference value described above, (b1) for each day determined to be fine, the amount of power generated by the other solar power generation device for which the day was identified as a fine day An average value may be calculated as the first reference value.

Furthermore, in the step of determining the presence or absence of an abnormality described above, (d1) the abnormality of the specific solar power generation device based on the Mahalanobis distance between the second reference value and the amount of power generated by the specific solar power generation device You may make it determine the presence or absence of. Accuracy of anomaly detection becomes higher.

A program can be created to cause a computer to perform the methods described above, and the program can be stored in various storage media.

In addition, the information processing apparatus that executes the above-described method may be realized by one computer or may be realized by a plurality of computers. Let's just call it a system.

100 network 200 information processing device 310, 320, 330 house

Claims (8)

a step of determining whether or not each day in a specific period is sunny based on the power generation status of other solar power generation devices associated with a specific solar power generation device among a plurality of solar power generation devices based on geographical conditions; ,
calculating a first reference value based on the amount of power generated by the other photovoltaic power generation device for each day determined to be fine;
calculating a second reference value by correcting the calculated first reference value based on the amount of power generated by the specific photovoltaic power generation device on days determined to be fine weather in a predetermined period; ,
determining whether or not there is an abnormality in the specific solar power generation device based on the second reference value and the amount of power generated by the specific solar power generation device;
and a computer-executed abnormality determination method. In the step of calculating the second reference value,
The first reference value is determined by the difference between the average of the first reference values in the predetermined period and the average amount of power generated by the specific solar power generation device on days determined to be fine in the predetermined period 2. The abnormality determination method according to claim 1, wherein the second reference value is calculated by correcting. The step of determining whether the weather is fine,
a identifying step of identifying, for each of the plurality of photovoltaic power generation devices, a sunny day at an installation location of the photovoltaic power generation device from the amount of power generated by the photovoltaic power generation device;
a determination step of determining whether or not each day of the specific period is sunny based on the number or ratio of the other solar power generation devices identified as sunny days;
The abnormality determination method according to claim 1 or 2, comprising: The determination step includes
4. The abnormality according to claim 3, further comprising the step of identifying a sunny day from the sunny days identified in the identifying step for each of the other photovoltaic power generation devices, based on a predetermined abnormality degree value related to the power generation amount. judgment method. In the step of calculating the first reference value,
5. The abnormality according to claim 3 or 4, wherein for each day determined to be fine, an average value of the amount of power generated by the other photovoltaic power generation devices for which the day is determined to be a fine day is calculated as a first reference value. judgment method. In the step of determining the presence or absence of the abnormality,
6. The presence or absence of an abnormality in the specific solar power generation device is determined based on the Mahalanobis distance between the second reference value and the amount of power generated by the specific solar power generation device. Abnormality determination method described. a step of determining whether or not each day in a specific period is sunny based on the power generation status of other solar power generation devices associated with a specific solar power generation device among a plurality of solar power generation devices based on geographical conditions; ,
calculating a first reference value based on the amount of power generated by the other photovoltaic power generation device for each day determined to be fine;
calculating a second reference value by correcting the calculated first reference value based on the amount of power generated by the specific photovoltaic power generation device on days determined to be fine weather in a predetermined period; ,
determining whether or not there is an abnormality in the specific solar power generation device based on the second reference value and the amount of power generated by the specific solar power generation device;
A program that causes a computer to run a means for determining whether or not each day in a specific period is sunny based on the power generation status of other solar power generation devices associated with a specific solar power generation device among a plurality of solar power generation devices based on geographical conditions; ,
means for calculating a first reference value based on the amount of power generated by the other photovoltaic power generation device for each day determined to be fine;
means for calculating a second reference value by correcting the calculated first reference value based on the amount of power generated by the specific photovoltaic power generation device on days determined to be fine weather in a predetermined period; ,
means for determining whether or not there is an abnormality in the specific solar power generation device based on the second reference value and the amount of power generated by the specific solar power generation device;
A system with

Images