Abnormality detection device, abnormality detection method, and abnormality detection program

Abstract

To appropriately detect an abnormality in a photovoltaic power generation system in response to a change in a solar radiation condition.SOLUTION: Based on output estimation information in each of a plurality of time zones included in one day, an estimation section calculates an estimation value of power generation output of a photovoltaic power generation system in each of the plurality of time zones from a measurement value of a solar radiation intensity in each of the plurality of time zones on a target day. Regarding each of the plurality of time zones on the target day, a detection section calculates a level of deviation between the estimation value of the power generation output and a measurement value of power generation output of the photovoltaic power generation system and, based on the level of deviation in each of the plurality of time zones, detects an abnormality in the photovoltaic power generation system.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to anomaly detection technology. [Background technology]

[0002] Regarding abnormality detection in solar power generation systems, a technique is known that accurately detects and reports defective photoelectric conversion elements (see, for example, Patent Document 1). A technique is also known that detects abnormalities in solar power generation panels with high accuracy (see, for example, Patent Document 2). A technique is also known that detects deterioration or abnormalities early based on a simple regression equation (see, for example, Patent Document 3). A technique is also known that appropriately determines equipment abnormalities in power generation equipment (see, for example, Patent Document 4). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 7-334767 [Patent Document 2] Japanese Patent Application Laid-Open No. 2014-93368 [Patent Document 3] Japanese Patent Application Laid-Open No. 2017-93275 [Patent Document 4] Japanese Patent Publication No. 2021-145509 Summary of the Invention [Problem to be solved by the invention]

[0004] With the techniques of Patent Documents 1 to 4, it is difficult to appropriately determine an abnormality in a solar power generation system depending on changes in solar radiation conditions caused by the installation status of the solar power generation system or the surrounding environment.

[0005] In one aspect, the present invention aims to appropriately determine an abnormality in a photovoltaic power generation system in response to changes in solar radiation conditions. [Means for solving the problem]

[0006] According to one embodiment, the anomaly detection device includes an estimation unit and a detection unit, and the estimation unit calculates an estimated value of the power generation output of the photovoltaic power generation system for each of a plurality of time periods from measured values ​​of solar radiation intensity for each of a plurality of time periods on a target day, based on output estimation information for each of a plurality of time periods included in a day.

[0007] The detection unit calculates the degree of deviation between the estimated power generation output and the measured power generation output of the solar power generation system for each of a plurality of time periods on the target day, and detects an abnormality in the solar power generation system based on the degree of deviation for each of the plurality of time periods. [Effects of the Invention]

[0008] According to one aspect, an abnormality in a photovoltaic power generation system can be appropriately determined in accordance with changes in solar radiation conditions. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a functional configuration diagram of an anomaly detection device according to an embodiment. [Figure 2] 10 is a flowchart of an abnormality detection process. [Figure 3] FIG. 1 is a configuration diagram of an anomaly detection system. [Figure 4] FIG. 1 is a configuration diagram of a solar power generation system. [Figure 5] FIG. 2 is a functional configuration diagram of the anomaly detection device. [Figure 6] FIG. 10 is a diagram showing performance data. [Figure 7] FIG. 10 is a diagram showing profile information. [Figure 8] FIG. 10 is a diagram showing the correlation between solar radiation intensity and generated power. [Figure 9] FIG. 10 is a diagram illustrating a deviation power conversion table. [Figure 10] FIG. 10 is a diagram showing the correspondence relationship between deviation power before conversion and deviation power after conversion. [Figure 11] 10 is a flowchart of a data selection process. [Figure 12]10 is a flowchart of a profile generation process. [Figure 13A] 10 is a flowchart (part 1) of an abnormality detection process. [Figure 13B] 10 is a flowchart (part 1) of an abnormality detection process. [Figure 14] FIG. 10 is a diagram showing the diagnosis results. [Figure 15] FIG. 2 is a hardware configuration diagram of an information processing device. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, embodiments will be described in detail with reference to the drawings.

[0011] 1 shows an example of the functional configuration of an anomaly detection device according to an embodiment. The anomaly detection device 101 in FIG. 1 includes an estimation unit 111 and a detection unit 112.

[0012] Fig. 2 is a flowchart showing an example of an anomaly detection process performed by the anomaly detection device 101 in Fig. 1. First, the estimation unit 111 obtains an estimated value of the power generation output of the photovoltaic power generation system for each of a plurality of time periods from the measured values ​​of the solar radiation intensity for each of the plurality of time periods on a target day, based on the output estimation information for each of the plurality of time periods included in one day (step 201).

[0013] Next, the detection unit 112 calculates the degree of discrepancy between the estimated power output and the measured power output of the solar power generation system for each of a plurality of time periods on the target day (step 202).Then, the detection unit 112 detects an abnormality in the solar power generation system based on the degree of discrepancy for each of the plurality of time periods (step 203).

[0014] The abnormality detection device 101 in FIG. 1 can appropriately determine whether an abnormality has occurred in the photovoltaic power generation system in accordance with changes in solar radiation conditions.

[0015] Fig. 3 shows an example of the configuration of an anomaly detection system including the anomaly detection device 101 of Fig. 1. The anomaly detection system of Fig. 3 includes a solar power generation system 301, a database server 302, and an anomaly detection device 303. The anomaly detection device 303 corresponds to the anomaly detection device 101 of Fig. 1.

[0016] The photovoltaic power generation system 301, the database server 302, and the anomaly detection device 303 can communicate with each other via a communication network 304. The communication network 304 is, for example, a wide area network (WAN).

[0017] Fig. 4 shows a configuration example of the solar power generation system 301 of Fig. 3. The solar power generation system 301 of Fig. 4 includes solar power generation units 411-1 to 411-N (N is an integer of 1 or more), junction boxes 412-1 to 412-N, a current collection box 413, a PCS (Power Conditioning System) 414, and a distribution board 415. The solar power generation system 301 further includes a measurement device 416, a pyranometer 417, a thermometer 418, a weather conversion box 419, and a communication device 420.

[0018] The solar power generation unit 411-i (i = 1 to N) includes strings 431-i-1 to 431-iM (M is an integer equal to or greater than 1). Each string 431-ij (j = 1 to M) includes a plurality of solar panels connected in series, and each solar panel includes a plurality of solar cells that convert light energy into electrical energy. The solar power generation unit 411-i is placed outdoors.

[0019] Each string 431-ij generates DC power by generating electricity according to the intensity of solar radiation and outputs the generated DC power to a junction box 412-i. Each junction box 412-i collects the DC power output from strings 431-i-1 to 431-iM and outputs it to a power collection box 413. The power collection box 413 collects the DC power output from junction boxes 412-1 to 412-N and outputs it to a PCS 414.

[0020] The PCS 414 converts the DC power output from the current collection box 413 into AC power and outputs it to the distribution board 415. The distribution board 415 distributes the AC power output from the PCS 414 to a distribution board, load equipment, etc. (not shown). The PCS 414 may be provided inside the connection box 412-i. When the PCS 414 is provided inside the connection box 412-i, the current collection box 413 can be omitted.

[0021] The measuring device 416 measures the DC power output from each connection box 412-i to the current collection box 413 at predetermined time intervals. Then, the measuring device 416 outputs the measured DC power as a measurement value of the power generated by the solar power generation unit 411-i to the communication device 420. The power generated by the solar power generation unit 411-i is an example of power generation output.

[0022] The measuring device 416 may measure the DC power output from each connection box 412-i to the power collection box 413 at time intervals shorter than the predetermined time interval. In this case, the measuring device 416 may obtain statistical values ​​of multiple DC powers at the predetermined time intervals and output the obtained statistical values ​​as measured values ​​of the power generated by the solar power generation unit 411-i to the communication device 420. The statistical values ​​may be an average value, a median value, a maximum value, a minimum value, or the like.

[0023] The pyranometer 417 measures the solar radiation intensity of sunlight at predetermined time intervals and outputs the measured value to a weather conversion box 419. The weather conversion box 419 converts the solar radiation intensity output from the pyranometer 417 from analog to digital (AD) and outputs the converted solar radiation intensity to a communication device 420 as a measured value of solar radiation intensity.

[0024] The thermometer 418 measures the outside air temperature at predetermined time intervals and outputs the measured value to a weather conversion box 419. The weather conversion box 419 converts the outside air temperature output from the thermometer 418 into an AD signal, and outputs the converted outside air temperature to the communication device 420 as a measured value of the outside air temperature.

[0025] The communication device 420 transmits the measured values ​​of the power generation, solar radiation intensity, and outside air temperature of each solar power generation unit 411-i to the database server 302 via the communication network 304. The database server 302 stores the measured values ​​of the power generation, solar radiation intensity, and outside air temperature of each solar power generation unit 411-i received from the solar power generation system 301.

[0026] The anomaly detection device 303 calculates the degree of anomaly that indicates the degree of decrease in the power generation of each solar power generation unit 411-i on the day to be diagnosed, using the measured values ​​of the power generation power and solar radiation intensity of each solar power generation unit 411-i stored in the database server 302. Then, the anomaly detection device 303 detects an anomaly in each solar power generation unit 411-i based on the anomaly degree of that solar power generation unit 411-i.

[0027] For example, when a solar cell deteriorates, the power generated by the solar power generation unit 411-i including that solar cell decreases. Therefore, by calculating the abnormality level indicating the degree of decrease in the power generated by the solar power generation unit 411-i, it is possible to detect an abnormality in the solar power generation unit 411-i.

[0028] However, the solar radiation conditions of each solar power generation unit 411-i change depending on the installation conditions or surrounding environment of the solar power generation unit 411-i.

[0029] For example, depending on the time of day, any solar cell included in any solar power generation unit 411-i may be shaded by surrounding trees, weeds, structures, etc. Examples of surrounding structures include utility poles, electric wires, and fences. When shaded, the power generated by the solar cell decreases. Furthermore, the irradiance intensity of sunlight incident on each solar cell included in each solar power generation unit 411-i changes depending on the season, and when the irradiance intensity weakens, the power generated by the solar cell decreases.

[0030] If a temporary drop in power generation is detected as an abnormality in the solar power generation unit 411-i without taking into account such changes in solar radiation conditions, an abnormality may be mistakenly detected even though the solar cell is not degraded.

[0031] Fig. 5 shows an example of the functional configuration of the anomaly detection device 303 in Fig. 3. The anomaly detection device 303 in Fig. 5 includes an acquisition unit 511, a generation unit 512, an estimation unit 513, a detection unit 514, an output unit 515, and a storage unit 516. The estimation unit 513 and the detection unit 514 correspond to the estimation unit 111 and the detection unit 112 in Fig. 1, respectively.

[0032] The anomaly detection device 303 performs a data selection process, a profile generation process, and an anomaly detection process. The profile generation process is performed after the data selection process, and the anomaly detection process is performed after the profile generation process.

[0033] In the data selection process, the acquisition unit 511 acquires the measured values ​​of the power generation power and solar irradiance intensity of each solar power generation unit 411-i for the month being analyzed in the year being analyzed from the database server 302 via the communication network 304. The acquisition unit 511 then generates performance data 521 including the acquired measurement values ​​and stores the data in the storage unit 516. The performance data 521 is generated for each solar power generation unit 411-i.

[0034] Next, the acquisition unit 511 edits the performance data 521 by deleting unnecessary measurement values ​​from the performance data 521 and adding to the performance data 521 measurement values ​​from the same month of a year prior to the year to be analyzed.

[0035] FIG. 6 shows an example of the performance data 521. In this example, the year to be analyzed is 2023, the month to be analyzed is January, and the predetermined time interval is 10 minutes. The performance data 521 in FIG. 6 shows the solar radiation intensity (kW / m 2 ) and generated power (kW). The solar radiation intensity and generated power represent measurements obtained from the database server 302.

[0036] For example, the measuring device 416 measures the DC power output from the connection box 412-i every minute, calculates the average of the most recent 10 measured values ​​every 10 minutes, and outputs the average to the communication device 420. Therefore, the generated power at each time represents the average value of the 10 generated power values. For example, the generated power at time 6:30 is the average value of the 10 generated power values ​​measured between time 6:21 and time 6:30.

[0037] The performance data 521 from January 2, 2023 to January 31, 2023 is also generated in the same format as in FIG.

[0038] Next, in the profile generation process, the generation unit 512 calculates an index S corresponding to the sunshine duration from the measured values ​​of the solar radiation intensity for one day or multiple days included in the performance data 521 of each solar power generation unit 411-i. Then, the generation unit 512 calculates a sunshine duration coefficient C1 for the analysis target month of the analysis target year using the index S for each solar power generation unit 411-i.

[0039] For example, when using the measured value of solar radiation intensity for one day, the generation unit 512 may determine that the time when the measured value of solar radiation intensity is equal to or greater than a threshold is a time when there is sunlight, and may use the number of times when there is sunlight as the index S. The threshold value of solar radiation intensity is 0.1 to 0.2 kW / m 2 It may be a value in the range

[0040] When using measured values ​​of solar radiation intensity over multiple days, the generating unit 512 may, for example, calculate the maximum value of the measured values ​​of solar radiation intensity over multiple days for each time period. The generating unit 512 may then determine that a time period when the maximum value is equal to or greater than a threshold is a time period with sunshine, and may use the number of times with sunshine as the index S.

[0041] The generation unit 512 calculates the sunshine duration coefficient C1 using, for example, the index S and the total number L of measured values ​​of solar radiation intensity in one day, according to the following formula.

[0042] C1=L 2 / S 2 (1)

[0043] For example, if the solar radiation intensity is measured every 10 minutes between 6:10 and 18:00, 72 solar radiation intensities will be recorded per day. If 60 of these solar radiation intensities are above the threshold, then L = 72 and S = 60. Therefore, the sunshine duration coefficient C1 is calculated as follows:

[0044] C1=(72×72) / (60×60)=1.44 (2)

[0045] According to equation (1), the shorter the sunshine hours, the smaller S becomes, and therefore C1 becomes larger. As the sunshine hours increase, S approaches L, and therefore C1 approaches 1. Therefore, C1 changes depending on the sunshine hours over one or more days.

[0046] Next, the generation unit 512 calculates a statistical value R of the measured values ​​of the solar radiation intensity for one day or multiple days included in the performance data 521 of each solar power generation unit 411-i. The statistical value R may be an average value, a median value, a maximum value, a minimum value, or the like. Then, the generation unit 512 calculates the solar radiation intensity coefficient C2 for the analysis target month of the analysis target year using the statistical value R for each solar power generation unit 411-i.

[0047] The generation unit 512 calculates the solar radiation intensity coefficient C2 by the following equation using, for example, the statistical value R and the maximum solar radiation intensity R0 that can be assumed as a normal measurement value.

[0048] C2=R0 / R (3)

[0049] The maximum solar radiation intensity R0 is 1.0 to 2.0 kW / m 2 According to equation (3), the smaller the statistical value R, the larger C2 becomes. As the statistical value R approaches R0, C2 approaches 1. Therefore, C2 changes according to the statistical value R.

[0050] Next, the generating unit 512 generates coefficient information 522 indicating the sunshine duration coefficient C1 and the solar radiation intensity coefficient C2 for the analysis target month of the analysis target year, and stores the generated coefficient information 522 in the storage unit 516.

[0051] Next, the generating unit 512 divides into a plurality of time slots the period of one day during which the measurement values ​​included in the performance data 521 exist. For example, if the length of each time slot is one hour, the period from 6:10 to 18:00 is divided into the following 12 time slots:

[0052] Time period T1: 6:10~7:00 Time period T2: 7:10~8:00 Time period T3: 8:10~9:00 Time period T4: 9:10~10:00 Time period T5: 10:10~11:00 Time period T6 11:10~12:00 Time period T7 12:10~13:00 Time period T8 13:10~14:00 Time period T9 14:10~15:00 Time period T10: 15:10~16:00 Time period T11: 16:10~17:00 Time period T12 17:10~18:00

[0053] Next, the generation unit 512 uses the performance data 521 of each solar power generation unit 411-i to generate profile information 523 for each solar power generation unit 411-i and for each time zone for the month being analyzed in the year being analyzed, and stores the information in the memory unit 516.

[0054] The profile information 523 represents the correlation between the solar radiation intensity and the power generated by each solar power generation unit 411-i in each time period, and is used to estimate the power generated by the solar power generation unit 411-i from the solar radiation intensity. The profile information 523 is an example of output estimation information.

[0055] FIG. 7 shows an example of the profile information 523. The profile information 523 in FIG. 7 includes solar radiation intensity (kW / m 2 ) and the estimated value of generated power (kW). Each estimated value of generated power is associated with a different solar radiation intensity.

[0056] For example, for each time period, the generation unit 512 extracts, from the measured values ​​of solar radiation intensity for all days belonging to the month being analyzed, the measured values ​​of solar radiation intensity within a predetermined range corresponding to each solar radiation intensity in the profile information 523. The lower limit of the predetermined range corresponding to a specific solar radiation intensity is obtained by subtracting a threshold value from the specific solar radiation intensity, and the upper limit of the predetermined range is obtained by adding a threshold value to the specific solar radiation intensity.

[0057] The generation unit 512 then obtains the measured value of the power generation at the same time as the extracted measured value of the solar radiation intensity from the performance data 521, and records the statistical value of the obtained measured value of the power generation as an estimated value of the power generation in the profile information 523. The statistical value may be an average value, a median value, a maximum value, a minimum value, or the like.

[0058] The range of solar radiation intensity recorded in the profile information 523 may be 0 to R0. 2 If the estimated value of the power generation corresponding to the solar radiation intensity exceeds 0 kW, the generation unit 512 may correct the estimated value of the power generation to 0 kW. Also, if the estimated value of the power generation exceeds the rated value of the PCS 414, the generation unit 512 may correct the estimated value of the power generation to the rated value of the PCS 414. In the example of FIG. 7, R0=1.2 kW / m 2 and the rated value of the PCS414 is 50kW.

[0059] Figure 8 shows an example of the correlation between solar radiation intensity and generated power. The horizontal axis shows solar radiation intensity (kW / m 2 ), and the vertical axis represents the power generation (kW). A straight line 801 shows the correlation represented by the profile information 523 in FIG. 7. The points around the straight line 801 represent the measured values ​​of the solar radiation intensity and the power generation used to generate the profile information 523.

[0060] As described above, the shadow cast on the solar cell of each solar power generation unit 411-i varies depending on the month or time of day, and therefore the correlation represented by the profile information 523 also varies depending on the month or time of day and is not necessarily represented by a straight line. The correlation represented by the profile information 523 may be represented by a different curve for each time of day, and even for the same time of day, it may be represented by a different curve for each month or season.

[0061] If the measured value of the solar radiation intensity does not exist within a predetermined range corresponding to a specific solar radiation intensity, the generation unit 512 may generate an estimate of the power generation power corresponding to the specific solar radiation intensity by performing linear interpolation using estimates of the power generation power corresponding to the solar radiation intensities before and after the specific solar radiation intensity.

[0062] Next, the generation unit 512 uses the profile information 523 to obtain, for each solar power generation unit 411-i and for each time period, an estimated value of power generation from the measured value of solar radiation intensity at each time used to generate the profile information 523. The generation unit 512 then obtains the divergence power at each time by subtracting the measured value of power generation at each time used to generate the profile information 523 from the estimated value of power generation at each time.

[0063] The power discrepancy calculated by the generating unit 512 is an example of an actual discrepancy. The measured values ​​of the solar radiation intensity and the generated power at each time used to generate the profile information 523 are an example of the measured values ​​of the solar radiation intensity and the generated power at each time on a day prior to the target diagnosis day.

[0064] Next, the generation unit 512 calculates, for each solar power generation unit 411-i and for each time period, a statistical value of the power deviation for each of a plurality of times included in the time period, generates variation information 524 indicating the calculated statistical value, and stores the information in the storage unit 516. The statistical value may be an average value, a median value, a maximum value, a minimum value, or the like. The statistical value of the power deviation indicated by the variation information 524 is an example of a predetermined value indicating a statistical value of the degree of deviation from the actual results for each of a plurality of times.

[0065] The generating unit 512 may generate coefficient information 522, profile information 523, and variation information 524 for each month of each past year by changing the year and month to be analyzed and repeating the profile generating process.

[0066] Next, in the anomaly detection process, the estimation unit 513 acquires the measured values ​​of the power generation power and solar radiation intensity of each solar power generation unit 411-i on the diagnosis target day from the database server 302 via the communication network 304. Then, the estimation unit 513 generates diagnosis target data 525 including the acquired measurement values ​​and stores it in the storage unit 516. The diagnosis target data 525 is generated for each solar power generation unit 411-i in the same format as in FIG. 6 .

[0067] Next, the estimation unit 513 uses the profile information 523 to calculate an estimated value of the power generation from the measured value of the solar radiation intensity at each time included in the diagnosis target data 525 for each solar power generation unit 411-i and for each time period on the diagnosis target day.

[0068] The detection unit 514 calculates the divergence power for each solar power generation unit 411-i and for each time slot on the diagnosis target day by subtracting the measured value of the power generation output at each time included in the diagnosis target data 525 from the estimated value of the power generation output at each time. The divergence power calculated by the detection unit 514 is an example of the degree of divergence between the estimated value of the power generation output and the measured value of the power generation output.

[0069] Next, the detection unit 514 uses the divergence power to determine the number K of consecutive power decreases at each time. The number K represents the number of times that the divergence power indicates that the measured value of the generated power is lower than the estimated value of the generated power.

[0070] The initial value of the consecutive number K is 0, and the consecutive number K is not updated while the power deviation at each time is 0 or less. Then, at the time when the power deviation becomes greater than 0, the consecutive number K is incremented by 1. Even when the time moves to the next time period, the consecutive number K is incremented by 1 at each time while the power deviation is greater than 0, and at the time when the power deviation becomes 0 or less, the consecutive number K is reset to 0. Therefore, the longer the time that the power deviation is greater than 0, the larger the consecutive number K becomes.

[0071] After determining the continuous count K, the detection unit 514 may change the power variance values ​​at each time that are negative to 0 kW. This allows the power variance values ​​at times when no power reduction occurs to be changed to 0 kW. The detection unit 514 then generates a power variance conversion table 526 using the power variance values ​​at each time and the variation information 524, and stores the table in the storage unit 516.

[0072] Since the measured value of the power generation varies from the estimated value estimated using the profile information 523, a power deviation smaller than the statistical value of the power deviation indicated by the variation information 524 is changed to 0 kW using the power deviation conversion table 526. This allows the power deviation smaller than the statistical value to be ignored, thereby reducing the effect that variation in the measured value of the power generation has on the calculation of the degree of anomaly.

[0073] Fig. 9 shows an example of the power deviation conversion table 526. The power deviation conversion table 526 in Fig. 9 includes power deviations (kW) before conversion and power deviations (kW) after conversion. Each converted power deviation is associated with each pre-conversion power deviation.

[0074] In this example, an average value is used as the statistical value of the power discrepancy indicated by the variation information 524, and the average value of the power discrepancy indicated by the variation information 524 is 0.64227 kW. Therefore, the power discrepancy after conversion corresponding to the power discrepancy before conversion of 0 to 0.6 kW is set to 0 kW. The maximum value of the power discrepancy before conversion and the power discrepancy after conversion, 50 kW, represents the rated value of the PCS 414. To reduce the amount of data in the power discrepancy conversion table 526, the resolution of the power discrepancy is set to 0.1 kW.

[0075] 10 shows an example of the correspondence relationship between the power deviation before conversion and the power deviation after conversion. The horizontal axis represents the power deviation before conversion (kW), and the vertical axis represents the power deviation after conversion (kW). A polygonal line 1001 shows the correspondence relationship represented by the power deviation conversion table 526 in FIG. 9.

[0076] Next, the detection unit 514 converts the divergence power at each time using the divergence power conversion table 526 to obtain the converted divergence power P. Then, the detection unit 514 acquires from the storage unit 516 the sunshine duration coefficient C1 and the solar radiation intensity coefficient C2 indicated by the coefficient information 522 for the same month as the diagnosis target date in any year prior to the diagnosis target date.

[0077] Next, the detection unit 514 calculates the degree of abnormality X for each time on the diagnosis target day using the divergence power P, the continuation number K, the sunshine duration coefficient C1, and the solar radiation intensity coefficient C2 according to the following formula.

[0078] X=P×K×C1×C2 (4)

[0079] Next, the detection unit 514 calculates the total value Z of the abnormality degrees X for all times on the target day for diagnosis, and compares the total value Z with a threshold value. The detection unit 514 then generates a diagnosis result 527 for the target day for diagnosis, including the determination result of the total value Z, and stores it in the storage unit 516.

[0080] If the total value Z is equal to or less than the threshold, the detection unit 514 records “normal” as the determination result of the total value Z in the diagnosis result 527. On the other hand, if the total value Z is greater than the threshold, the detection unit 514 detects an abnormality in the solar power generation unit 411-i, and records “abnormal” as the determination result of the total value Z in the diagnosis result 527. The output unit 515 outputs the diagnosis result 527.

[0081] According to formula (4), by using the deviation power P, a temporary drop in the power generation at each time is reflected in the degree of abnormality X, and by using the continuation number K, the duration for which the drop in power generation continues is reflected in the degree of abnormality X. Therefore, by using the deviation power P and the continuation number K, a drop in the power generation of the solar power generation unit 411-i can be detected as an abnormality.

[0082] The longer the time period for diagnosis on the diagnosis date, the larger the continuous number K becomes, and the larger the abnormality level X becomes. Therefore, when comparing the abnormality level X or total value Z on different days, it is desirable to standardize the time period for diagnosis.

[0083] When calculating the deviation power P, different profile information 523 is used for each time period on the day to be diagnosed to calculate an estimated value of the power generation, and thus the change in the deviation power P depending on the time period is reflected in the abnormality degree X. Therefore, the fluctuation in the abnormality degree X due to the change in the solar radiation conditions depending on the time period is corrected, and the accuracy of the abnormality determination based on the abnormality degree X is improved. This makes it possible to appropriately determine an abnormality in the solar power generation unit 411-i depending on the change in the solar radiation conditions depending on the time period.

[0084] Furthermore, by calculating an estimated value of the power generation power using profile information 523 that differs for each month to which the diagnosis target date belongs, the change in the deviation power P depending on the season of the diagnosis target date is reflected in the abnormality degree X. Therefore, fluctuations in the abnormality degree X due to seasonal changes in solar radiation conditions are corrected, and the accuracy of abnormality determination based on the abnormality degree X is improved. This makes it possible to appropriately determine an abnormality in the solar power generation unit 411-i depending on seasonal changes in solar radiation conditions.

[0085] By using the sunshine duration coefficient C1 for the same month as the diagnosis target date, the change in sunshine duration according to the season of the diagnosis target date is reflected in the abnormality level X. For example, in summer, the sunshine duration is long, so the sunshine duration coefficient C1 is small, and in winter, the sunshine duration is short, so the sunshine duration coefficient C1 is large.

[0086] Furthermore, by using the solar radiation intensity coefficient C2 of the same month as the diagnosis target date, changes in solar radiation intensity according to the season of the diagnosis target date are reflected in the abnormality degree X. For example, in summer, the solar radiation intensity is strong, so the solar radiation intensity coefficient C2 is small, and in winter, the solar radiation intensity is weak, so the solar radiation intensity coefficient C2 is large.

[0087] Therefore, by using the sunshine duration coefficient C1 and the solar radiation intensity coefficient C2, the fluctuation of the abnormality degree X due to the seasonal change in solar radiation conditions is further corrected, and the accuracy of the abnormality determination based on the abnormality degree X is further improved. This makes it possible to appropriately compare the diagnosis results 527 of different seasons.

[0088] The detection unit 514 may calculate the degree of abnormality X using another calculation formula, instead of formula (4), in which the degree of abnormality X increases as the deviation power P and the continuation count K increase. If it is not necessary to reflect seasonal changes in the sunshine duration or solar radiation intensity in the degree of abnormality X, C1 or C2 in formula (4) can be omitted.

[0089] If the determination result of the total value Z is abnormal, the administrator of the solar power generation system 301 may notify the user of the solar power generation system 301 of the abnormality of the solar power generation unit 411-i. Furthermore, if the determination result of the total value Z is abnormal, the administrator may dispatch an engineer to the installation location of the solar power generation unit 411-i to inspect, repair, or the like the solar power generation unit 411-i.

[0090] Fig. 11 is a flowchart showing an example of data selection processing performed by the anomaly detection device 303 in Fig. 5. First, the acquisition unit 511 sets the year and month to be analyzed as the year and month to be collected (step 1101).

[0091] Next, the acquisition unit 511 acquires the measured values ​​of the power generation power and solar radiation intensity of each solar power generation unit 411-i for the target month of the target year from the database server 302. Then, the acquisition unit 511 generates performance data 521 including the acquired measured values ​​for each solar power generation unit 411-i (step 1102).

[0092] Next, the acquisition unit 511 deletes the measurement values ​​that are outside the measurable ranges of the measurement device 416 and the pyranometer 417 from the performance data 521 (step 1103).

[0093] Next, the acquisition unit 511 calculates the integrated value of the measured values ​​of solar radiation intensity and the integrated value of the measured values ​​of the generated power for each day, and calculates the power generation efficiency by dividing the integrated value of the measured values ​​of the generated power by the integrated value of the measured values ​​of solar radiation intensity.The acquisition unit 511 then deletes the measured values ​​of the generated power and solar radiation intensity for days with power generation efficiency equal to or lower than the threshold value from the performance data 521 (step 1104).

[0094] Next, the acquisition unit 511 deletes the measurement value of the power generation that is equal to or less than the threshold value from the performance data 521 (step 1105). For example, if any of the solar power generation units 411-i is stopped at a specific time on a specific day, the measurement value of the power generation at that time will be equal to or less than the threshold value, and therefore the measurement value is deleted.

[0095] Next, the acquiring unit 511 deletes the measurement values ​​of the power generation power and the solar irradiance intensity on the excluded day from the performance data 521 (step 1106). The excluded day is, for example, a day on which the photovoltaic power generation system 301 was stopped due to output control or equipment inspection.

[0096] Next, the acquisition unit 511 deletes the measurement values ​​to be excluded from the performance data 521 (step 1107). The measurement values ​​to be excluded are, for example, the measurement values ​​of the power generation power and the solar radiation intensity on a day that was determined to be abnormal in the past abnormality detection process, and are measurement values ​​that are specified to be excluded in the profile generation process.

[0097] Next, the acquisition unit 511 checks whether the performance data 521 contains the number of measurement values ​​required to generate the profile information 523 (step 1108). The number required to generate the profile information 523 is specified in advance.

[0098] If the required number of measurement values ​​are included in the performance data 521 (step 1108, YES), the acquisition unit 511 records in the performance data 521 that it is possible to generate profile information 523 using the performance data 521 (step 1109).

[0099] If the required number of measured values ​​are not included in the performance data 521 (step 1108, NO), the acquisition unit 511 checks whether the year of collection is a specific year (step 1110).

[0100] The specific year represents the earliest year for which additional measurement values ​​can be collected. For example, if the equipment conditions of the solar power generation unit 411-i are changed to 2015, 2015 is designated as the specific year. The equipment conditions of the solar power generation unit 411-i include the number of strings 431-ij included in the solar power generation unit 411-i, the number of solar panels included in each string 431-ij, etc.

[0101] If the year to be collected is a specific year (step 1110, YES), the acquisition unit 511 records in the performance data 521 that it is not possible to generate the profile information 523 using the performance data 521 (step 1111).

[0102] If the collection target year is not a specific year (step 1110, NO), the acquisition unit 511 updates the collection target year by setting the year previous to the collection target year as the new collection target year (step 1112).Then, the acquisition unit 511 repeats the processing from step 1102 onwards.

[0103] As a result, in step 1102, the measured values ​​of the power generation power and solar radiation intensity for the same month of the new collection year are obtained from the database server 302 and added to the performance data 521. By specifying the year in which the equipment conditions of the photovoltaic power generation unit 411-i were changed as the specific year, it is possible to prevent the measured values ​​of the power generation power before the change in the equipment conditions from being added.

[0104] FIG. 12 is a flowchart showing an example of a profile generation process performed by the anomaly detection device 303 of FIG.

[0105] 12 is performed for a month for which it is recorded in the performance data 521 that the generation of profile information 523 is possible in step 1109 of Fig. 11, using the performance data 521 of that month. Therefore, for a month for which it is recorded in the performance data 521 that it is not possible to generate profile information 523, profile information 523 is not generated.

[0106] First, the generation unit 512 acquires the performance data 521 of each photovoltaic power generation unit 411-i for the month of interest in the year of interest from the storage unit 516 (step 1201).

[0107] Next, the generation unit 512 calculates an index S corresponding to the sunshine duration from the measured values ​​of the solar radiation intensity for one day or multiple days included in the performance data 521. Then, the generation unit 512 calculates a sunshine duration coefficient C1 for the analysis target month of the analysis target year using the index S for each solar power generation unit 411-i (step 1202).

[0108] Next, the generation unit 512 calculates a statistical value R of the measured values ​​of the solar radiation intensity for one day or multiple days included in the performance data 521. Then, the generation unit 512 calculates, for each solar power generation unit 411-i, a solar radiation intensity coefficient C2 for the month being analyzed in the year being analyzed, using the statistical value R (step 1203).

[0109] Next, the generating unit 512 generates coefficient information 522 indicating the sunshine duration coefficient C1 and the solar radiation intensity coefficient C2 (step 1204).

[0110] Next, the generating unit 512 divides the period of the day in which the measurement values ​​included in the performance data 521 exist into a plurality of time slots (step 1205), and selects the earliest time slot as the time slot to be processed (step 1206).

[0111] Next, the generation unit 512 uses the performance data 521 to generate, for each photovoltaic power generation unit 411-i, profile information 523 for the time period to be processed in the month to be analyzed in the year to be analyzed (step 1207).

[0112] Next, the generation unit 512 uses the profile information 523 to generate, for each solar power generation unit 411-i, variation information 524 for the time period to be processed in the month to be analyzed in the year to be analyzed (step 1208).

[0113] Next, the generation unit 512 checks whether all time periods have been selected as time periods to be processed (step 1209). If there are any unselected time periods remaining (step 1209, NO), the generation unit 512 selects the next time period as the time period to be processed (step 1212) and repeats the processing from step 1207 onwards.

[0114] If all time periods have been selected (step 1209, YES), the generation unit 512 checks whether there is a time period for which profile information 523 has not been generated (step 1210). For example, if measurement values ​​are missing due to a malfunction of the measurement device 416 or the pyranometer 417, a communication failure, or the like, there is a possibility that the profile information 523 will not be generated.

[0115] If there is a time period for which profile information 523 has not been generated (step 1210, YES), the generation unit 512 copies the profile information 523 of the time period before or after that time period. Then, the generation unit 512 stores the copied profile information 523 in the storage unit 516 as the profile information 523 of that time period (step 1211). If profile information 523 has been generated for all time periods (step 1210, NO), the generation unit 512 ends the process.

[0116] 13A and 13B are flowcharts showing an example of anomaly detection processing performed by the anomaly detection device 303 in Fig. 5. First, the estimation unit 513 acquires the measured values ​​of the power generation power and solar irradiance intensity of each solar power generation unit 411-i on the target diagnosis day from the database server 302. Then, the estimation unit 513 generates diagnosis target data 525 including the acquired measured values ​​for each solar power generation unit 411-i (step 1301).

[0117] Next, the estimation unit 513 compares the number of measurement values ​​of the power generation and the solar radiation intensity included in the diagnosis target data 525 that are outside the measurable ranges of the measurement device 416 and the pyranometer 417 with a threshold value (step 1302).

[0118] If the number of measurement values ​​outside the measurable range is equal to or less than the threshold value (step 1302 , YES), the estimation unit 513 performs the process of step 1303 .

[0119] If the number of measurement values ​​outside the measurable range is greater than the threshold (step 1302, NO), the estimation unit 513 records an abnormality in the number of out-of-range data in the diagnosis result 527 (step 1306), and performs the process of step 1303. An abnormality in the number of out-of-range data occurs when the measurement value falls outside the measurable range due to, for example, a malfunction of the measuring device 416 or the pyranometer 417, a communication failure, or the like.

[0120] In step 1303, the estimation unit 513 compares the number of measurement values ​​of the generated power and the solar radiation intensity included in the diagnosis object data 525 that continuously show a constant value with a threshold value.

[0121] If the number of measurement values ​​that continuously show a constant value is equal to or less than the threshold value (step 1303 , YES), the estimation unit 513 performs the process of step 1304 .

[0122] If the number of measurement values ​​that continuously show a constant value is greater than the threshold value (step 1303, NO), the estimation unit 513 records an abnormality in the number of constant value data in the diagnosis result 527 (step 1307), and performs the process of step 1304. An abnormality in the number of constant value data occurs when the measurement value becomes a constant value due to, for example, a malfunction of the measuring device 416 or the pyranometer 417, a communication failure, or the like.

[0123] However, even if the measured value of solar radiation intensity continuously indicates a constant value, if the measured value of generated power is equal to or less than the threshold value, the estimation unit 513 may not record an abnormality in the number of constant value data points of solar radiation intensity. Also, even if the measured value of generated power continuously indicates a constant value, if the measured value of solar radiation intensity is equal to or less than the threshold value or if the measured value of generated power is less than the rated value of the PCS 414, the estimation unit 513 may not record an abnormality in the number of constant value data points of generated power.

[0124] In step 1304, the estimation unit 513 checks, for each solar power generation unit 411-i and for each time zone, whether or not profile information 523 for the same month of the previous year of the diagnosis target date exists in the storage unit 516. The same month of the previous year of the diagnosis target date refers to the same month one year prior to the month to which the diagnosis target date belongs.

[0125] If profile information 523 for the same month of the previous year to the diagnosis target date exists (step 1304, YES), profile information 523, coefficient information 522, and variation information 524 for the same month of the previous year to the diagnosis target date are obtained from storage unit 516 (step 1305).

[0126] If there is no profile information 523 for the same month of the previous year of the target diagnosis date (step 1304, NO), the estimation unit 513 checks whether there is profile information 523 for the same month before the target diagnosis date in the storage unit 516 (step 1308). The same month before the target diagnosis date refers to the month before the month to which the target diagnosis date belongs.

[0127] If profile information 523 for the previous month of the diagnosis target date exists (step 1308, YES), profile information 523, coefficient information 522, and variation information 524 for the previous month of the diagnosis target date are obtained from storage unit 516 (step 1309).

[0128] When the processing of step 1305 or step 1309 has been performed for all the photovoltaic power generation units 411-i and all the time periods, the estimation unit 513 performs the processing of step 1311.

[0129] If profile information 523 for the previous month of the same month as the diagnosis target date does not exist (step 1308, NO), the estimation unit 513 performs the process of step 1310. In step 1310, the estimation unit 513 associates the information indicating that diagnosis is not possible with the solar power generation unit 411-i for which profile information 523 does not exist, and records the information in the storage unit 516. Then, the estimation unit 513 performs the process of step 1311.

[0130] In step 1311, the estimation unit 513 calculates an estimate of the power generation at each time for each time period for the solar power generation unit 411-i for which information indicating that diagnosis is not possible is not recorded, using the acquired profile information 523 and the diagnosis target data 525.

[0131] Next, the detection unit 514 calculates the deviation power by subtracting the measured value of the generated power from the estimated value of the generated power at each time for each solar power generation unit 411-i and for each time period (step 1312).

[0132] Next, the detection unit 514 generates a power variance conversion table 526 for each solar power generation unit 411-i and for each time period, using the power variance at each time and the variation information 524. The detection unit 514 then converts the power variance at each time using the power variance conversion table 526, thereby obtaining the converted power variance P (step 1313).

[0133] Next, the detection unit 514 uses the power deviation P to obtain the number K of consecutive power drops at each time (step 1314).

[0134] Next, the detection unit 514 calculates the degree of abnormality X for each time period for each solar power generation unit 411-i using the power deviation P and the continuation count K, and the sunshine duration coefficient C1 and solar radiation intensity coefficient C2 indicated by the acquired coefficient information 522 according to formula (4) (step 1315).The detection unit 514 then calculates the total value Z of the degrees of abnormality X for all times for each solar power generation unit 411-i (step 1316) and compares the total value Z with a threshold value (step 1317).

[0135] If the total value Z is equal to or less than the threshold value (step 1317, YES), the output unit 515 outputs the diagnosis result 527 (step 1318). If the total value Z is greater than the threshold value (step 1317, NO), the detection unit 514 records an abnormality as the determination result of the total value Z in the diagnosis result 527 (step 1319). Then, the output unit 515 outputs the diagnosis result 527 (step 1318).

[0136] 14 shows an example of a diagnosis result 527. The index represents an index of abnormality, the evaluation value represents the recorded index value, the threshold represents the threshold for the index value, and the judgment result represents either normal or abnormal.

[0137] The evaluation value of the number of out-of-range data of solar radiation intensity represents the number of measured values ​​of solar radiation intensity that are outside the measurable range of the pyranometer 417. This evaluation value is "0", which is smaller than the threshold value of "36", so "normal" is recorded as the judgment result.

[0138] The evaluation value of the number of data points outside the power generation range represents the number of measured values ​​of power generation that are outside the measurable range of the measurement device 416. This evaluation value is "0", which is smaller than the threshold value of "36", so "normal" is recorded as the judgment result.

[0139] The evaluation value for the number of constant solar radiation intensity data points represents the number of solar radiation intensity measurements that continue to show a constant value. This evaluation value is "10," which is greater than the threshold value of "6," so an abnormality is recorded as the judgment result.

[0140] The evaluation value for the number of constant power generation data points represents the number of measured power generation values ​​that continue to show a constant value. This evaluation value is "2," which is smaller than the threshold value of "6," so the judgment result is recorded as normal.

[0141] The abnormality level of the power drop represents the total value Z of the abnormality levels X at all times. This evaluation value is "120", which is smaller than the threshold value "400", so "normal" is recorded as the judgment result.

[0142] According to the anomaly detection process of FIGS. 13A and 13B, a decrease in the power generation of the solar power generation unit 411-i can be detected as an anomaly, and a malfunction of the measuring device 416 or the actinometer 417, a communication failure, etc. can be detected as an anomaly of out-of-range data or a certain value data number.

[0143] The anomaly detection system of FIG. 3 may perform the data selection process, profile generation process, and anomaly detection process using current, voltage, etc. as the power generation output instead of the power generated by the solar power generation unit 411-i.

[0144] The configurations of the anomaly detection device 101 in Fig. 1 and the anomaly detection device 303 in Fig. 5 are merely examples, and some of the components may be omitted or changed depending on the application or conditions of the anomaly detection device. For example, in the anomaly detection device 303 in Fig. 5, if the profile information 523, the coefficient information 522, and the variation information 524 are generated by an external device, the acquisition unit 511 and the generation unit 512 can be omitted.

[0145] The configuration of the anomaly detection system in FIG. 3 is merely an example, and some of the components may be omitted or changed depending on the application or conditions of the anomaly detection system.

[0146] The configuration of the solar power generation system 301 in FIG. 4 is merely an example, and some of the components may be omitted or changed depending on the application or conditions of the solar power generation system 301.

[0147] 2 and 11 to 13B are merely examples, and some of the processes may be omitted or changed depending on the configuration or conditions of the anomaly detection device 101 or the anomaly detection system. For example, in the anomaly detection process of Figures 13A and 13B, if it is not necessary to detect anomalies in out-of-range data and a certain number of data values, the processes of steps 1302, 1303, 1306, and 1307 can be omitted.

[0148] The solar radiation intensity and generated power shown in Figures 6 to 8 are merely examples, and the measured values ​​of solar radiation intensity and generated power and the estimated values ​​of generated power change depending on the configuration or conditions of the anomaly detection system. The power deviation conversion table 526 shown in Figures 9 and 10 is merely an example, and the power deviation conversion table 526 changes depending on the power deviation before conversion and the variation information 524. The diagnosis result 527 shown in Figure 14 is merely an example, and the diagnosis result 527 changes depending on the diagnosis target date.

[0149] Equations (1) to (4) are merely examples, and the anomaly detection system may use other calculation formulas to perform the profile generation process and the anomaly detection process.

[0150] Fig. 15 shows an example of the hardware configuration of an information processing device (computer) used as the anomaly detection device 101 in Fig. 1 and the anomaly detection device 303 in Fig. 5. The information processing device in Fig. 15 includes a CPU (Central Processing Unit) 1501, a memory 1502, an input device 1503, an output device 1504, an auxiliary storage device 1505, a media drive device 1506, and a network connection device 1507. These components are hardware and are connected to each other by a bus 1508.

[0151] The memory 1502 is, for example, a semiconductor memory such as a read-only memory (ROM) or a random access memory (RAM), and stores programs and data used in processing. The memory 1502 may operate as the storage unit 516 in FIG.

[0152] 1 by executing a program using the memory 1502. The CPU 1501 (processor) also operates as the acquisition unit 511, generation unit 512, estimation unit 513, and detection unit 514 in FIG. 5 by executing a program using the memory 1502.

[0153] The input device 1503 is, for example, a keyboard, a pointing device, etc., and is used to input instructions or information from the administrator. The output device 1504 is, for example, a display device, a printer, etc., and is used to output inquiries or instructions to the administrator and processing results. The output device 1504 may operate as the output unit 515 in FIG. 5, and the processing results may be the diagnosis results 527.

[0154] The auxiliary storage device 1505 is, for example, a magnetic disk device, an optical disk device, a magneto-optical disk device, a tape device, or the like. The auxiliary storage device 1505 may be a hard disk drive or a solid state drive (SSD). The information processing device stores programs and data in the auxiliary storage device 1505 and can use them by loading them into the memory 1502. The auxiliary storage device 1505 may operate as the storage unit 516 in FIG. 5.

[0155] The medium drive device 1506 drives the portable recording medium 1509 and accesses the recorded contents thereof. The portable recording medium 1509 is a memory device, a flexible disk, an optical disk, a magneto-optical disk, etc. The portable recording medium 1509 may be a CD-ROM (Compact Disk Read Only Memory), a DVD (Digital Versatile Disk), a USB (Universal Serial Bus) memory, etc. The administrator can store programs and data in the portable recording medium 1509 and load them into the memory 1502 for use.

[0156] In this way, the computer-readable recording medium that stores the program and data used in the processing is a physical (non-transitory) recording medium such as the memory 1502, the auxiliary storage device 1505, or the portable recording medium 1509.

[0157] The network connection device 1507 is a communication circuit connected to the communication network 304 and performs data conversion associated with communication. The information processing device receives programs and data from external devices via the network connection device 1507 and loads them into the memory 1502 for use. The network connection device 1507 may operate as the output unit 515 in FIG. 5.

[0158] 15, some components may be omitted or changed depending on the purpose or conditions of the information processing device. For example, if an interface with an administrator is not required, the input device 1503 and the output device 1504 may be omitted. If the portable recording medium 1509 is not used, the medium drive device 1506 may be omitted.

[0159] The database server 302 in FIG. 3 can be an information processing device similar to that in FIG.

[0160] While the disclosed embodiments and their advantages have been described in detail above, those skilled in the art may make various modifications, additions, and omissions without departing from the scope of the invention as clearly set forth in the claims. [Explanation of symbols]

[0161] 101, 303 Anomaly detection device 111, 513 Estimation part 112, 514 Detection unit 301 Solar Power Generation System 302 Database Server 304 Communication Network 411-1~411-N Solar Power Generation Department 412-1~412-N Junction Box 413 Current collector box 414 PCS 415 Switchboard 416 Measuring Equipment 417 Pyranometer 418 Thermometer 419 Weather Conversion Box 420 Communication Equipment 431-1-1~431-1-N string 511 Acquisition Department 512 Generation part 515 Output section 516 Storage section 521 Performance Data 522 Coefficient Information 523 Profile Information 524 Variation Information 525 diagnostic data 526 Deviation Power Conversion Table 527 Diagnosis Results 801 straight line 1001 Line 1501 CPU 1502 memory 1503 Input Device 1504 Output Device 1505 Auxiliary storage device 1506 Media drive unit 1507 Network connection device 1508 Bus 1509 Portable recording media

Claims

1. an estimation unit that calculates an estimated value of the power generation output of the photovoltaic power generation system for each of a plurality of time periods included in one day from measured values ​​of solar radiation intensity for each of the plurality of time periods on a target day based on output estimation information for each of the plurality of time periods included in one day; a detection unit that calculates a degree of discrepancy between the estimated value of the power generation output and a measured value of the power generation output of the photovoltaic power generation system for each of the plurality of time periods on the target day, and detects an abnormality in the photovoltaic power generation system based on the degree of discrepancy for each of the plurality of time periods; An anomaly detection device comprising:

2. the detection unit calculates the deviation degree for each of a plurality of times included in each of the plurality of time periods, and detects an abnormality in the photovoltaic power generation system based on the deviation degree for each of the plurality of times included in each of the plurality of time periods and the duration of each of the plurality of times included in each of the plurality of time periods; 2. The anomaly detection device according to claim 1, wherein the consecutive number of times represents the number of times during which the deviation continuously indicates that the measured value of the power generation output is smaller than the estimated value of the power generation output.

3. the detection unit calculates an anomaly degree for each of a plurality of times included in each of the plurality of time periods using the deviation degree, the continuation number, and a sunshine duration coefficient for the target day, and detects an anomaly in the photovoltaic power generation system based on the anomaly degree; 3. The anomaly detection device according to claim 2, wherein the sunshine duration coefficient varies depending on the sunshine duration in one or more days.

4. The detection unit calculates the degree of abnormality by further using a solar radiation intensity coefficient for the target day, 4. The anomaly detection device according to claim 3, wherein the solar radiation intensity coefficient changes according to a statistical value of solar radiation intensity over one or more days.

5. When the deviation degree is smaller than a predetermined value, the detection unit changes the deviation degree to 0 and calculates the abnormality degree; the predetermined value represents a statistical value of the degree of deviation of the actual results at each of the plurality of times; the actual deviation degree at each of the plurality of times represents a deviation degree between an estimated value of the power generation output of the photovoltaic power generation system at each of the plurality of times on a day before the target day and a measured value of the power generation output of the photovoltaic power generation system at each of the plurality of times on a day before the target day; 5. The anomaly detection device according to claim 3, wherein the estimated value of the power generation output of the solar power generation system at each of the plurality of times on the day before the target day is estimated from the measured value of solar radiation intensity at each of the plurality of times on the day before the target day, based on the output estimation information for a time period including the plurality of times.

6. calculating an estimated value of the power generation output of the photovoltaic power generation system for each of the plurality of time periods on a target day from measured values ​​of the solar radiation intensity for each of the plurality of time periods based on output estimation information for each of the plurality of time periods included in one day; calculating a degree of discrepancy between the estimated power output and a measured power output of the photovoltaic power generation system for each of the plurality of time periods on the target day; detecting an abnormality in the photovoltaic power generation system based on the deviation degree for each of the plurality of time periods; An anomaly detection method characterized in that processing is executed by a computer.

7. calculating an estimated value of the power generation output of the photovoltaic power generation system for each of the plurality of time periods on a target day from measured values ​​of the solar radiation intensity for each of the plurality of time periods based on output estimation information for each of the plurality of time periods included in one day; calculating a degree of discrepancy between the estimated power output and a measured power output of the photovoltaic power generation system for each of the plurality of time periods on the target day; detecting an abnormality in the photovoltaic power generation system based on the deviation degree for each of the plurality of time periods; An anomaly detection program that causes a computer to execute a process.

Images