JP7435073B2 - Anomaly detection device, anomaly detection method, and anomaly detection program - Google Patents

Description

The present invention relates to an abnormality detection device, an abnormality detection method, and an abnormality detection program for a solar power generation device.

A power generation analysis technique for checking whether the power generated by a solar power generation device is appropriate has been disclosed. These technologies estimate the power generated by a power generation device from the solar radiation intensity, compare the measured power generation with the estimated power generation, and if the two values deviate significantly, it is determined that the power generation device is abnormal. It is determined that

For example, the power generated on the target day is estimated based on the power generation data (generated power, solar radiation intensity) of the solar power generation device between 6:00 and 17:59 on days when the solar radiation intensity is above the average solar radiation intensity in the previous month. An estimation model is generated, and if there is a large discrepancy between the estimated power generation calculated based on the estimation model and the actually measured power generation, it is possible to detect that a failure or trouble in the power generation equipment, a change in the environment, etc. has occurred. A detection technique has been disclosed (Patent Document 1).

JP2014-179464

In the abnormality detection technique as described above, it is preferable to be able to suitably determine an equipment abnormality of the power generation device.

However, in the conventional technology, for example, due to occurrence of events (1) to (5) shown below, it may not be possible to correctly determine equipment abnormality of the power generation device.
(1) When the solar power generation device is overloaded, that is, when the actual power generation output of the device is greater than the rated output of the PCS (power conditioning system) (2) When the PCS power factor control setting is performed according to the instructions of the electric power company. (3) When the power generation output of the PCS is suppressed due to a disturbance such as a rise in power grid voltage (4) When there is an output control instruction (instruction to stop power generation) by the power company (5) Self-consumption power generation When equipment output control is required (suppression of PCS output due to a decrease in domestic power demand during holidays or lunch breaks, etc.) In addition, with conventional technology, in addition to the above-mentioned events, Due to the influence of shadows, etc., there were cases in which it was not possible to correctly determine equipment abnormalities in the power generation equipment.

In view of the above-mentioned problems, the present invention aims to provide a technique that can suitably determine equipment abnormality of a power generation device.

An abnormality detection device according to one aspect of the present invention includes a first acquisition unit that acquires power generation information indicating power generated by a solar power generation device, and a first acquisition unit that acquires power generation information indicating the power generated by the solar power generation device; an estimation model generated by referring to the second acquisition unit that acquires the electricity, the generated power actually generated in the predetermined sunshine conditions and time, and the actual solar radiation intensity in the predetermined sunshine conditions and time. The solar radiation intensity indicated by the solar radiation intensity information acquired by the second acquisition unit is calculated from the solar radiation intensity under the predetermined solar radiation conditions and time using an estimation model that estimates the generated power based on the solar radiation intensity. an estimating unit that estimates the generated power in the predetermined sunshine condition and time zone; an estimated generated power that is the generated power estimated by the estimating unit; and the generated power indicated by the generated power information acquired by the first acquisition unit. and an abnormality detection unit that determines the presence or absence of an abnormality by comparing the generated power in the predetermined sunlight condition and time, and the predetermined sunlight condition is determined when the solar radiation intensity is below a predetermined threshold. However, the time period does not include morning, evening, or night.

According to the above configuration, an abnormality in the solar power generation device can be accurately detected by generating a power generation estimation model limited to data for a predetermined sunshine condition and time period. Furthermore, by reducing the amount of data to be targeted, the processing load on the abnormality detection device can be reduced.

In the anomaly detection device according to one aspect of the present invention, the estimation model used by the anomaly detection unit to determine anomaly on a target date is generated with reference to generated power and solar radiation intensity in a plurality of periods in the past than the target date. It may also be an estimated model.

According to the above configuration, by comparing the generated power in a plurality of past periods with the generated power on the target day, it is possible to accurately detect an abnormality in the power generation device.

In the anomaly detection device according to one aspect of the present invention, the estimation model used by the anomaly detection unit to determine an anomaly on a target date is based on the generated power and solar radiation intensity in the month preceding the month including the target date, and the target date. The estimation model may be generated by referring to the generated power and solar radiation intensity in one month in the past including the month.

According to the above configuration, an abnormality in the power generation device can be accurately detected by performing a comparison with the generated power of the previous month and a comparison with the generated power of the same month in the past.

In the anomaly detection device according to one aspect of the present invention, the estimation model may include a regression curve including a nonlinear term.

According to the above configuration, it is possible to generate a model that can estimate generated power from solar radiation intensity with high accuracy, and it is possible to detect abnormalities in the power generation device with high accuracy.

In the abnormality detection device according to one aspect of the present invention, the predetermined threshold value is 0.6 kW/m ² or less, and the time period is any time period from 11:00 to 12:59. Good too.

According to the above configuration, it is possible to generate a power estimation model by eliminating as much as possible the power data during the time period when the solar power generation device or the pyranometer is affected by the shadow, and the power data at the time when the power is restricted. By referring to , it is possible to detect abnormalities in the power generation equipment with higher accuracy. Furthermore, the amount of data used to estimate generated power can be reduced, and the processing load on the abnormality detection device can be reduced.

The abnormality detection device according to one aspect of the present invention may include an estimated model generation unit that generates the estimated model.

According to the above configuration, an abnormality in the power generation device can be detected with a simpler configuration.

In the abnormality detection device according to one aspect of the present invention, the first acquisition unit acquires power generation information indicating the power generated by each of the plurality of solar power generation devices, and the second acquisition unit includes: The abnormality detection unit acquires solar radiation intensity information indicating the solar radiation intensity in each of the plurality of solar power generation devices, and the abnormality detection unit calculates estimated generated power and the first acquisition for each of the plurality of solar power generation devices. The presence or absence of an abnormality is determined using the degree of deviation from the generated power indicated by the generated power information acquired by the section, and if it is determined that a predetermined number or more of solar power generation devices are abnormal, the plurality of solar power generation devices The degree of deviation regarding each solar power generation device may be corrected by referring to the statistical value derived from the degree of deviation in each power generation device, and the presence or absence of an abnormality may be determined based on the degree of deviation after correction. .

According to the above configuration, by correcting the data of each power generation device by referring to statistical values derived from data of multiple solar power generation devices, abnormalities in the solar power generation device can be detected even under unstable weather conditions. Can be accurately detected.

An estimation model generation device according to one aspect of the present invention includes a first acquisition unit that acquires power generation information indicating power generated by a solar power generation device, and a second acquisition unit that acquires solar radiation intensity information indicating solar radiation intensity. an estimation unit that estimates the generated power based on the solar radiation intensity by referring to the acquisition unit, the generated power actually generated under the predetermined solar radiation conditions and time, and the actual solar radiation intensity under the predetermined solar radiation conditions and time; The apparatus includes an estimated model generating section that generates a model, and the predetermined sunshine condition includes that the solar radiation intensity is less than or equal to a predetermined threshold, and the time period does not include morning, evening, and night.

According to the above configuration, the same effects as the abnormality detection unit device according to one aspect of the present invention can be obtained.

An abnormality detection method according to one aspect of the present invention includes a first acquisition step of acquiring power generation information indicating power generated by a solar power generation device, and solar radiation intensity information indicating solar radiation intensity at the solar power generation device. The estimation model is generated by referring to the second acquisition step, the generated power actually generated in the predetermined sunlight conditions and time, and the actual solar radiation intensity in the predetermined sunlight conditions and time. The solar radiation intensity indicated by the solar radiation intensity information acquired in the second acquisition step is calculated from the solar radiation intensity under the predetermined solar radiation conditions and time using an estimation model that estimates the generated power based on the solar radiation intensity. , an estimation step of estimating the generated power under the predetermined sunshine conditions and time period, an estimated generated power that is the generated power estimated in the estimation step, and a generated power indicated by the generated power information acquired in the first acquisition step. an abnormality determination step of determining the presence or absence of an abnormality by comparing the electric power generated under the predetermined sunshine conditions and time period;
The predetermined sunshine condition includes that the solar radiation intensity is less than or equal to a predetermined threshold, and the time period does not include morning, evening, and night.

According to the above configuration, the same effects as the abnormality detection device according to one aspect of the present invention can be obtained.

An anomaly detection program according to an aspect of the present invention is an anomaly detection program for causing a computer to function as an anomaly detection device according to an aspect of the present invention, and for causing a computer to function as the estimating section and the anomaly detecting section. It may be an abnormality detection program.

According to the above configuration, the same effects as the abnormality detection device according to one aspect of the present invention can be obtained.

According to the abnormality detection device according to one aspect of the present invention, it is possible to suitably determine an equipment abnormality of a power generation device.

1 is a block diagram showing an example of the configuration of a solar power generation system including an abnormality detection device according to an embodiment of the present invention. FIG. 1 is a block diagram showing an example of the configuration of an abnormality detection device according to an embodiment of the present invention. 5 is a graph showing a range of data used in a power generation estimation model in power generation data of the abnormality detection device according to an embodiment of the present invention. It is a flowchart which shows the flow of estimation processing of generated power in an abnormality detection device concerning one embodiment of the present invention. It is a graph showing the relationship between solar radiation intensity and generated power acquired by the abnormality detection device according to an embodiment of the present invention. It is a flowchart which shows the flow of abnormality detection processing in an abnormality detection device concerning one embodiment of the present invention. It is a table for explaining correction of the degree of deviation of generated power data in an abnormality detection device according to another embodiment of the present invention.

Embodiments of the present invention will be described in detail below with reference to the drawings. Note that for convenience of explanation, members having the same functions as the members described above may be denoted by the same reference numerals, and the description thereof may be omitted.

[Embodiment 1]
[Overall configuration of solar power generation system]
FIG. 1 is a diagram schematically showing an example of the overall configuration of a solar power generation system 1 including an abnormality detection device 19 according to the present embodiment. However, the illustration and description of the configuration of the solar power generation system 1 that is not related to this embodiment will be omitted.

In the solar power generation system 1, power generation data of the solar cell module (solar power generation device) 2 is collected by a data collection unit 11, and the generated power data is analyzed and diagnosed via the Internet or 3G line, etc. The generated power is sent to the server 18, and the generated power analysis/diagnosis server 18 detects an abnormality in the generated power.

The solar cell module 2 is composed of a solar cell array in which a solar cell string in which a plurality of solar panels are wired in series is connected in parallel to obtain a predetermined output.The solar cell module 2 generates electric power according to the solar radiation intensity. Output.

Each solar cell string is housed in a junction box 3, respectively. The DC current collector box 4 collects DC power from the connection box 3 and relays it to a power conditioner 5.

The power conditioner 5 is an inverter that receives DC power from the DC current collection box 4 and converts it into stable AC power that can be used for power sale or power storage. The power conditioner 5 may be integrated with the connection box 3, and in this case, the DC current collection box 4 is not required.

The DC measurement unit 6 measures the power generated by the solar cell module 2 collected in the DC current collection box at predetermined intervals. The measured generated power is sent to the data collection unit 11.

The pyranometer 8 is installed near the solar cell module 2 and measures the solar radiation intensity of sunlight at predetermined intervals. The measured solar radiation intensity signal is sent to the weather conversion box 10.

Thermometer 9 is installed near solar cell module 2 and measures the temperature near solar cell module 2 . The temperature signal is sent to the weather conversion box 10.

The weather conversion box 10 A/D converts measurement signals such as signals related to solar radiation intensity and signals related to temperature, and sends the converted signals to a data collection unit 11 .

The data collection unit 11 collects measurement data such as solar radiation intensity, temperature, and generated power.

The data collected by the data collection unit 11 is sent to the cloud server 16 via the Internet, 3G, LTE line, etc. 15.

The converter 13 performs protocol conversion of the power generation information received from the power conditioner 5, etc. A programmable logic controller (also abbreviated as PLC) 14 collects calculation data from the DC measurement unit 6 and the weather conversion box 10 . The data collection/transmission unit 12 relays the data collected from the converter 13 and the PLC 14 and transmits it to the cloud server 16.

The cloud server 16 includes a visualization server 17, a generated power analysis/diagnosis server 18, and an abnormality detection device 19 of the present invention. The visualization server 17 displays solar radiation intensity data, generated power data, etc. collected from the data collection unit 11 on a display device (not shown) so that the data can be visually understood. The abnormality detection device 19 detects an abnormality in the solar cell module 2 by referring to the data acquired from the generated power analysis/diagnosis server 18 .

When the abnormality detection device 19 detects an abnormality, the customer may be notified of the abnormality, or a technician may be dispatched to the site to repair the solar cell module 2, or other measures may be taken. The configuration of the abnormality detection device 19 will be described below with reference to FIG. 2.

[Configuration of anomaly detection device]
Next, the configuration of the abnormality detection device 19 will be described with reference to FIG. 2. FIG. 2 is a block diagram showing the configuration of the abnormality detection device 19 according to this embodiment.

As shown in FIG. 2, the abnormality detection device 19 includes a first acquisition section 60, a second acquisition section 61, an estimation section 57, an abnormality detection section 62, and a storage section 50. Note that the storage unit 50 may be provided outside the abnormality detection device 19. Further, the configuration including the first acquisition section 60, the second acquisition section 61, and the estimation section 57 is also referred to as an "estimated model generation device."

The storage unit 50 stores solar radiation intensity data acquired from the data collection unit 11 and power generation data of the solar cell module 2. Therefore, the storage unit 50 stores the generated power data 51 of the previous month of the target date of determination, the generated power data of the same month in the past 52, the solar irradiance intensity data 53 of the previous month of the target date, and the solar irradiance intensity data 54 of the same month in the past. ing. Here, "the same month in the past" refers to the same month in the past of the month that includes the target date of determination. Examples include the same month in the previous year of the month that includes the target date, and two years before the month that includes the target date. This may include the same month, the same month three years before the month that includes the target date, etc.

The first acquisition unit 60 acquires generated power information indicating the generated power generated by the solar cell module 2. That is, the first acquisition unit 60 acquires the generated power information stored in the storage unit 50.

The second acquisition unit 61 acquires solar radiation intensity information indicating the solar radiation intensity measured by the pyranometer 8. That is, the second acquisition unit 61 acquires the solar radiation intensity information stored in the storage unit 50.

The estimation unit 57 estimates the generated power on the target day from the solar radiation intensity on the target day with reference to an estimation model that estimates generated power based on the solar radiation intensity.

The estimation model is generated with reference to the power actually generated under a predetermined sunshine condition and time and the actual solar radiation intensity under the predetermined sunshine condition and time. The predetermined sunshine condition includes that the solar radiation intensity is less than or equal to a predetermined threshold.
That is, an estimation model is generated by referring only to data on cloudy or rainy days. In this way, by generating an estimation model limited to data where the solar radiation conditions are below the threshold, we can reduce the effects of the following events (1) to (5) and the impact on the solar cell module 2 and pyranometer 8. A suitable estimation model that is less affected by shadows can be generated.
(1) When the solar power generation device is overloaded, that is, when the actual power generation output of the device is greater than the rated output of the PCS (power conditioning system) (2) When the PCS power factor control setting is performed according to the instructions of the electric power company. (3) When the power generation output of the PCS is suppressed due to a disturbance such as a rise in power grid voltage (4) When there is an output control instruction (instruction to stop power generation) by the power company (5) Self-consumption power generation When equipment output control is required (reduction of PCS output due to reduced domestic power demand during holidays, lunch break, etc.) Additionally, the above time periods do not include morning, evening, and night. If the data is from the daytime period, the incident angle of sunlight that enters the solar cell module 2 is relatively high, so the generated power is less affected by shadows on the solar cell module 2 and the pyranometer 8. Therefore, by limiting the target data to daytime data, a highly accurate estimation model can be generated.

The estimating unit 57 uses the estimation model to calculate the predetermined solar radiation condition and time from the solar radiation intensity indicated by the solar radiation intensity information acquired by the second acquisition unit 61 and in the predetermined solar radiation condition and time. Estimate the power generated in the zone. That is, the estimation unit 57 estimates the generated power on the target day from the solar radiation intensity on the target day using the estimation model generated based on the data with limited sunlight conditions and time zones described above. Thereby, the estimation unit 57 can estimate the generated power on the target day with high accuracy.

The estimation model may include a regression curve including a nonlinear term.
For example, the generated power (Y) may be expressed as a quadratic function of the solar radiation intensity (x) as shown below.
[Math. 1] Y=ax ² +bx+c
Note that the estimation unit 57 may include an estimation model generation unit 58 that generates an estimation model.

Here, in the abnormality detection device 19, the configuration including the first acquisition section 60, the second acquisition section 61, and the estimated model generation section 58 constitutes the estimated model generation device according to the present embodiment.

The abnormality detection unit 62 detects the estimated generated power that is the generated power estimated by the estimation unit 57 and the generated power indicated by the generated power information acquired by the first acquisition unit 60, which is the generated power under the predetermined sunshine conditions and time period. The presence or absence of an abnormality in the solar cell module 2 is determined by comparing the power and the power. For example, the abnormality detection unit 62 compares the estimated generated power obtained by substituting the solar radiation intensity measured on the target day into the equation shown in [Equation 1] above, and the actually measured generated power on the target day. Then, an abnormality in the solar cell module 2 is detected.

Here, the estimation model used by the abnormality detection unit 62 to determine abnormality on the target day may have an average solar radiation intensity threshold of 0.6 kW/m ² , for example. Since the events (1) to (5) above often occur under weather conditions where the solar radiation intensity is 0.6 kW/m ² or more, the predetermined threshold value is set to 0.6 kW/m ² or less. By doing so, it is possible to avoid the days when events (1) to (5) occur. However, it is more preferable that the solar radiation intensity threshold is 0.4 kW/m ² . Alternatively, a different solar radiation intensity threshold may be set for each season. For example, the threshold for average solar radiation intensity is set to 0.3 kW/m 2 from December to February, which is the period that includes the winter solstice, and 0.3 kW/m ² from March to May and from September to November, which are the periods that include the spring and autumn equinoxes. It may be set to 4 kW/m ² and 0.6 kW/m ² from June to August, which is the period including the summer solstice.

Further, the estimation model used by the abnormality detection unit 62 to determine abnormality on the target day may be, for example, any time period from 11:00 to 13:00.

FIG. 3 is a graph showing changes in the solar radiation intensity (kW/m ² , thick line) and the generated power (kW, thin line) of the solar cell module 2 on the target day. When generating an estimation model for generated power, for example, reference may be made only to data for 120 minutes from 11:00 to 12:59, which is included in frame A in FIG. 3. During this time period, the angle of incidence of sunlight that enters the solar cell module 2 is close to a right angle, and the generated power is hardly affected by shadows.

Furthermore, the estimation model used by the abnormality detection unit 62 to determine the abnormality on the target date may be an estimation model generated with reference to the generated power and solar radiation intensity in a plurality of periods in the past than the target date. By comparing the generated power in a plurality of past periods with the generated power on the target date, it is possible to accurately detect an abnormality in the solar cell module 2.

For example, the estimation model used by the anomaly detection unit 62 to determine an abnormality on a target date is based on the generated power and solar radiation intensity in the month preceding the month that includes the target date, and the power generation and solar radiation intensity in the past month that includes the target date. The estimation model may be generated with reference to electric power and solar radiation intensity. In this way, in addition to comparing the generated power with the previous month, by comparing with the estimated generated power of the same month in the past, it is possible to check how much the generated power fluctuates on the target day. In other words, the power generated by the solar cell module 2 is considered to gradually decrease due to aging of the device, etc., but in the same month in the past, conditions such as the angle of incidence of sunlight on the solar cell module 2 are almost the same. Therefore, it is effective to compare the generated power.

[Operation of abnormality detection device 19]
Next, an example of the flow of processing in the abnormality detection device 19 according to this embodiment will be described with reference to FIGS. 4 and 6. FIG. 4 is a flowchart showing the process of estimating the generated power, and FIG. 6 is a flowchart showing the process of determining abnormality of the solar cell module 2.

<Estimated flow of generated power>
First, with reference to FIG. 4, an example of the flow of the process of estimating the generated power in the solar cell module 2 will be described.

(Step S10)
In step S10, the estimating unit 57 obtains, from the storage unit 50, minute-by-minute solar radiation intensity data and generated power data for the month preceding the target date and the past month for the same month. Subsequently, the process advances to step S12.

(Step S12)
In step S12, the estimation unit 57 determines whether the acquired data is data for a predetermined sunshine condition and time zone. As the sunshine condition and time period, for example, 0.6 W/m ² or less and 120 minutes from 11:00 to 12:59 may be selected. Here, the predetermined sunshine condition is more preferably 0.4 kW/m ² . Alternatively, different values of sunlight conditions may be set depending on the season. If the acquired data is data for a predetermined sunshine condition and time zone (YES in step S12), the process advances to step S16. If the acquired data is not data for the predetermined sunshine conditions and time zone (NO in step S12), the process proceeds to step S14.

In addition, in this step, as an example, data in the predetermined time period (11:00 to 12:59) is extracted from the data acquired in step S10, and it is determined whether the extracted data satisfies the above sunshine condition. It is also possible to have a configuration in which the determination is made.

(Step S14)
In step S14, the estimation unit 57 does not refer to the acquired data.

(Step S16)
In step S16, the estimating unit 57 determines whether the acquired data is data for a period during which an instruction to stop power generation was issued by the electric power company. If the acquired data is not data during the power generation stop instruction period, equipment failure, or inspection stop period (NO in step S16), the process advances to step S20. If the acquired data is data during the power generation stop instruction period, equipment failure, or inspection stop period (YES in step S16), the process advances to step S18.

(Step S18)
In step S18, the estimation unit 57 does not refer to the acquired data.

(Step S20)
In step S20, an estimation model for estimating generated power from solar radiation intensity, that is, an approximate formula for estimated generated power, is derived for each of the previous month's data and the past same month's data by referring to the acquired data. . As an example of an approximate formula for the estimated generated power, the following formula can be obtained.
Y=ax ² +bx+c (x is solar radiation intensity, Y is estimated generated power)
Subsequently, the process advances to step S22.

(Step S22)
In step S22, the derived estimated estimated power generation power approximation formula for the data of the previous month and the estimated power generation power approximation formula for the past data of the same month are stored in the storage unit 50 for each solar cell module 2 (power plant, equipment).

(Step S24)
This completes the generated power estimation process.

<Correlation and approximate formula between solar radiation intensity and estimated generated power>
FIG. 5 is a graph in which the solar radiation intensity (kW/m ² ) is plotted on the horizontal axis and the actually measured generated power (kW) in the solar cell module 2 is plotted on the vertical axis. As shown in Figure 5, it has been shown that there is a high correlation between solar radiation intensity and generated power, and the function (Y = ax ² + bx + c) that estimates generated power (Y) from solar radiation intensity (x) is It is derived as a power estimation model.

For example, in FIG. 5, the following calculation formula was derived as an estimation model by substituting the solar radiation intensity and the actually measured generated power in the solar cell module 2.
Y=-4.0784×x ² +24.744×x+0.3414
(a=-4.0784, b=24.744, c=0.3414)
In the above example, the correlation coefficient between the solar radiation intensity (x) and the generated power (Y) was 0.9735, and a high correlation coefficient was obtained.

<Power generation equipment abnormality determination processing flow>
Next, the flow of the abnormality determination process for the solar cell module 2 on the target day will be described using the above-described estimation model (estimated power generation approximation formula) for the estimated power generation.

(Step S30)
In step S30, the first acquisition unit 60 and the second acquisition unit 61 acquire solar radiation intensity data and generated power data for each minute of the determination target day.

(Step S32)
In step S32, the estimation unit 57 determines whether the acquired data is data for a predetermined sunshine condition and time zone. As the sunshine condition and time period, for example, 0.6 kW/m ² or less and 120 minutes from 11:00 to 12:59 may be selected. Here, the predetermined sunshine condition is more preferably 0.4 kW/m ² . If the acquired data is data for a predetermined sunshine condition and time zone (YES in step S32), the process advances to step S36. If the acquired data is not data for the predetermined sunshine conditions and time zone (NO in step S32), the process advances to step S34.

In addition, in this step, as an example, data in the predetermined time period (11:00 to 12:59) is extracted from the data acquired in step S30, and it is determined whether the extracted data satisfies the above sunshine condition. It is also possible to have a configuration in which the determination is made.

(Step S34)
In step S34, the estimation unit 57 does not refer to the acquired data.

(Step S36)
In step S36, the estimating unit 57 determines whether the acquired data is data for a period during which an instruction to stop power generation was issued by the electric power company. If the acquired data is not data during the power generation stop instruction period, equipment failure, or inspection stop period (NO in step S36), the process advances to step S40. If the acquired data is data during the power generation stop instruction period, equipment failure, or inspection stop period (YES in step S36), the process advances to step S38.

(Step S38)
In step S38, the estimation unit 57 does not refer to the acquired data.

(Step S40)
In step S40, the estimated power generation amount on the target day is calculated by substituting the value of the solar radiation intensity data into the estimated power generation power approximation equation for the previous month and the estimated power generation power approximation equation for the past month.

(Step S42)
In step S42, the abnormality detection unit 62 compares the calculated estimated generated power and the actually measured generated power on the target day, and determines whether the degree of deviation is within a threshold value. For example, the threshold value of the deviation degree may be set to 7%. The actually measured generated power on the target day is compared with the estimated generated power obtained by substituting the solar radiation intensity on the target day into the estimated generated power approximation formula based on the previous month's generated power. Furthermore, the actually measured generated power on the target day is compared with the estimated generated power obtained by substituting the solar radiation intensity on the target day into the estimated generated power approximation formula based on the past generated power of the same month. In these two comparisons, if both degrees of deviation are less than or equal to the threshold value (YES in step S42), the process advances to step S44. In the above two comparisons, if any of the degrees of deviation exceeds the threshold value (NO in step S42), the process proceeds to step S46.

(Step S44)
In step S44, the abnormality detection unit 62 determines that the operation of the solar cell module 2 is normal.

(Step S46)
In step S46, the abnormality detection unit 62 determines that the operation of the solar cell module 2 is abnormal.

(Step S48)
This completes the abnormality detection process for the solar cell module 2.

As described above, according to the abnormality detection device 19 according to the present embodiment, when an event in which the generated power decreases due to a malfunction of the solar cell module 2 or the like occurs, an abnormality can be accurately detected. Therefore, for example, it is possible to take measures such as notifying the customer of an abnormality that has occurred in the solar cell module 2 and quickly recovering from the failure. Furthermore, by reducing the amount of data that is used when generating estimation models and determining abnormalities, it is expected that server capacity and processing time will be reduced. Furthermore, by accurately and early detecting the occurrence of abnormalities that result in electricity sales losses at solar power plants, it is possible to minimize the electricity sales losses of power generation companies (customers, investors). Furthermore, it is possible to reduce the need for individual personnel when determining whether or not there is an abnormality in the device, and to reduce maintenance costs for the power generation device.
[Embodiment 2]
In this embodiment, the abnormality detection device 19 derives a statistical value of the degree of deviation calculated for a plurality of solar cell modules 2 having the same configuration, and corrects the degree of deviation of each solar cell module 2 based on the statistical value. An example of detecting an abnormality in the solar cell module 2 will be described.

In the abnormality detection device 19 of this embodiment, the first acquisition unit 60 acquires power generation information indicating the power generated by each of the plurality of solar cell modules 2, and the second acquisition unit 61 acquires power generation information indicating the power generated by each of the plurality of solar cell modules 2. Solar radiation intensity information indicating the solar radiation intensity measured by the pyranometer 8 is acquired.

In this case, the abnormality detection unit 62 detects an abnormality for each of the plurality of solar cell modules 2 using the degree of deviation between the estimated generated power and the generated power indicated by the generated power information acquired by the first acquisition unit 60. Determine the presence or absence of.

However, depending on the weather conditions, the solar radiation intensity on the target day may change rapidly in a short period of time, and on days with such unstable weather conditions, the actual power generated may differ from the generated power estimated from the solar radiation intensity. It tends to be lower. Therefore, if abnormalities are detected by directly referring to the deviation value calculated for each solar power generation device, there is a possibility that many solar power generation devices will be determined to be abnormal even though there are no defects. There is.

For example, the left column of the table in FIG. 7 shows the degree of deviation between the actually measured generated power and the estimated generated power estimated from the estimation model for each of the 20 power conditioners (PCS1 to PCS20) with the same configuration according to this embodiment. show. When the deviation degree threshold is set to 7%, as shown in Figure 7, in this example, in 15 of the 20 solar power generation modules, the deviation exceeds the threshold of 7% and is considered abnormal. You will be judged.

Therefore, in this embodiment, when it is determined that a predetermined number or more of the solar cell modules 2 among the plurality of power conditioners 5 or solar cell modules 2 are abnormal, each of the plurality of power conditioners 5 or solar cell modules 2 The deviation degree regarding each power conditioner 5 or solar cell module 2 may be corrected with reference to the statistical value derived from the deviation degree in , and the presence or absence of an abnormality may be determined based on the corrected deviation degree.

For example, the median value -8.655 is derived from the degree of deviation calculated for each of the 20 power conditioners 5 shown in the left column of Table 7. Then, as shown in the right column of the table in FIG. 7, the degree of deviation in the power conditioner 5 is corrected using the median value of -8.655. That is, the value of (calculated deviation degree - median value of deviation degrees) is set as the corrected deviation degree, and abnormality determination of each power conditioner 5 is performed based on this corrected deviation degree. As the statistical value used for correction, in this embodiment, the median value of the generated power of the plurality of power conditioners 5 is referred to, but it is not limited to this, and for example, the average value of the generated power may be referred to.

Based on the corrected degree of deviation, three of the 20 power conditioners 5 (PCS1, PCS11, and PCS18) exceed 7%, which is the threshold value of the degree of deviation, and are determined to be abnormal.

In this way, if there are multiple solar power generation devices with the same configuration, calculate the statistical value such as the median of the calculated deviation degree of all the devices, and calculate the deviation after correction by (calculated deviation value - statistical value). By determining the abnormality of the power generation device based on the corrected deviation degree, unnecessary abnormality determination can be avoided even on days with unstable weather conditions.

In addition, in this embodiment, for a plurality of solar cell modules with the same configuration, an estimation model of generated power is generated for each solar cell module, the degree of deviation is calculated for each solar cell module, and abnormalities are detected for each solar cell module. An example of performing detection has been described. However, the present invention is not limited to this, and generates one estimation model by referring to data acquired from a plurality of solar cell modules (power generation devices and equipment). Abnormality may be detected for each solar cell module by referring to the above.

[Example of implementation using software]
The estimation unit 57 and the abnormality detection unit 62 of the abnormality detection device 19 may be realized by a logic circuit (hardware) formed in an integrated circuit (IC chip) or the like, or may be realized by software.

In the latter case, the estimating unit 57 and the abnormality detecting unit 62 are equipped with a computer that executes instructions of a program that is software that implements each function. This computer includes, for example, one or more processors and a computer-readable recording medium that stores the above program. In the computer, the processor reads the program from the recording medium and executes the program, thereby achieving the object of the present invention. As the processor, for example, a CPU (Central Processing Unit) can be used. As the recording medium, in addition to "non-temporary tangible media" such as ROM (Read Only Memory), tapes, disks, cards, semiconductor memories, programmable logic circuits, etc. can be used. Further, the computer may further include a RAM (Random Access Memory) for expanding the above program. Furthermore, the program may be supplied to the computer via any transmission medium (communication network, broadcast waves, etc.) that can transmit the program. Note that one aspect of the present invention can also be realized in the form of a data signal embedded in a carrier wave, in which the program is embodied by electronic transmission.

[Additional notes]
The present invention is not limited to the embodiments described above, and various modifications can be made within the scope of the claims, and embodiments obtained by appropriately combining technical means disclosed in different embodiments. are also included within the technical scope of the present invention.

1 Solar power generation system 2 Solar cell module 3 Connection box 4 DC collector box 5 Power conditioner
6 DC measurement unit 8 Pyranometer 9 Thermometer 10 Weather conversion box 11 Data collection unit 15 Internet, 3G, LTE line 16 Cloud server 17 Visualization server 18 Generated power analysis/diagnosis server 19 Abnormality detection device 50 Storage unit 51, 52 Power generation Power data 53, 54 Solar radiation intensity data 57 Estimating unit 58 Estimated model generating unit 60 First acquiring unit 61 Second acquiring unit 62 Abnormality detecting unit

Claims (8)

a first acquisition unit that acquires power generation information indicating power generated by the solar power generation device;
a second acquisition unit that acquires solar radiation intensity information indicating solar radiation intensity in the solar power generation device;
It is an estimation model that is generated by referring to the power actually generated in a given sunlight condition and time zone and the actual solar radiation intensity in the given sunlight condition and time zone, and the generated power is calculated based on the solar radiation intensity. Using an estimation model that estimates the solar radiation intensity indicated by the solar radiation intensity information acquired by the second acquisition unit, from the solar radiation intensity under the predetermined solar radiation condition and time zone, an estimation unit that estimates generated power;
Comparing the estimated generated power, which is the generated power estimated by the estimation unit, and the generated power, which is indicated by the generated power information acquired by the first acquisition unit, and which is generated under the predetermined sunshine conditions and time zone. It is equipped with an anomaly detection section that determines the presence or absence of an anomaly.
The predetermined sunshine condition includes that the solar radiation intensity is below a predetermined threshold,
The abnormality detection device does not include morning, evening, and night in the time period,
The first acquisition unit acquires power generation information indicating the power generated by each of the plurality of solar power generation devices,
The second acquisition unit acquires solar radiation intensity information indicating solar radiation intensity in each of the plurality of solar power generation devices,
The abnormality detection unit is
For each of the plurality of solar power generation devices, specifying the degree of deviation between the estimated generated power and the generated power indicated by the generated power information acquired by the first acquisition unit,
The deviation degree for each solar power generation device is corrected by referring to the statistical value derived from the deviation degree for each of the plurality of solar power generation devices, and the presence or absence of an abnormality is determined based on the corrected deviation degree. judge
An anomaly detection device characterized by: The estimation model used by the anomaly detection unit to determine anomaly on the target day is as follows:
The abnormality detection device according to claim 1, wherein the estimation model is generated with reference to generated power and solar radiation intensity in a plurality of periods in the past than the target date. The estimation model used by the anomaly detection unit to determine anomaly on the target day is as follows:
It is characterized by being an estimation model generated by referring to the generated power and solar radiation intensity in the month preceding the month that includes the target date, and the generated power and solar radiation intensity in the past month that includes the target date. The abnormality detection device according to claim 2. The abnormality detection device according to any one of claims 1 to 3, wherein the estimation model includes a regression curve including a nonlinear term. The predetermined threshold value is 0.6kW/ ^m2 ,
The abnormality detection device according to any one of claims 1 to 4, wherein the time slot is any time slot from 11:00 to 12:59. The abnormality detection device according to any one of claims 1 to 5, further comprising an estimated model generation unit that generates the estimated model. a first acquisition step of acquiring power generation information indicating the power generated by the solar power generation device;
a second acquisition step of acquiring solar radiation intensity information indicating solar radiation intensity in the solar power generation device;
It is an estimation model that is generated by referring to the power actually generated in a given sunlight condition and time zone and the actual solar radiation intensity in the given sunlight condition and time zone, and the generated power is calculated based on the solar radiation intensity. Using an estimation model that estimates the solar radiation intensity indicated by the solar radiation intensity information acquired in the second acquisition step, from the solar radiation intensity in the predetermined solar radiation condition and time period, an estimation step of estimating the generated power in;
Compare the estimated generated power that is the generated power estimated in the estimation step and the generated power indicated by the generated power information acquired in the first acquisition step under the predetermined sunshine conditions and time zone. an abnormality determination step of determining the presence or absence of an abnormality,
The predetermined sunshine condition includes that the solar radiation intensity is below a predetermined threshold,
The abnormality detection method does not include morning, evening, and night in the time period,
In the first acquisition step, power generation information indicating the power generated by each of the plurality of solar power generation devices is acquired;
In the second acquisition step, solar radiation intensity information indicating solar radiation intensity in each of the plurality of solar power generation devices is acquired,
In the abnormality determination step,
For each of the plurality of solar power generation devices, specifying the degree of deviation between the estimated generated power and the generated power indicated by the generated power information acquired in the first acquisition step,
The deviation degree for each solar power generation device is corrected by referring to the statistical value derived from the deviation degree for each of the plurality of solar power generation devices, and the presence or absence of an abnormality is determined based on the corrected deviation degree. judge
An anomaly detection method characterized by: An abnormality detection program for causing a computer to function as the abnormality detection device according to any one of claims 1 to 6 , the abnormality detection program for causing the computer to function as the estimation section and the abnormality detection section.

Images