JP7376270B2 - Solar power generation prediction system - Google Patents

Description

The present invention relates to a solar power generation amount prediction system that predicts the amount of power generated by a solar power generation unit.

Conventionally, the technology of a solar power generation amount prediction system that predicts the amount of power generation by a solar power generation unit is publicly known. For example, as described in Patent Document 1.

Patent Document 1 describes a power generation amount prediction device that predicts the amount of power generated by a solar power generation device.

The power generation amount prediction device described in Patent Document 1 is configured to predict the power generation amount of a solar power generation device using weather information and the power generation capacity (installed capacity) of the solar power generation device.

However, when predicting the power generation amount using the installed capacity of the solar power generation device, it is necessary to input the installed capacity of the solar power generation device for each direction, which may increase the burden of input work.

JP 2019-47584 Publication

The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a solar power generation amount prediction system that can reduce the burden of predicting the amount of power generated by a solar power generation unit. It provides:

The problem to be solved by the present invention is as described above, and next, means for solving this problem will be explained.

That is, in claim 1, there is provided a first predicted power generation amount calculation section that calculates a first predicted power generation amount of the solar power generation section based on prediction information including at least weather information; and a correction coefficient that corrects the first predicted power generation amount, a second predicted power generation amount calculation unit that calculates a second predicted power generation amount using the first predicted power generation amount, and a correction coefficient updating unit that can update the correction coefficient. The correction coefficient updating unit includes: a total of the actual power generation amount of the solar power generation unit collected during the data collection period; a total of the second predicted power generation amount collected during the data collection period; The correction coefficient is updated by multiplying the correction coefficient before updating by the ratio of .

In claim 2, the second predicted power generation amount calculation unit is configured such that, during the data collection period, a value obtained by dividing the total of the actual power generation amount by the second total of the predicted power generation amount is less than or equal to a first threshold value, or When the second threshold value, which is larger than the first threshold value, is exceeded, the second predicted power generation amount is calculated based on the correction coefficient before updating.

In claim 3, the correction coefficient updating unit does not update the correction coefficient when a value obtained by dividing the second total predicted power generation amount by the correction coefficient becomes a predetermined reference value or less. It is.

In claim 4, if the value obtained by dividing the total of the actual power generation amount by the second total of the predicted power generation amount is equal to or less than the first threshold value or equal to or more than the second threshold value, the value The apparatus includes a notification section that can notify that the current value is less than or equal to the first threshold value or greater than or equal to the second threshold value.

In claim 5, the data collection period includes a first data collection period in which the actual power generation amount and the second predicted power generation amount are collected for the first update of the correction coefficient, and a first data collection period in which the actual power generation amount and the second predicted power generation amount are collected for the first update of the correction coefficient. The actual power generation amount and the second predicted power generation amount for the update after the update are collected, and includes a second data collection period that is longer than the first data collection period.

The present invention has the following effects.

In the first aspect, it is possible to reduce the burden of predicting the amount of power generated by the solar power generation unit. Further, the predicted power generation amount of the solar power generation unit can be suitably calculated.

In the second aspect, the predicted power generation amount of the solar power generation unit can be suitably calculated.

In claim 3, if sufficient data for updating the correction coefficients has not been collected, the correction coefficients may not be updated.

In claim 4, it is possible to notify that there is a possibility that some kind of abnormality has occurred in the solar power generation unit.

According to claim 5, the first update of the correction coefficient can be performed relatively early.

1 is a block diagram showing the configuration of a power supply system in which a solar power generation amount prediction system according to an embodiment of the present invention is used. FIG. 2 is a block diagram showing the configuration of a control device. Flowchart showing processing executed by the solar power generation amount prediction system. A table showing an example of correction coefficients. 12 is a graph showing an example of an error between the corrected predicted power generation amount and the pre-correction predicted power generation amount with respect to the actual power generation amount.

Below, a solar power generation amount prediction system 1 according to an embodiment of the present invention will be described using FIGS. 1 and 2. The solar power generation amount prediction system 1 predicts the amount of power generated by the solar power generation unit 11 of a predetermined house.

Below, first, using FIG. 1, a power supply system A in which a solar power generation amount prediction system 1 is used will be explained. The power supply system A supplies power from a predetermined power supply source to a residence. The power supply system A is installed in a residence (such as a detached house or an apartment complex). The power supply system A mainly includes a power path L and a power storage system 10.

The power path L is a path that supplies power to the load H. Here, the load H is something that consumes electric power, such as electrical appliances included in the house. Power path L connects system power supply K and load H.

The power storage system 10 stores power generated using sunlight and supplies it to a load H. The power storage system 10 includes a solar power generation unit 11, a storage battery 12, and a hybrid power conditioner 13.

The solar power generation unit 11 is a device that generates power using sunlight. The solar power generation unit 11 is composed of a solar battery panel and the like. The solar power generation unit 11 is installed in a sunny place, such as on the roof of a house, for example.

The storage battery 12 is used to charge and discharge power from the system power supply K and power (generated power) from the solar power generation unit 11 as appropriate. The storage battery 12 is composed of, for example, a lithium ion battery. The storage battery 12 is connected to the solar power generation unit 11 by a predetermined power distribution line via a hybrid power conditioner 13, etc., which will be described later.

The hybrid power conditioner 13 is a device (hybrid power conditioner) that converts electric power as appropriate. The hybrid power conditioner 13 is capable of outputting the electric power generated by the solar power generation unit 11 and the electric power discharged from the storage battery 12 to the load H, and also outputs the electric power generated by the solar power generation unit 11 and the electric power from the grid power supply K. is configured to be able to output to the storage battery 12. Hybrid power conditioner 13 is connected to power path L through a predetermined power distribution line.

The power supply system A supplies the power generated by the solar power generation unit 11, the power discharged from the storage battery 12, and the power from the system power supply K to the load H according to the power demand of the load H.

Next, the solar power generation amount prediction system 1 will be explained. The solar power generation amount prediction system 1 can predict the amount of power generated by the solar power generation unit 11 of the power supply system A. The solar power generation amount prediction system 1 includes a control device 20 .

The control device 20 shown in FIG. 2 is capable of processing various types of information. Control device 20 is connected to hybrid power conditioner 13 via a predetermined communication means. Further, the control device 20 can acquire information on the amount of power generated by the solar power generation unit 11 via the hybrid power conditioner 13 . The control device 20 includes a storage section 21, a control section 22, a display section 23, and an input section 24.

The storage unit 21 stores various programs, information from the hybrid power conditioner 13, and the like. The storage unit 21 is composed of an HDD, RAM, ROM, and the like.

The control unit 22 executes a program stored in the storage unit 21. The control unit 22 is composed of a CPU.

The display section 23 displays various information. The display section 23 is composed of a liquid crystal display or the like.

The input unit 24 is for inputting various information. The input unit 24 includes a keyboard, a mouse, and the like.

In this way, a general personal computer or the like can be used as the control device 20. Further, the control device 20 may be installed in a residence or a business office that provides services related to the solar power generation amount prediction system 1.

The solar power generation amount prediction system 1 calculates the amount of power that the solar power generation unit 11 is expected to generate on a prediction target day (hereinafter referred to as a "prediction target date"). ) (kWh) can be executed.

Here, the predicted power generation amount is the power generation amount generated by the solar power generation unit 11 at each time on the prediction target day. Further, in this embodiment, the prediction target date is the day following the day on which the power generation amount prediction process is executed (hereinafter referred to as the "prediction execution date"). The power generation amount prediction process is repeatedly executed every day. Further, the power generation amount prediction process is executed at a certain time (for example, 7:00 p.m.) in a time period after the power generation time period (daytime) in which the solar power generation unit 11 generates electricity (time period after sunset).

In the power generation prediction process, as explained in step S18 of the flowchart of FIG. 3, which will be described later, the predicted power generation amount before correction (hereinafter referred to as "pre-correction predicted power generation amount") calculated based on the predicted solar radiation amount and the predicted temperature is calculated based on the predicted solar radiation amount and the predicted temperature. ) is multiplied by a correction coefficient to calculate the corrected predicted power generation amount (hereinafter referred to as "corrected predicted power generation amount").

Here, the predicted solar radiation amount is the predicted solar radiation amount, and is the solar radiation amount on the prediction target day in the area where the house is located. Further, as the predicted solar radiation amount, it is possible to employ global solar radiation amount, direct solar radiation amount, scattered solar radiation amount, and horizontal surface solar radiation amount.

The predicted temperature is the predicted outside temperature, and is the outside temperature on the prediction target day in the area where the house is located.

For the predicted amount of solar radiation and predicted temperature, weather prediction information such as a weather forecast transmitted from a public organization related to weather can be used. In this case, the weather forecast information is acquired via an appropriate communication means and is stored in the storage unit 21.

Further, the correction coefficient is a coefficient for correcting the predicted power generation amount before correction, taking into consideration the influence of factors such as deterioration of the solar power generation unit 11 on the prediction of the power generation amount other than the above-mentioned predicted solar radiation amount and predicted temperature.

Furthermore, in the power generation amount prediction process, the correction coefficient is updated as described in step S15 of the flowchart of FIG. 3, which will be described later. The correction coefficient is updated when a period (hereinafter referred to as the "data collection period") for collecting data (actual power generation amount and predicted power generation amount, described later) for updating the correction coefficient has elapsed. Executed every time.

In this embodiment, the data collection period up to the first update (initial update) of the correction coefficient will be referred to as an "adaptation period". The adaptation period is set to be a period during which a certain amount of data for performing the first update of the correction coefficient can be collected. In this embodiment, the acclimatization period is set to 7 days.

Furthermore, in this embodiment, the data collection period for updating the correction coefficient after the first update (normal update) will be referred to as a "normal update period". In this embodiment, the normal update period is one month. That is, in this embodiment, the correction coefficient is updated for the first time after the adaptation period (7 days) has elapsed, and thereafter, the normal update of the correction coefficient is repeatedly performed every time the normal update period (1 month) has elapsed.

Further, the correction coefficient is set to an initial value as a temporary value until the first update is performed. The initial value is desirably set to a relatively small value from the viewpoint of avoiding overestimation of the predicted power generation amount calculated using the initial value. In this embodiment, the initial value of the correction coefficient is 1 (kW). The initial value of the correction coefficient is input via the input section 24. The initial value of the correction coefficient is updated in the first update. Further, after the first update, the correction coefficient is updated every time the data collection period after the first update elapses. Note that a calculation method for updating the correction coefficient will be described later.

Below, the process performed by the control unit 22 in the power generation amount prediction process will be explained using the flowchart of FIG. 3.

In step S10, the control unit 22 acquires the actual power generation amount on the prediction execution date, and adds the actual power generation amount to the period performance total. Here, the actual power generation amount is the power generation amount actually generated by the solar power generation unit 11 in one day. Moreover, the period performance total is the cumulative value of the performance power generation amount in the data collection period (acclimation period or normal update period) to which the prediction execution date belongs. The period performance total is updated by adding the performance power generation amount in step S10.

The control unit 22 calculates the actual power generation amount of the solar power generation unit 11 on the predicted execution day based on the electric power generation amount of the solar power generation unit 11 acquired via the hybrid power conditioner 13, and also calculates the actual power generation amount on the predicted execution date. Add the amount to the period actual total. Further, the control unit 22 causes the storage unit 21 to store the period performance total. After executing the process of step S10, the control unit 22 moves to the process of step S11.

In step S11, the control unit 22 acquires the predicted power generation amount with the prediction execution date as the prediction target day (the predicted power generation amount on the prediction execution date calculated by the power generation amount prediction process on the previous day), and Add the predicted power generation amount to the period forecast total. Here, the predicted power generation amount is the corrected predicted power generation amount calculated by the process of step S18, which will be described later. Moreover, the period prediction total is the cumulative value of the predicted power generation amount in the data collection period (acclimation period or normal update period) to which the prediction execution date belongs. The period predicted total is updated by adding the predicted power generation amount in step S11.

The control unit 22 calls the predicted power generation amount (corrected predicted power generation amount) on the predicted execution date calculated on the previous day by the process of step S18 described later and stored in the storage unit 21, and also calls the predicted power generation amount on the predicted execution date. Add to the period forecast total. Further, the control unit 22 causes the storage unit 21 to store the period prediction total. After executing the process of step S11, the control unit 22 moves to the process of step S12.

In step S12, the control unit 22 determines whether the predicted execution date is after the end of the familiarization period (7 days). If the control unit 22 determines that the predicted execution date is after the end date of the familiarization period (step S12: YES), the process proceeds to step S13. On the other hand, if the control unit 22 determines that the predicted execution date is not after the end date of the familiarization period (step S12: NO), the process proceeds to step S18.

In step S13, the control unit 22 determines whether the correction coefficient has not been updated for the first time. That is, the control unit 22 determines whether the first update of the correction coefficient has not been performed yet.

When the control unit 22 determines that the correction coefficient has not been updated for the first time (step S13: YES), the process proceeds to step S14. On the other hand, when the control unit 22 determines that the correction coefficient has not been updated for the first time (step S13: NO), the process proceeds to step S19.

In step S14, the control unit 22 determines whether the value obtained by dividing the period prediction total by the correction coefficient is larger than a predetermined reference value α. Here, the reference value α is a value that serves as a reference for determining whether or not enough data for updating the correction coefficient has been collected. As the reference value α, a value corresponding to the average amount of power generation (kWh) of the solar power generation unit 11 over a predetermined period (for example, one week) can be adopted. In this embodiment, the reference value α is set to 10 kWh. If the value obtained by dividing the period forecast total by the correction coefficient is less than or equal to the reference value α, there is a possibility that sufficient data has not been collected due to some reason such as missing data.

When the control unit 22 determines that the value obtained by dividing the period prediction total by the correction coefficient is larger than the reference value α (step S14: YES), the process proceeds to step S15. On the other hand, when the control unit 22 determines that the value obtained by dividing the period prediction total by the correction coefficient is less than or equal to the reference value α (step S14: NO), the process proceeds to step S18.

In step S15, the control unit 22 updates the correction coefficient. The control unit 22 calculates a post-update correction coefficient using the period actual total, the period forecast total, and the pre-update correction coefficient.

The updated correction coefficient is calculated by the following formula (1).
Correction coefficient after update = period actual total / period forecast total × correction coefficient before update... (1)
After executing the process of step S15, the control unit 22 moves to the process of step S16.

In step S16, the control unit 22 determines whether the value obtained by dividing the period actual total by the period predicted total is greater than the first threshold value β1 and less than the second threshold value β2. Here, the first threshold value β1 and the second threshold value β2 determine whether the discrepancy between the period actual total and the period forecast total is large due to the period forecast total being too large or too small relative to the period actual total. This is the value to judge. In this embodiment, the first threshold value β1 is set to 0.5, and the second threshold value β2 is set to 2.0.

If the control unit 22 determines that the value obtained by dividing the period actual total by the period forecast total is greater than the first threshold β1 and less than the second threshold β2 (step S16: YES), the control unit 22 performs step S17. Move to processing. On the other hand, if the control unit 22 determines that the value obtained by dividing the period actual total by the period forecast total is less than or equal to the first threshold value β1 or greater than or equal to the second threshold value β2 (step S16: NO), Then, the process moves to step S20.

In step S17, the control unit 22 resets the period actual total and the period predicted total. That is, the control unit 22 erases all period performance totals and period prediction totals accumulated in the storage unit 21, and makes it possible to newly collect period performance totals and period prediction totals. In this way, by resetting the period's actual total and period's predicted total each time the correction coefficient is updated, the trend of the actual power generation amount and predicted power generation amount during the data collection period (acclimation period or normal update period) can be reflected in the correction coefficient. It can be made easier to reflect. After executing the process of step S17, the control unit 22 moves to the process of step S18.

In step S18, the control unit 22 calculates the predicted power generation amount (corrected predicted power generation amount) of the solar power generation unit 11.

The corrected predicted power generation amount of the solar power generation unit 11 is calculated by the following formula (2).
Predicted power generation amount after correction = (a0 + a1 x predicted solar radiation + a2 x predicted temperature) x correction coefficient... (2)

The value in parentheses in the above formula (2) indicates the predicted power generation amount before correction. The predicted power generation amount before correction is the power generation amount before correction per 1 kW of the solar power generation unit 11, which is calculated based on the predicted solar radiation amount, the predicted temperature, and predetermined coefficients a0, a1, and a2. That is, in step S18, the predicted power generation amount after correction is calculated by multiplying the predicted value of the power generation amount before correction by a correction coefficient.

Moreover, as the above-mentioned predetermined coefficients a0, a1, and a2, appropriate values that can calculate a desired value (pre-correction predicted power generation amount) using the above-mentioned formula (2) can be set. a0, a1, and a2 can be calculated, for example, based on the relationship between the past predicted solar radiation amount and predicted temperature, and the actual power generation amount of the solar power generation unit 11. In addition, as a method for calculating the above-mentioned a0, a1, and a2, an appropriate calculation method such as multiple regression analysis or machine learning such as a decision tree or a neural network can be adopted.

The control unit 22 causes the storage unit 21 to store the calculation result of the corrected predicted power generation amount. Further, the control unit 22 causes the display unit 23 to display the calculation result of the corrected predicted power generation amount. After executing the process of step S18, the control unit 22 ends the power generation amount prediction process.

In step S19, which is proceeded to when it is determined in step S13 that the correction coefficient has not been updated for the first time, the control unit 22 determines whether the predicted execution date is the update date. Here, the update date is the day on which the normal update period (1 month) passes after the first update (the day on which the normal update is performed).

When the control unit 22 determines that the predicted execution date is the update date (step S19: YES), the process proceeds to step S14. On the other hand, if the control unit 22 determines that the predicted execution date is not the update date (step S19: NO), the process proceeds to step S18.

In step S20, the control unit 2 determines whether it is the first update.

When the control unit 22 determines that it is the first update (step S20: YES), the process proceeds to step S17. On the other hand, if the control unit 22 determines that it is not the first update (step S20: NO), the process proceeds to step S21.

In step S21, the control unit 22 issues a warning. The control unit 22 is configured to provide notification by displaying a predetermined warning on the display unit 23. Note that the warning is not limited to the above-mentioned display, and a configuration that provides notification using an appropriate alarm, lamp, etc. may be adopted.

The above-mentioned warning can notify that there is a possibility that some abnormality has occurred since the discrepancy between the period actual total and the period predicted total is large. Factors that increase the discrepancy between the period's actual total and the period's forecast total include, for example, the occurrence of an abnormality such as a failure in the solar power generation section 11, or the fact that the area around the location where the solar power generation section 11 is installed. , changes in the external environment such as the construction of a building, etc. Further, in the present embodiment, in the first update where there is relatively little data for updating the correction coefficient (step S20: YES), when the discrepancy between the period actual total and the period forecast total is large (step S16: NO) However, the configuration is such that the warning is not executed. After executing the process of step S21, the control unit 22 moves to the process of step S17.

Further, below, an example of the power generation amount prediction process executed in the solar power generation amount prediction system 1 will be explained by dividing it into cases for each different prediction execution date. In addition, below, the process after adding the actual power generation amount to the period performance total and adding the predicted power generation amount to the period prediction total (step S10, step S11) will be explained.

First, a case where the prediction execution date is before the end date of the adaptation period will be described. In this case, the prediction execution date is not after the acclimatization period end date (step S12: NO), and sufficient data (actual power generation amount and predicted power generation amount) for performing the first update of the correction coefficient has not been collected.

Therefore, in this case, the control unit 22 calculates the predicted power generation amount (corrected predicted power generation amount) without updating the correction coefficient for the first time (step S18). In this case, since the correction coefficient has not been updated for the first time, the initial value is used as the correction coefficient.

Next, a case where the predicted execution date is the end date of the adaptation period will be described. In this case, the prediction execution date is after the familiarization period end date (step S12: YES). Moreover, in this case, it has not been updated for the first time (step S13: YES).

Therefore, in this case, if the value obtained by dividing the period prediction total by the correction coefficient is larger than the reference value α (10 kWh) (step S14: YES), the control unit 22 updates the correction coefficient as the initial value for the first time. (Step S15).

Further, the control unit 22 determines whether the value obtained by dividing the period actual total by the period predicted total is greater than the first threshold value β1 and less than the second threshold value β2 (step S16). Here, since this is the first update (step S20: YES), whether the value obtained by dividing the period actual total by the period forecast total is greater than the first threshold β1 and less than the second threshold β2. Regardless, the period actual total and the period predicted total are reset (step S17), and the predicted power generation amount (corrected predicted power generation amount) is calculated (step S18). In this case, the correction coefficient after the first update is used as the correction coefficient.

Note that if the value obtained by dividing the period forecast total by the correction coefficient is equal to or less than the reference value α (step S14: NO), the control unit 22 determines that sufficient data for performing the initial update of the correction coefficient has not been collected. Then, the predicted power generation amount (corrected predicted power generation amount) is calculated without updating the correction coefficient (step S18). In this case, since the correction coefficient has not been updated for the first time, the initial value is used as the correction coefficient.

Next, a case where the predicted execution date is after the end date of the familiarization period and before the update date of the first normal update will be described. In this case, the prediction execution date is after the end date of the familiarization period (step S12: YES), and the first update has been performed (step S13: NO). Furthermore, in this case, the predicted execution date is not the update date (step S19: NO).

Therefore, in this case, the control unit 22 calculates the predicted power generation amount (corrected predicted power generation amount) without normally updating the correction coefficient (step S18). In this case, the correction coefficient after the first update is used as the correction coefficient.

Next, a case where the predicted execution date is the update date of normal update will be described. In this case, the prediction execution date is after the end date of the familiarization period (step S12: YES), and the first update has been performed (step S13: NO). Moreover, in this case, the predicted execution date is the update date (step S19: YES).

Therefore, in this case, if the value obtained by dividing the period prediction total by the correction coefficient is larger than the reference value α (10 kWh) (step S14: YES), the control unit 22 normally updates the correction coefficient as the initial value. (Step S15).

Further, the control unit 22 determines whether the value obtained by dividing the period actual total by the period predicted total is greater than the first threshold value β1 and less than the second threshold value β2 (step S16). If the value obtained by dividing the period actual total by the period forecast total is greater than the first threshold β1 and less than the second threshold β2 (step S16: YES), the control unit 22 divides the period actual total and the period forecast. The total is reset (step S17), and the predicted power generation amount (corrected predicted power generation amount) is calculated (step S18). In this case, the correction coefficient after normal update is used as the correction coefficient.

Further, when the control unit 22 determines that the value obtained by dividing the period actual total by the period forecast total is less than or equal to the first threshold value β1 or greater than or equal to the second threshold value β2 (step S16: NO), determines whether it is the first update (step S20). This time, since it is a normal update (step S20: NO), the control unit 22 issues a warning (step S21). Further, the control unit 22 resets the period actual total and the period predicted total (step S17), and calculates the predicted power generation amount (corrected predicted power generation amount) (step S18).

According to the solar power generation amount prediction system 1 as described above, it is possible to calculate the predicted power generation amount by the solar power generation section 11 without inputting the installed capacity of the solar power generation section 11, which requires a relatively heavy input work. Become. In addition, the system calculates the corrected predicted power generation amount by multiplying the pre-correction predicted power generation amount by a correction coefficient. It becomes possible to calculate the predicted power generation amount.

Moreover, according to the solar power generation amount prediction system 1, the correction coefficient is updated every time the data collection period (acclimation period, normal update period) elapses, so that the solar power generation unit 11 may deteriorate over time, It is possible to calculate the predicted power generation amount (corrected predicted power generation amount) in consideration of the increase/decrease in the power generation amount by the solar power generation section 11 due to the installation state of the solar power generation section 11.

Further, in this embodiment, the correction coefficient is normally updated every month. Thereby, by updating the correction coefficient based on the actual power generation amount of the solar power generation unit 11 in the previous month, it is possible to reduce the error in the corrected predicted power generation amount throughout the year. The above effects will be explained below using FIGS. 4 and 5.

The table shown in FIG. 4 shows an example of the adaptation period and the correction coefficient for each month from January to December. In this embodiment, as described above, the correction coefficient for the adaptation period is set to the initial value of 1 (kW). In addition, the correction coefficients for each month are 4.1 (kW), 4.6 (kW), 4.5 (kW), 4.3 (kW), 4.3 ( kW), 3.9 (kW), 3.8 (kW), 3.8 (kW), 4.2 (kW), 4.3 (kW), 4.7 (kW), 5.1 (kW) ).

The graph shown in FIG. 5 shows the predicted power generation amount after correction using the correction coefficients for each month from January to December shown in FIG. This figure shows an example of the error in the actual power generation amount for each month. Note that the corrected predicted power generation amount, pre-correction predicted power generation amount, and actual power generation amount are the averages of the power generation amounts for each month. The above graph also shows the average error in the predicted power generation amount after correction and the predicted power generation amount before correction for the year (January to December). In the above graph, the closer the value indicating the error is to 0%, the smaller the error is.

As shown in FIG. 5, in most months, the error in the corrected predicted power generation amount was smaller than the error in the pre-correction predicted power generation amount (close to 0%). It was also shown that the annual average error in the corrected predicted power generation amount was smaller than the annual average error in the uncorrected predicted power generation amount. In other words, it was shown that the error in the predicted power generation amount after correction was smaller than that in the predicted power generation amount before correction throughout the year.

Furthermore, in this embodiment, the adaptation period (7 days) is shorter than the normal update period (1 month). Thereby, the first update of the correction coefficient can be performed relatively early. By making the normal update period relatively long, a relatively large amount of data (actual power generation amount and predicted power generation amount) can be used in the normal update of the correction coefficient.

In addition, according to the solar power generation forecasting system 1, a warning is issued when there is a large discrepancy between the period's actual total and the period's predicted total, so that some kind of abnormality such as a failure of the solar power generation unit 11 occurs. It is possible to notify that there is a possibility that the

As described above, the solar power generation amount prediction system 1 according to an embodiment of the present invention is
Calculating the pre-correction predicted power generation amount (first predicted power generation amount) of the solar power generation unit 11 based on the predicted solar radiation amount and the predicted temperature (prediction information including at least weather information) (step S18) First predicted power generation amount A quantity calculation unit (control unit 22),
Calculating the corrected predicted power generation amount (second predicted power generation amount) using the pre-correction predicted power generation amount and a correction coefficient for correcting the pre-correction predicted power generation amount (step S18) Second predicted power generation amount calculation (control unit 22);
It is equipped with the following.

Such a configuration can reduce the burden of predicting the amount of power generated by the solar power generation unit 11. That is, it is possible to calculate the predicted power generation amount (corrected predicted power generation amount) by the solar power generation section 11 without inputting the installed capacity of the solar power generation section 11, which requires a relatively large input work burden. In addition, the configuration is such that the predicted power generation amount after correction is calculated by multiplying the predicted power generation amount before correction by a correction coefficient, so it has no effect on the prediction of power generation amount other than the above predicted solar radiation and predicted temperature (prediction information). It becomes possible to calculate the predicted power generation amount (corrected predicted power generation amount) taking into account the

In addition, the solar power generation amount prediction system 1 is
The total amount of actual power generation of the solar power generation unit 11 (period performance total) collected during the data collection period (acclimation period, normal update period) and the total amount of actual power generation of the solar power generation unit 11 collected during the data collection period (acclimation period, normal update period) The apparatus includes a correction coefficient updating section (control section 22) that can update the correction coefficient (step S15) based on the ratio of the corrected predicted power generation total (period predicted total).

With such a configuration, the predicted power generation amount (corrected predicted power generation amount) of the solar power generation unit 11 can be suitably calculated. That is, by updating the correction coefficient based on the period actual total and the period forecast total, the amount of power generated by the solar power generation section 11 due to aging deterioration of the solar power generation section 11 or the installation state of the solar power generation section 11 can be reduced. It becomes possible to calculate the predicted power generation amount (corrected predicted power generation amount) taking into account the increase/decrease in .

Further, the correction coefficient updating unit (control unit 22)
If the value obtained by dividing the corrected total predicted power generation amount (period predicted total) by the correction coefficient is less than or equal to the predetermined reference value α (step S14: NO), the correction coefficient is not updated. .

With this configuration, if sufficient data for updating the correction coefficients has not been collected, the correction coefficients can be configured not to be updated.

In addition, the solar power generation amount prediction system 1 is
A second value in which the total of the actual power generation amount (period actual total) divided by the corrected total predicted power generation amount (period prediction total) is less than or equal to the first threshold β1 or greater than the first threshold β1. If the value is equal to or greater than the threshold β2 (step S16: NO), the display section 23 (step S21) capable of issuing a warning (that the value is equal to or less than the first threshold or equal to or greater than the second threshold) is displayed. It is equipped with a notification section).

With such a configuration, it is possible to notify that there is a possibility that some kind of abnormality has occurred in the solar power generation unit 11. That is, when the discrepancy between the period's actual total and the period's predicted total is large, it is possible to notify that some abnormality, such as a failure of the solar power generation unit 11, may have occurred by issuing an alarm.

In addition, the data collection period is
an adaptation period (first data collection period) during which the actual power generation amount and the corrected predicted power generation amount are collected for the first update of the correction coefficient;
The actual power generation amount and the corrected predicted power generation amount for the normal update (update after the first update) of the correction coefficient are collected, and the period is longer than the adaptation period (second data collection period);
This includes:

With such a configuration, the correction coefficient can be updated for the first time relatively quickly. Further, in the normal update of the correction coefficient, a relatively large amount of data (actual power generation amount and corrected predicted power generation amount) can be used.

Note that the control unit 22 according to the present embodiment is one form of the first predicted power generation amount calculation unit, the second predicted power generation amount calculation unit, and the correction coefficient updating unit according to the present invention.
Further, the pre-correction predicted power generation amount according to the present embodiment is one form of the first predicted power generation amount according to the present invention.
Further, the corrected predicted power generation amount according to the present embodiment is one form of the second predicted power generation amount according to the present invention.
Further, the predicted solar radiation amount and predicted temperature according to the present embodiment are one form of weather information according to the present invention.
Further, the adaptation period according to the present embodiment is one form of the first data collection period according to the present invention.
Further, the normal update period according to the present embodiment is one form of the second data collection period according to the present invention.
Further, the display section 23 according to the present embodiment is one form of the notification section according to the present invention.

Although the embodiments of the present invention have been described above, the present invention is not limited to the above configuration, and various changes can be made within the scope of the invention described in the claims.

For example, in the present embodiment, the equation in parentheses in the linear equation (2) is used as a calculation method for indicating the predicted power generation amount before correction, but the method is not limited to this. For example, the calculation result of a prediction model based on machine learning such as a decision tree or a neural network can be used as a calculation method showing the predicted power generation amount before correction.

Furthermore, in the present embodiment, the initial value of the correction coefficient is 1 (kW), but the present invention is not limited to such an aspect. Various values can be adopted as the initial value of the correction coefficient.

Further, in this embodiment, the adaptation period is set to 7 days, but it is not limited to such an aspect. Various periods (number of days) can be adopted as the acclimatization period.

Further, in this embodiment, the normal update period is one month, but it is not limited to such an aspect. Various periods (number of days) can be adopted as the normal update period.

Further, as the predicted solar radiation amount in this embodiment, for example, the total solar radiation amount can be adopted. Furthermore, the predicted amount of solar radiation is not limited to such a mode, and direct solar radiation, scattered solar radiation, horizontal surface solar radiation, and inclined surface solar radiation can be adopted.

Further, in this embodiment, the predicted solar radiation amount and the predicted temperature are used as the prediction information for calculating the predicted power generation amount before correction, but the present invention is not limited to such an aspect. As the prediction information, it is possible to employ weather information other than the predicted amount of solar radiation and the predicted temperature, such as humidity. Further, as the prediction information, in addition to weather information, various information such as the installed capacity of the solar power generation unit 11 can be employed.

Further, in the present embodiment, the period actual total and the period predicted total are reset each time the correction coefficient is updated, but the present invention is not limited to such an aspect. Instead of resetting the period's actual total and period's forecast total each time the correction coefficient is updated, the actual and predicted power generation amount can be calculated for a period (for example, one year) that exceeds the data collection period (acclimation period or normal update period). It may be accumulated. In this case, the correction coefficient can reflect the long-term trend of the actual power generation amount and the predicted power generation amount.

Further, in the present embodiment, the period actual total and the period predicted total are all reset, but the present invention is not limited to such an aspect. For example, a portion of the period actual total and the period predicted total may be reset.

Further, in the present embodiment, when a warning is notified (step S21), the predicted power generation amount is calculated using the updated correction coefficient (step S18), but the present invention is not limited to such a mode. . For example, when a warning is issued, the predicted power generation amount may be calculated without using the updated correction coefficient. In this case, the correction coefficient updated in step S15 may be returned to the value before the update, and the predicted power generation amount may be calculated using the correction coefficient before the update. Thereby, when the discrepancy between the period actual total and the period predicted total is large, the predicted power generation amount can be calculated without using the correction coefficient updated using the data.

Furthermore, in the present embodiment, the solar power generation amount prediction system 1 is configured to predict the amount of power generated by the solar power generation section 11 of a residence, but the present invention is not limited to such an aspect. The solar power generation unit 11 whose power generation amount is predicted by the solar power generation amount prediction system 1 is not limited to residences, but may be installed in business offices such as commercial facilities and factories, and public facilities such as government offices and parks. Good too.

1 Solar power generation prediction system 11 Solar power generation unit 22 Control unit (first predicted power generation calculation unit, second predicted power generation calculation unit, correction coefficient update unit)
23 Display section (notification section)

Claims (5)

a first predicted power generation amount calculation unit that calculates a first predicted power generation amount of the solar power generation unit based on prediction information including at least weather information;
a second predicted power generation amount calculation unit that calculates a second predicted power generation amount using the first predicted power generation amount and a correction coefficient that corrects the first predicted power generation amount;
a correction coefficient updating unit capable of updating the correction coefficient;
Equipped with
The correction coefficient updating unit includes:
The ratio between the total actual power generation amount of the solar power generation unit collected during the data collection period and the total of the second predicted power generation amount collected during the data collection period is set as the correction coefficient before updating. updating the correction coefficient by multiplying;
Solar power generation prediction system. The second predicted power generation amount calculation unit includes:
During the data collection period, the value obtained by dividing the total actual power generation amount by the second total predicted power generation amount is equal to or less than a first threshold value or equal to or greater than a second threshold value that is larger than the first threshold value. If so, calculate the second predicted power generation amount based on the correction coefficient before updating;
The solar power generation amount prediction system according to claim 1. The correction coefficient updating unit includes:
If the value obtained by dividing the second total predicted power generation amount by the correction coefficient is less than or equal to a predetermined reference value, the correction coefficient is not updated;
The solar power generation amount prediction system according to claim 2. If the value obtained by dividing the total of the actual power generation amount by the second total of the predicted power generation amount is less than or equal to the first threshold value or greater than or equal to the second threshold value, the value is equal to or less than the first threshold value. comprising a notification unit that can notify that the value is below or above the second threshold;
The solar power generation amount prediction system according to claim 2 or 3. The data collection period is
a first data collection period during which the actual power generation amount and the second predicted power generation amount are collected for the first update of the correction coefficient;
a second data collection period in which the actual power generation amount and the second predicted power generation amount for updating after the first update of the correction coefficient are collected, and which is longer than the first data collection period;
including,
The solar power generation amount prediction system according to any one of claims 2 to 4.

Images