JP2022183783A - Information processing device, information processing method, program, and power generation system - Google Patents

Abstract

To predict an amount of power generated by a power plant.SOLUTION: An information processing device includes a position calculation unit, a power amount calculation unit, and an output unit. The position calculation unit calculates a target position of a first time zone result value of an amount of power generated by one or more power plants on a target day with respect to a first sample group including a plurality of first time zone result values of an amount of power generated by the one or more power plants on a plurality of reference days. The power amount calculation unit calculates a power amount corresponding to the target position, in a second sample group including a plurality of prediction target time zone result values of an amount of power generated by the one or more power plants on the plurality of reference days. The output unit outputs the power amount corresponding to the target position, as a predicted power amount which is predicted to be generated by the one or more power plants in the prediction target time zone.SELECTED DRAWING: Figure 1

Description

TECHNICAL FIELD Embodiments of the present invention relate to an information processing device, an information processing method, a program, and a power generation system.

The method of purchasing electricity generated by photovoltaic power generation will shift from a fixed-price purchase system (FIT) to a market-price-linked purchase system (FIP). In the purchase system linked to the market price, in order to match the planned value and the actual value of the generated power, for example, utilization of the one-hour-ahead market, in which power is traded one hour before the power supply, is being considered. For this reason, in the purchase system linked to the market price, there is a demand for high-precision short-term prediction technology for predicting the amount of power to be generated in the relatively near future.

Conventionally, as short-term forecasting techniques in photovoltaic power generation, there are a method of using only actual power generation values and a method of using weather satellites and weather forecasts in addition to actual power generation values.

As a method of using only actual power generation values, a continuous forecasting method of a clear weather index is known. However, the continuous forecasting method of the clear weather index is complicated in mutual conversion between solar radiation intensity and power generation amount, and the error is large.

As a method of using weather satellites and weather forecasts in addition to actual power generation values, a method of searching for days with similar weather is known. However, this method complicates data processing as the amount of data acquired from the outside increases, and as a result, the risk of problems such as failure to acquire data in real time also increases.

Japanese Patent Application Laid-Open No. 2020-108188 JP 2013-84736 A Japanese Patent No. 6785971

Stuart Coles, “An Introduction to Statistical Modeling of Extreme Values,” Springer, 2001 Sokol Dervishi and Ardeshir Mahdavi, "Computing diffuse fraction of global horizontal solar radiation: A model comparison", in Solar Energy. 2012 Jun; 86(6): 1796－1802.

The problem to be solved by the present invention is to provide an information processing device, an information processing method, a program, and a power generation system that easily and accurately predict the amount of electric power generated by a power plant using a small number of types of data. be.

An information processing apparatus according to an embodiment includes a position calculator, a power amount calculator, and an output unit. The position calculation unit performs power generation by the one or more power plants with respect to a first sample group including a plurality of actual values of the amount of electric power generated by the one or more power plants in a first time period on a plurality of reference days. The target position of the actual value for the first time period on the target day of the electric energy calculated is calculated. The electric energy calculation unit corresponds to the target position in a second sample group including a plurality of actual values of the electric energy generated by the one or more power plants in the prediction target time period on the plurality of reference days. Calculate the amount of power. The output unit outputs the electric energy corresponding to the target position as a predicted electric energy expected to be generated by the one or more power plants in the prediction target time period.

The figure which shows the functional structure of the prediction apparatus which concerns on 1st Embodiment. A flow chart showing a flow of processing of a prediction device. FIG. 4 is a diagram showing actual values of electric energy generated by a power plant for each time zone; The figure which shows a prediction object time zone. The figure which shows the actual value of the 1st time slot|zone of an object day. The figure which shows the actual value of the 1st time slot|zone of several reference days. The figure which shows the actual value of the prediction object time zone of several reference days. The figure which shows an example of the change of an actual value. The figure which shows the functional structure of the prediction apparatus which concerns on 2nd Embodiment. The figure which shows the block structure of a power plant model. Numerical examples of Erbs, Orgill, and Reindl, which are linear separation models. The figure which shows the calculation example of the direct separation model at the time of using Orgill. The figure which shows the example of a numerical value of a sun direction. The figure which shows the example of a numerical value of an inclined surface solar radiation intensity. The figure which shows the calculation example of a sun direction. A diagram showing the difference between the module temperature and the ambient air temperature. The figure showing the conversion efficiency of the inverter with respect to a load factor. 4 is a flowchart showing the flow of processing of a power plant model; The figure which shows the hardware constitutions of a prediction apparatus.

A prediction device 10 according to an embodiment will be described below with reference to the drawings. The prediction device 10 calculates the amount of power generated by one or more power plants in a prediction target time period based on past power amounts generated by one or more power plants.

(First embodiment)
FIG. 1 is a diagram showing the functional configuration of a prediction device 10 according to the first embodiment. The prediction device 10 is, for example, a single computer or a computer such as a server device. The prediction device 10 may be one computer, or may be composed of a plurality of computers like a cloud system. The computer functions as the prediction device 10 by executing a given program.

The prediction device 10 includes a reception unit 22, a storage unit 24, an input unit 26, a target sample acquisition unit 28, a first sample group acquisition unit 30, a second sample group acquisition unit 32, and a performance value correction unit 34. , a position calculator 36 , a power amount calculator 38 , and an output unit 40 .

The receiving unit 22 receives the actual value of the amount of electric power generated by the power plant for each time period. The power plant is a photovoltaic power plant in this embodiment. It should be noted that the power plants may be other types of power plants such as hydroelectric power plants, wind power plants or geothermal power plants. Also, the power plant may be a facility installed by a business operator such as an electric power company, or may be a private power generator installed in a home or the like.

In addition, the receiving unit 22 may receive actual values for each time period of the amount of power generated by a plurality of power plants that can estimate the power to be generated by the same power generation model. For example, the receiving unit 22 detects the amount of power generated by a plurality of photovoltaic power plants that are located close to each other and use the same equipment, in which the energy obtained from sunlight is almost the same and the weather conditions are the same. You may receive the performance value for each.

The storage unit 24 stores the performance value for each time period received by the reception unit 22 . For example, the storage unit 24 is managed by a database and stores a set of dates, time zones, and performance values. By designating the year, month, day and time period, the storage unit 24 retrieves and outputs the actual values for the designated year, month, day and time period. The storage unit 24 may be implemented within the prediction device 10 or may be implemented as a database server or the like external to the prediction device 10 .

The input unit 26 receives an input of the prediction target time period on the prediction date. The input unit 26 may accept, for example, a prediction target time period input by an operation by a user, may accept a prediction target time period received from another device, or may receive a prediction target time period automatically input periodically. Obi may be accepted. The input unit 26 provides the received prediction target time period to the second sample group acquisition unit 32 and the actual value correction unit 34 .

The target sample acquiring unit 28 obtains actual values in one time period before the current time, among the actual values for each time period stored in the storage unit 24, during which substantial power generation was performed. to get For example, the target sample acquiring unit 28 acquires the actual value in the most recent time period in which substantial power generation was performed among the actual values for each time period stored in the storage unit 24 . Subsequently, the target sample acquiring unit 28 specifies the time zone of the acquired performance values as the first time zone. In addition, the target sample acquisition unit 28 identifies the power generation date of the acquired actual value as the target date. The target sample acquisition unit 28 provides the specified first time period to the first sample group acquisition unit 30 and the performance value correction unit 34 . In addition, the target sample acquisition unit 28 gives the acquired performance value to the position calculation unit 36 as the performance value for the first time slot on the target day.

The first sample group acquisition unit 30 acquires from the storage unit 24 a first sample group including a plurality of actual values in a first time period on a plurality of reference dates. The multiple reference dates are multiple days selected under a predetermined rule prior to the forecast date. The first sample group acquisition unit 30 provides the first sample group to the performance value correction unit 34 .

The second sample group acquisition unit 32 acquires from the storage unit 24 a second sample group including a plurality of actual values for prediction target time periods on a plurality of reference dates. The plurality of reference dates from which the second sample group acquisition unit 32 acquires the actual values are the same as the actual values for the first time period acquired by the first sample group acquisition unit 30 . The second sample group acquisition unit 32 provides the second sample group to the performance value correction unit 34 .

The actual value correction unit 34 converts each of the plurality of actual values included in the first sample group into a target date calculated value calculated based on a predetermined power plant model and a reference date calculated based on the power plant model. Corrected by the ratio to the calculated value. The power plant model is a model that calculates the power generated by the power plant or a value proportional to the power generated by the power plant. The target day calculated value is the target day power generated by the power plant in the corresponding time period on the target day, or a value proportional to the target day power, which is calculated based on the power plant model. In addition, the reference day calculation value is proportional to the reference day power or the reference day power generated by the power plant in the corresponding time period on the corresponding reference day of the plurality of reference days calculated based on the power plant model. value.

Further, the actual value correcting unit 34 converts each of the plurality of actual values included in the second sample group into the target date calculated value calculated based on the power plant model and the reference date calculated value calculated based on the power plant model. Corrected by the ratio of The actual value correction unit 34 gives the corrected first sample group to the position calculation unit 36 . In addition, the actual value correction unit 34 supplies the corrected second sample group to the power amount calculation unit 38 .

The position calculation unit 36 calculates the amount of electric power generated by the power plant on the target day for the first sample group including a plurality of actual values of the amount of electric power generated in the first time period on a plurality of reference days. Calculate the target position of the actual value for one hour. In this embodiment, the target position is the actual value of the first time period on the target date in the data string in which the multiple actual values included in the first sample group are arranged in the first order of either ascending order or descending order. is the order of power consumption corresponding to .

Note that the ranking may be expressed not only by natural numbers but also by real numbers. Also, the rank may be a percentile rank normalized between 0% and 100%. The position calculator 36 provides the calculated target position to the power amount calculator 38 .

The power amount calculation unit 38 calculates the power amount corresponding to the target position in the second sample group including a plurality of actual values of the power amount generated by the power plant in the prediction target time period on a plurality of reference days. Here, the power amount corresponding to the target position is the power amount corresponding to the target position in the data string in which the multiple actual values included in the second sample group are arranged in the first order.

In the present embodiment, the power amount calculation unit 38, as the power amount corresponding to the target position, in the data string in which the plurality of actual values included in the second sample group are arranged in the first order, at the target position (order) Calculate the corresponding power consumption. The power amount calculator 38 provides the calculated power amount to the output unit 40 .

The output unit 40 outputs the power amount corresponding to the target position received from the power amount calculation unit 38 as the predicted power amount for the prediction target time zone to be generated.

FIG. 2 is a flowchart showing the processing flow of the prediction device 10. As shown in FIG. The flow of processing of the prediction device 10 will be described with reference to the flowchart shown in FIG. 2 and FIGS. 3 to 7. FIG.

First, as shown in FIG. 3, the storage unit 24 stores the actual value of the amount of electric power generated by the power plant for each time period. In the present embodiment, the storage unit 24 stores power consumption for every 30 minutes. More specifically, in the present embodiment, the storage unit 24 stores the electric energy from 00 minutes 00 seconds to 29 minutes 59 seconds and from 30 minutes 00 seconds to 59 minutes 59 seconds at each time. Note that the time period is not limited to every 30 minutes, and may be any time unit such as every 5 minutes, 10 minutes, or 60 minutes.

In S<b>11 , the input unit 26 receives an input of the prediction target time period on the prediction date. The prediction target time period is the time period for which the electric energy generated by the power plant is to be predicted. The input unit 26 accepts a future time period from the current time as a prediction target time period on the prediction date. The prediction target time zone corresponds to the time unit for obtaining the actual value. For example, the input unit 26 receives the prediction target time period on the prediction date as shown in FIG. 4 at the current time as shown in FIG.

Subsequently, in S12, the target sample acquisition unit 28 selects the actual value for each time period stored in the storage unit 24 for the time period prior to the current time, when the power generation was substantially performed. Get the actual value for one time period. In the present embodiment, the target sample acquiring unit 28 acquires the actual value in the latest time period in which substantial power generation was performed among the actual values for each time period stored in the storage unit 24 .

The time period during which substantial power generation is performed does not mean the time period in which the actual value is 0 or less than or equal to the predetermined value, but the time period in which the actual value is 0 or greater than the predetermined value. For example, if the current time is 13:10, and power is generated from 12:30 to 12:59 on the predicted date with an amount of power that is greater than the predetermined value, the target sample acquisition unit 28 acquires Then, the actual values from 12:30 to 12:59 on the forecast day are acquired. Further, for example, if the current time is 8:00 and power generation by the power plant is stopped from 17:00 to 8:59 the next morning, the target sample acquisition unit 28 Acquire the actual value for the time period from 30 minutes to 16:59.

Subsequently, in S<b>13 , the target sample acquisition unit 28 specifies the time zone of the acquired performance values as the first time zone. In addition, the target sample acquisition unit 28 identifies the power generation date of the acquired actual value as the target date. For example, when the actual value of the predicted date is obtained, the target sample obtaining unit 28 identifies the predicted date as the target date. Further, for example, when the actual values for the day before the prediction date are obtained, the target sample obtaining unit 28 identifies the day before the prediction date as the target date.

Subsequently, in S<b>14 , the first sample group acquisition unit 30 acquires from the storage unit 24 a first sample group including a plurality of actual values for the first time period on a plurality of reference dates. The multiple reference dates are multiple days selected under a predetermined rule prior to the forecast date. In this embodiment, the multiple reference dates are 30 days included in the period from 1 day to 30 days before the predicted date. For example, if the prediction date is May 25, 2021, the first sample group acquisition unit 30 collects 30 days from April 25, 2021 to May 24, 2021, as shown in FIG. are identified as a plurality of reference dates, and the actual values for the first time period on these days are obtained. Note that the multiple reference dates are not limited to such a period, and may be multiple days selected under other rules.

Subsequently, in S<b>15 , the second sample group acquisition unit 32 acquires from the storage unit 24 a second sample group including a plurality of actual values for the prediction target time period on a plurality of reference dates. The plurality of reference dates from which the second sample group acquisition unit 32 acquires the actual values are the same as the actual values for the first time period acquired by the first sample group acquisition unit 30 . For example, when the first sample group acquisition unit 30 acquires actual values for the first time period of 30 days from April 25, 2021 to May 24, 2021, the second sample group acquisition unit As shown in FIG. 7, 32 acquires the actual values of the prediction target time period for 30 days from April 25, 2021 to May 24, 2021.

Subsequently, in S16, the performance value correction unit 34 converts each of the plurality of performance values included in the first sample group into the target day calculation value calculated based on the predetermined power plant model and the power plant model. Corrected by the ratio with the reference date calculated value calculated based on More specifically, the actual value correcting unit 34 combines each of the multiple actual values included in the first sample group with the target day calculated value for the first time slot on the target day and the first hour on the corresponding reference date. It is corrected by multiplying the ratio with the reference date calculation value of the band.

For example, the performance value correcting unit 34 makes corrections as shown in Equation (1).

Actual value for the 1st time period after correction = Actual value for the 1st time period before correction x (Calculated value for the 1st time period on the target date / Calculated value for the 1st time period on the corresponding reference date ) (1)

Here, the power plant model is a model that calculates the power generated by the power plant or a value proportional to the power generated by the power plant. The target day calculated value is the target day power generated by the power plant in the corresponding time period on the target day, or a value proportional to the target day power, which is calculated based on the power plant model. In addition, the reference day calculated value is the reference day power generated by the power plant in the corresponding time period on the corresponding reference day of the plurality of reference days calculated based on the power plant model, or the reference day power It is a proportional value.

If the power plant is a solar power plant, for example, the power plant model calculates the power generated by the power plant at a specified time on the target date or reference date based on facility information and weather information. Calculated as a value or reference date calculation value.

Equipment information includes layout angle, module capacity, and inverter capacity. The layout angle is the normal direction of the solar cell modules installed in the solar power plant. The layout angle may be represented, for example, by the orientation and tilt angle of the solar cell module. Module capacity is the maximum output of a solar cell module installed in a solar power plant. Inverter capacity is the maximum output of a DC to AC converter in a photovoltaic power plant.

The equipment information does not change for each day and for each time zone, and does not affect the ratio between the target date calculation value and the reference date calculation value. Therefore, the power plant model may use standardized values that predetermine layout angles, module capacities, and inverter capacities. In this case, in the power plant model, the orientation of the solar cell module may be south (0°) and the tilt angle of the solar cell module may be approximately 30° to 40°. In the power plant model, the capacity of the solar cell module may be approximately 1.0 to 2.0, and the inverter capacity may be approximately 1.0. If the latitude of the installation position of the power plant is high, the power plant model may have the inclination angle of the solar cell module larger than 30° to 40°.

The weather information includes, for example, all-sky horizontal solar radiation intensity, upper atmospheric horizontal solar radiation intensity, sun direction, ground surface albedo, temperature, wind speed, and the like.

The global solar radiation intensity is the solar radiation intensity for the horizontal plane. The horizontal surface solar radiation intensity at the top of the atmosphere is the horizontal surface solar radiation intensity at the top of the atmosphere. The sun direction is the solid angle of the direction of the sun with respect to the installation position of the power plant, represented by the azimuth and altitude of the sun with respect to the installation position of the power plant. The surface albedo is the rate at which the surface of the earth reflects sunlight.

The actual value correction unit 34 may acquire the latitude, longitude and time of the location of the power plant instead of the solar direction, and calculate the solar direction based on the latitude, longitude and time of the location of the power plant. Further, the actual value correcting unit 34 can acquire the all-sky horizontal solar radiation intensity, the upper atmospheric horizontal solar radiation intensity, and the surface albedo from a weather database, a weather providing service server, or the like. In addition, the actual value correction unit 34 can acquire the air temperature and the wind speed from measuring equipment provided at the power plant. Also, the actual value correction unit 34 may acquire the temperature and wind speed from a weather database or a weather providing service server.

Note that the power plant model may calculate the power generated by the power plant using a preset value for the ground surface albedo when the ground surface albedo cannot be obtained. In addition, when the power plant model cannot obtain the global solar radiation intensity or the upper horizontal solar radiation intensity, the average value of the area where the power plant is installed is used as the global solar radiation intensity or the upper horizontal solar radiation intensity. A preset value may be used to calculate the power generated by the power plant. The power plant model may also calculate the power generated by the power plant using past average values when temperature or wind speed cannot be obtained.

Moreover, the solar radiation intensity with respect to the solar cell module provided in the solar power plant is called the inclined plane solar radiation intensity. Inclined solar radiation intensity is approximately proportional to the power generated by the photovoltaic power plant. Therefore, the power plant model may calculate the slope solar radiation intensity as a value proportional to the power generated by the photovoltaic power plant. In this case, the power plant model calculates the slope solar radiation intensity at the specified time on the target date or reference date as the target date calculation value or reference date calculation value.

When calculating the sloped surface solar radiation intensity, the solar cell module uses facility information and the all-sky horizontal solar radiation intensity, top-of-atmosphere horizontal surface solar radiation intensity, solar direction, and ground surface albedo among the weather information. Even in this case, if the ground surface albedo cannot be acquired, the power plant model may calculate the inclined plane solar radiation intensity using a value preset as the ground surface albedo. In addition, when the power plant model cannot obtain the global solar radiation intensity or the upper horizontal solar radiation intensity, the average value of the area where the power plant is installed is used as the global solar radiation intensity or the upper horizontal solar radiation intensity. A preset value may be used to calculate the slanted surface solar radiation intensity.

An example of the power plant model will be described later in detail with reference to FIGS. 10 to 18. FIG.

Subsequently, in S17, the actual value correction unit 34 sets each of the plurality of actual values included in the second sample group to the target day calculated value calculated based on the power plant model and the calculated value calculated based on the power plant model. Corrected by the ratio with reference date calculation value. More specifically, the actual value correcting unit 34 sets each of the plurality of actual values included in the second sample group to the target day calculated value for the forecast target time period on the target day and the forecast target time on the corresponding reference date. It is corrected by multiplying the ratio with the reference date calculation value of the band.

For example, the performance value correcting unit 34 makes corrections as shown in Equation (2).

Actual value of the forecast target time period after correction = Actual value of the forecast target time period before correction x (calculated value of target day of forecast time period on target day / calculated value of reference date of target time period of corresponding reference date ) (2)

The power plant model is the same as the model obtained by correcting a plurality of actual values included in the first sample group in S16.

Subsequently, in S18, the position calculation unit 36 calculates the target position of the actual value of the amount of power generated by the power plant in the first time period on the target day for the corrected first sample group.

Here, the target position is the performance of the first time period on the target day in the data string in which the multiple performance values included in the corrected first sample group are arranged in the first order of either ascending order or descending order. It is the order of the power amount corresponding to the value. In the present embodiment, the position calculation unit 36 calculates, as the target position, the power corresponding to the actual value of the first time period on the target day in the data string in which the multiple actual values included in the first sample group are arranged in ascending order. Calculate quantity ranking.

For example, when the first sample group includes N actual values (N is an integer equal to or greater than 2), the position calculation unit 36 generates a data string in which the N actual values are arranged as shown in Equation (3). do.

Y[1]≤Y[2]≤Y[3]≤...≤Y[N-2]≤Y[N-1]≤Y[N] (3)

Here, Y[·] represents the actual value included in the first sample group. The brackets of Y[·] indicate the order.

The position calculation unit 36 compares, for example, the data string arranged in this way with the actual values for the first time slot on the target day, and calculates the order, which is the target position.

As an example, the position calculation unit 36 detects the closest value to the actual value for the first time slot on the target day in the data string. Then, the position calculation unit 36 may use the order of the detected values as the target position. In addition, when the data string contains a plurality of values that are the same as the actual value of the first time period on the target date, the position calculation unit 36 determines the rank of the midpoint of the plurality of the same values as the target position. may be In this case, the position calculation unit 36 uses the order represented by the real number as the target position.

As another example, the position calculation unit 36 calculates, among the plurality of values included in the data string, the highest ranking value that is equal to or lower than the actual value for the first time period on the target day, Get the lowest rank value equal to or greater than the actual value. Then, the position calculation unit 36 may use the order obtained by performing linear interpolation based on the two acquired values as the target position.

Specifically, let X be the actual value of the first time slot on the target day, let Y[n] be the highest ranking value below the actual value of the first time slot on the target day included in the first sample group, Let Y[n+1] be the lowest ranking value equal to or higher than the actual value for the first time period on the target day included in the first sample group. In this case, the magnitude relationship among X, Y[n] and Y[n+1] is given by Equation (4).
Y[n]≤X≤Y[n+1] (4)

Then, the position calculation unit 36 calculates nr, which is the ranking corresponding to the actual value of the first time period on the target day, by calculating the expression (5).
nr=n×{(Y[n+1]−X)/(Y[n+1]−Y[n])}+(n+1)×{X−(Y[n])/(Y[n+1]−Y[n ])}…(5)

Note that the rank may be a percentile rank normalized between 0% and 100%.

Further, as another example, the position calculation unit 36 sets each of two or more predetermined number of time periods before the latest time period as the first time period, and sets two or more predetermined number of time periods as the first time period. A position may be calculated. In addition, when the position calculation unit 36 has already calculated the target position for another time period in the past prediction processing of the predicted power amount, the position calculation unit 36 may be read. Then, the position calculation unit 36 calculates the average value of the calculated predetermined number of target positions, and outputs the calculated average value of the predetermined number of target positions as the target position of the actual value for the first time period on the target date. . As a result, the position calculation unit 36 can calculate the predicted power amount with high accuracy even when the actual value in the most recent time zone changes abruptly. For example, when the actual value in the most recent time period has changed by a predetermined value or more from the time period immediately before the most recent time period, the position calculation unit 36 calculates such an average value as the first hour of the target day. You may output as a target position of the actual value of a band.

Subsequently, in S19, the power amount calculator 38 calculates the power amount corresponding to the target position in the corrected second sample group. Here, the power amount corresponding to the target position is the power amount corresponding to the target position in the data string in which the multiple actual values included in the second sample group are arranged in the first order.

In this embodiment, the power amount calculator 38 corresponds to the target position (order) in the data string in which the actual values included in the second sample group are arranged in ascending order as the power amount corresponding to the target position. Calculate the amount of power to be used.

For example, when the second sample group includes N actual values, the power amount calculation unit 38 generates a data string in which the N actual values are arranged, as in Equation (6).

Z[1]≤Z[2]≤Z[3]≤...≤Z[N-2]≤Z[N-1]≤Z[N] (6)

Here, Z[·] represents the actual value included in the second sample group. The brackets of Z[·] indicate the order.

For example, the power amount calculation unit 38 compares the data string arranged in this way with the target position (ranking), and calculates the power amount corresponding to the target position (ranking).

As an example, the power amount calculation unit 38 detects the actual value of the order closest to the target position in the data string. Then, the power amount calculation unit 38 may use the detected actual value as the power amount corresponding to the target position (order).

As another example, the power amount calculation unit 38 calculates the actual value of the highest ranking below the target position (ranking) and the lowest ranking above the target position (ranking) among the plurality of values included in the data string. Acquire achievement values. Then, the power amount calculation unit 38 may use the power amount obtained by performing linear interpolation based on the two obtained actual values as the power amount corresponding to the target position.

Specifically, let nr be the target position (rank), let Z[n] be the actual value of the highest rank below the target position (rank) included in the second sample group, and let Z[n] be the target position included in the second sample group. Let Z[n+1] be the actual value of the lowest ranking equal to or higher than (ranking). In this case, the magnitude relation among nr, n and (n+1) is given by equation (7).
n≦nr≦(n+1) (7)

Then, the power amount calculation unit 38 calculates X[nr], which is the power amount corresponding to the target position (order), by calculating Expression (8).
X[nr]=Z[n]×{(n+1)−nr}+Z[n+1]×{nr−n} (8)

Also in S19, the rank may be a percentile rank normalized between 0% and 100%.

Subsequently, in S20, the output unit 40 outputs the electric power amount corresponding to the target position (order) calculated in S19 as the predicted electric power amount that is predicted to be generated by the power plant during the prediction target time period on the prediction date. do.

The prediction device 10 repeats the above processing from S11 to S20 each time a prediction target time period is input.

In S18, the position calculation unit 36 may set a preset position (order) as the target position when the first time period on the target date is earlier than the preset time from the current time.

For example, if the power generation period of the power plant in one day is from 9:00 to 16:59 and the current time is 8:00, the first time slot on the target day is from 16:30 to 17:00 on the previous day. It will be 59 minutes. In such a case, the first time period on the target date is earlier than the preset time from the current time. Also, if the receiving unit 22 cannot receive the actual value due to a communication failure for a preset time or longer, the first time period on the target day is set ahead of the current time by a preset time or longer.

In such a case, the correlation between the actual value for the first time slot on the target day and the expected power consumption for the prediction target time slot decreases, and the prediction device 10 can accurately calculate the expected power consumption. may not be possible. Therefore, when the first time period on the target date is earlier than the preset time from the current time, the prediction device 10 predicts the power amount by setting the preset position (order) as the target position. Accuracy can be improved.

In addition, in S18, if the first time period on the target date is earlier than the preset time from the current time, the position calculation unit 36 determines that the actual values included in the first sample group have the highest frequency. The position (order) may be the target position. Even in this way, the prediction device 10 can improve the prediction accuracy of the power amount when there is a possibility that the predicted power amount cannot be calculated with high accuracy.

FIG. 8 is a diagram showing an example of changes in performance values for a plurality of reference dates and changes in performance values on a target date.

The amount of power generated by a photovoltaic power plant varies from day to day, but the trend of change during a day in close seasons, such as the last 30 days, is similar. Therefore, the ranking of the power amounts in the first time period arranged in ascending order on a plurality of reference days such as the most recent 30 days approximates the order of the power amounts in the prediction target time period arranged in ascending order.

Therefore, the prediction device 10 calculates the target position (order) of the actual values of the first time period on the target date with respect to the multiple actual values of the first time period on the multiple reference dates, Calculate the amount of electric power corresponding to the target position (rank) in a plurality of actual values for the time period. Then, the prediction device 10 outputs the power amount corresponding to the target position as the predicted power amount that is predicted to be generated in the prediction target time zone.

According to such a prediction device 10, the predicted amount of power is calculated from the actual value of the amount of power generated by the power plant, so it is possible to easily and accurately calculate the predicted amount of power using a small number of types of data. can.

In addition, the prediction device 10 corrects a plurality of actual values for the first time period on a plurality of reference dates and a plurality of actual values for the prediction target time period on a plurality of reference dates using the power plant model. As a result, the prediction device 10 can correct the actual value based on the weather information including the direction of the sun and the like, so that the prediction device 10 can output a more accurate predicted power amount.

(Second embodiment)
Next, the prediction device 10 according to the second embodiment will be described. In the description of the second embodiment, components having substantially the same configurations and functions as those of the components described in the previous embodiments are denoted by the same reference numerals, and detailed descriptions will be given except for the differences. omitted.

FIG. 9 is a diagram showing the functional configuration of the prediction device 10 according to the second embodiment. The prediction device 10 according to the second embodiment further includes a position corrector 52 as compared with the first embodiment.

The position correction unit 52 corrects the target position (order) calculated by the position calculation unit 36 based on additional information such as weather information, and supplies the corrected target position to the power amount calculation unit 38 . For example, the position correction unit 52 corrects the target position (ranking) based on the predicted change rate of the position in the prediction target time zone with respect to the position in the first time zone on the target day.

For example, the position correction unit 52 may receive weather information for the first time slot on the target date and forecast weather information for the forecast target time slot on the forecast date, and calculate the forecast change rate. For example, the position correction unit 52 lowers the target position (rank) when the weather is fine in the first time slot on the target day, but the weather suddenly changes and it becomes rainy in the prediction target time slot. corrected to In addition, for example, the position correction unit 52 increases the target position (rank) when the weather is rainy in the first time period on the target day, but the weather suddenly changes and becomes sunny in the prediction target time period. Correct to

Since the prediction device 10 according to the second embodiment corrects the target position (order) based on the difference between the weather information of the first time period and the weather information of the prediction target time period, etc., the accuracy is improved. Predicted power consumption can be output.

(Power plant model)
Next, the power plant model of the photovoltaic power plant will be explained.

FIG. 10 is a diagram showing a block configuration of a power plant model of a photovoltaic power plant. The power plant model receives layout angles, module capacities and inverter capacities as facility information. Furthermore, the power plant model receives, as weather information, the global solar radiation intensity, the top horizontal solar radiation intensity, the direction of the sun, and the ground surface albedo at the target time. Then, the power plant model calculates the power (kW) generated by the solar power plant at the target time.

Note that the target time is the time at which power is estimated. In the present embodiment, the target time is any time within the first time period or any time within the prediction target time period. For example, the target time may be the center time of the first time period or the prediction target time period.

The power plant model includes a solar conversion unit 112 , a temperature correction unit 114 and a power conversion unit 116 .

The solar radiation converter 112 receives the layout angle, the global solar radiation intensity, the upper atmospheric horizontal solar radiation intensity, the solar direction, and the ground surface albedo. Based on these pieces of information, the solar radiation conversion unit 112 performs direct scattering separation processing and inclined surface conversion processing to calculate the inclined surface solar radiation intensity.

The temperature correction unit 114 receives the air temperature, the wind speed, and the inclined surface solar radiation intensity calculated by the solar radiation conversion unit 112 . The temperature correction unit 114 calculates a temperature correction coefficient based on these pieces of information.

The power conversion unit 116 receives the module capacity, the inverter capacity, the inclined surface solar radiation intensity calculated by the solar conversion unit 112 , and the temperature correction coefficient calculated by the temperature correction unit 114 . The power converter 116 calculates the power (kW) generated by the solar power plant based on these pieces of information.

(Direct scattering separation treatment)
The power plant model deals with the sloped surface solar radiation intensity, which is the solar radiation intensity with respect to the sloped surface of the solar cell module. The solar radiation intensity handled in weather prediction and weather observation is the all-sky horizontal solar radiation intensity, which is the solar radiation intensity with respect to the horizontal plane. For this reason, the power plant model converts the global solar radiation intensity into the inclined surface solar radiation intensity.

In order to convert the all-sky solar radiation intensity into the inclined-plane solar radiation intensity, first, the solar radiation conversion unit 112 separates the all-sky horizontal solar radiation intensity into a direct light component and a scattered light component using a direct scattering model. Executes the direct scattering separation process. When the global plane solar radiation intensity is G, the direct light component is G _b , and the scattered light component is G _d , the relationship between G, G _b and G _d is expressed as in Equation (11).

The direct dispersion model calculates the ratio of the scattered light component in the global solar radiation intensity. The direct scattering separation model is an empirical formula that expresses the qualitative tendency that the scattered light component is small in fine weather, that is, the direct light component is large, and the scattered light component is large in cloudy weather, that is, the direct light component is small. is.

The global solar radiation intensity depends on the solar altitude as well as the cloud cover. In order to eliminate the influence of the sun's altitude, the direct-scattering separation model uses a clearness index obtained by normalizing the global solar radiation intensity by the horizontal surface solar radiation intensity at the top of the atmosphere. When the clearness index is represented by k, and the horizontal surface solar radiation intensity at the upper end of the atmosphere is represented by _G0 , the clearness index is expressed as in Equation (12). k is a real number of 0 or more and 1 or less.

A direct dispersion model is represented by Formula (13).

f _d (k) is a function for calculating the ratio of the scattered light component in the all-sky horizontal solar radiation intensity based on the clearness index. When k is small, k _d is large, and when k is large, k _d is small. It is a function that expresses the tendency of

Also, the direct light component is represented by Equation (14).

From the above, in the direct scattering separation process, the solar radiation conversion unit 112 performs the calculations of formulas (15-1), (15-2), and (15-3) to convert the direct light component and the scattered light component into calculate.

FIG. 11 is a diagram showing numerical examples of Erbs, Orgill, and Reindl, which are linear separation models. Any of the Erbs, Orgill and Reindl direct segregation models will give approximate numerical values. Therefore, the solar radiation conversion unit 112 can output an approximated calculation result using any direct-scattering separation model.

FIG. 12 is a diagram showing an example of computation of a direct-discrete model using Orgill. For example, when Orgill's direct-scattering separation model is used, the solar radiation conversion unit 112 executes calculations as shown in FIG. 12 .

(Tilted surface conversion processing)
The degree of dependence of the direct light component and the scattered light component on the slanted surface solar radiation intensity changes depending on the relationship between the layout angle of the solar cell module and the direction of the sun. Equation (16) is a relational expression representing the inclined surface solar radiation intensity, taking into consideration the angular relationship between the layout angle of the solar cell module and the direction of the sun.

S represents the slanted surface solar radiation intensity. n̂ is a unit vector representing a solid angle (layout angle) in the normal direction of the solar cell module. s^ is a unit vector representing the solid angle in the direction of the sun. z^ is a unit vector representing a solid angle in the vertical direction. Note that n̂, ŝ, and ẑ are provided with a hat symbol immediately above the alphabet in the mathematical expressions. Also, ρ represents the ground surface albedo, which is the reflectance of the ground surface.

The first term on the right side of Equation (16) represents the direct light component with respect to the inclined surface of the solar cell module. The second term on the right side of Equation (16) represents the scattered light component for the inclined surface of the solar cell module. The third term on the right side of Equation (16) represents the reflected light component with respect to the inclined surface of the solar cell module.

The direct light component to the inclined surface is a directional solar radiation component, and becomes maximum when the normal direction (n̂) of the solar cell module coincides with the sun direction (ŝ). That is, the value (G _b /(s^, z^)) obtained by normalizing the direct light component (G _b ) with respect to the horizontal plane by the solid angle (s^, z^) in the direction of the sun with respect to the vertical direction is the direct light component. maximum value. Therefore, the first term on the right side of equation (16) is a value obtained by multiplying the maximum value of the direct light component by the solid angle (ŝ, n̂) in the direction of the sun with respect to the normal direction of the solar cell module.

The scattered light component on the inclined surface is a solar radiation component without directivity, and is a solid angle average of the components (r^, n^) incident on the solar cell module with the same intensity from each direction (r^) of the celestial sphere. proportional to the angle. Therefore, the scattered light component for the inclined plane is proportional to the value expressed in equation (17).

In equation (17), <·> _r̂ represents a solid angle average value for each direction (r̂) of the celestial sphere. Therefore, the second term on the right side of equation (16) is the value obtained by multiplying the right side of equation (17) by _Gd .

The reflected light component on the inclined surface is a solar radiation component with no directivity, and is similar to the scattered light component if the reflected light from the ground surface is interpreted as the scattered light with the intensity of the scattered light of _ρGd . Therefore, the third term on the right side of equation (16) is the value obtained by multiplying the right side of equation (17) by _ρGd .

From the above, the solar radiation converter 112 calculates n^, s^, and z^ based on the latitude and longitude, time, and layout angle of the solar power plant. Then, the solar radiation conversion unit 112 calculates the inclined surface solar radiation intensity based on the relational expression (16).

FIG. 13 is a diagram showing numerical examples of the sun direction. FIG. 14 is a diagram showing numerical examples of the inclined surface solar radiation intensity when the direction of the sun changes as shown in FIG.

In the examples of FIGS. 13 and 14, k is 0.8 and the latitude and longitude of the power station are 35.00N and 135.00E. In the examples of FIGS. 13 and 14, the azimuth from south to west is represented by 0° to 90°, and the altitude from the horizontal to the zenith is represented by 0° to 90°. In FIG. 14, the tilt angle of the solar cell module is changed between 0°, 30° and 60°, and the azimuth of the solar cell module is 30°. Therefore, in the example of FIG. 14, when the inclination angle is large, the direct light component does not reach early in the morning, and the inclined surface solar radiation intensity becomes small.

FIG. 15 is a diagram showing an example of calculation of the direction of the sun. Sun direction can be calculated by a function that depends on latitude, longitude and time. The power plant model may calculate the direction of the sun using a function as shown in FIG. 15 based on the latitude, longitude and target time of the solar power plant.

(Calculation of temperature correction coefficient)
The efficiency of a photovoltaic power plant mainly depends on the conversion efficiency of the photovoltaic modules. In general, the lower the module temperature, the higher the conversion efficiency of a solar cell module. Module temperature is affected by ambient air temperature, wind speed, and sloping surface insolation intensity. The module temperature becomes close to the surrounding air temperature when the slanting surface solar radiation intensity is weak, but becomes higher than the surrounding air temperature when the slanting surface solar radiation intensity is strong. Also, the temperature rise of the module due to solar radiation can be suppressed when the wind speed is high.

If S is the solar radiation intensity on the inclined surface, a is the absorption rate of the solar radiation intensity on the inclined surface, h is the heat transfer efficiency, T is the ambient air temperature, and _Tm is the module temperature, the heat flux q that escapes from the solar cell module is It is expressed as in formula (18-1). Therefore, _Tm , which is the module temperature, is expressed as in Equation (18-2).

The heat transfer efficiency, h, generally increases with wind speed, but cannot be easily calculated because it depends on flow conditions and installation conditions. Therefore, the temperature correction unit 114 may calculate the module temperature based on the empirical formula shown in formula (19) including the sloping surface solar radiation intensity, the air temperature, and the wind speed. In addition, in Formula (19), V is a wind speed. Also, in equation (19), A and B are constants. For example, in equation (19), A is approximately 50 and B is approximately 0.4.

FIG. 16 is a diagram showing the difference between the module temperature calculated using equation (19) and the ambient air temperature. The temperature correction unit 114 calculates the module temperature from the inclined surface solar radiation intensity and the wind speed based on the numerical values shown in FIG. 16, for example.

Also, the output current-output voltage characteristics of a solar cell module are mainly characterized by short-circuit current and open-circuit voltage. The short-circuit current increases according to the slanted surface solar radiation intensity. However, the open circuit voltage decreases as the module temperature increases. The conversion efficiency of a solar cell module is a value obtained by dividing the generated electric power by the slanted surface solar radiation intensity, and decreases as the temperature rises. The temperature correction coefficient of the solar cell module is the conversion efficiency normalized at a module temperature of 25.0°C.

The relationship between the temperature correction coefficient and the module temperature can be approximated by an empirical formula using the temperature coefficient. When the temperature correction coefficient is _KPV , the temperature correction coefficient is represented by Equation (20). Note that α is a value of about -0.4/100.

From the above, after calculating the module temperature, the temperature correction unit 114 calculates the temperature correction coefficient based on Equation (20). For example, when the module temperature is 55° C., the temperature correction coefficient is a numerical value that reduces the power generation efficiency by about 12%.

(Power calculation process)
The main parameters of a photovoltaic power plant are module capacity and inverter capacity. A ratio of the module capacity to the inverter capacity is called an overload rate, and is set to a value of about 1.0 to 2.0 in order to increase the operating rate. In this case, the DC output power of the solar cell module is expressed by equation (21-1), and the AC output power of the inverter is expressed by equation (21-2).

Y _DC is the DC output power of the solar module. Y _AC is the AC output power of the inverter. C _PV is the module capacitance. C _PCS is the inverter capacity. K _PV is the inverter conversion efficiency. η is the overload rate. min(x,y) is a function that selects the smaller of x and y.

It should be noted that the module capacity is usually defined as the output power under standard conditions of module temperature of 25° C. and slant surface irradiance of 1.0 kW/m ₂ . Therefore, the DC output power of the solar cell module can be expressed by a simple formula proportional to the slanted surface solar radiation intensity.

The final output power of the photovoltaic power plant is the AC output power of the inverter. The AC output power of the inverter is a value obtained by limiting the DC output power of the solar cell module by the inverter capacity.

FIG. 17 is a diagram showing the conversion efficiency of the inverter with respect to the load factor. The power plant model may consider the conversion efficiency of the inverter to be approximately a constant 1.0. However, the power plant model may use a function containing empirical parameters to calculate the inverter conversion efficiency. For example, the power plant model may calculate the inverter conversion efficiency using the functions of equations (22-1) and (22-2).

x represents the load factor. In equation (22-2), A, B and C represent empirical parameters. A is about 0.05. B is about 1.0. C is about 0.02. The function of formula (22-2) has a maximum value of 0/97 (K _PCS (x)=1/(B+2√AC) when the load factor is 0.63 (x=√(C/A)). ). The power plant model can improve the accuracy of estimating power at low load by calculating the inverter conversion efficiency using the function of equation (22-2).

(Processing flow of power plant model)
FIG. 18 is a flow chart showing the process flow of the power plant model. To summarize the above description, the power plant model executes processing according to the flow shown in FIG.

First, in S51, the power plant model acquires equipment information shown in the following equation (23).

The power plant model may obtain standardized values that are predetermined for the layout angle, module capacity, and inverter capacity.

Subsequently, in S52, the power plant model acquires the weather information shown in Equation (24) below.

Note that the power plant model may acquire a preset value as the ground surface albedo when the ground surface albedo cannot be acquired. In addition, when the power plant model cannot obtain the global solar radiation intensity or the upper horizontal solar radiation intensity, the average value of the area where the power plant is installed is used as the global solar radiation intensity or the upper horizontal solar radiation intensity. A preset value may be acquired. The power plant model may also calculate the power generated by the power plant using past average values when temperature or wind speed cannot be obtained.

Subsequently, in S53, the power plant model calculates the following equations (25-1), (25-2), (25-3) and (25-4) to perform direct scattering separation processing. Run.

Subsequently, in S54, the power plant model calculates the following equations (26-1), (26-2), (26-3) and (26-4) to perform the slope conversion process. Run. Thereby, the power plant model can calculate the inclined surface solar radiation intensity (S).

Subsequently, in S55, the power plant model calculates the temperature correction coefficient by calculating the following equations (27-1) and (27-2).

Subsequently, in S56, the power plant model computes the following equations (28-1) and (28-2) to calculate the power output from the solar power plant.

Then, in S57, the power plant model outputs the power shown in Equation (29) as the target day calculated value or the reference day calculated value.

Note that the power plant model may execute the processes from S51 to S54 and output the inclined plane solar radiation intensity (S) as the calculated value for the target day or the calculated value for the reference day. In this case, the power plant model does not need to acquire temperature and wind speed in S52.

As described above, the power plant model can calculate calculated values for correcting a plurality of actual values included in the first sample group and the second sample group.

(Hardware configuration)
FIG. 19 is a diagram illustrating an example of the hardware configuration of the prediction device 10 according to the embodiment. The prediction device 10 according to this embodiment is realized by, for example, an information processing device having a hardware configuration as shown in FIG. 19 . The prediction device 10 includes a CPU (Central Processing Unit) 201 , a RAM (Random Access Memory) 202 , a ROM (Read Only Memory) 203 , an operation input device 204 , a storage device 206 and a communication device 207 . These units are connected by a bus.

The CPU 201 is a processor that executes arithmetic processing, control processing, and the like according to programs. The CPU 201 uses a predetermined area of the RAM 202 as a work area and executes various processes in cooperation with programs stored in the ROM 203 and the storage device 206 or the like.

A RAM 202 is a memory such as SDRAM (Synchronous Dynamic Random Access Memory). A RAM 202 functions as a work area for the CPU 201 . The ROM 203 is a memory that non-rewritably stores programs and various information.

The operation input device 204 is an input device such as a mouse and keyboard. The operation input device 204 receives information input by a user as an instruction signal, and outputs the instruction signal to the CPU 201 .

The storage device 206 is a device that writes data to and reads data from a semiconductor storage medium such as a flash memory, or a magnetically or optically recordable storage medium. The storage device 206 writes data to and reads data from a storage medium under the control of the CPU 201 . A communication device 207 communicates with an external device via a network under the control of the CPU 201 .

The program executed by the prediction device 10 of the present embodiment includes a reception module, an input module, a target sample acquisition module, a first sample group acquisition module, a second sample group acquisition module, a performance value correction module, It comprises a position calculation module, a power calculation module and an output module. Additionally, the program may include a position correction module. This program is expanded on the RAM 202 and executed by the CPU 201 (processor), thereby operating the information processing apparatus as a receiving unit 22, an input unit 26, a target sample acquiring unit 28, a first sample group acquiring unit 30, a second sample, and so on. It functions as a group acquisition unit 32 , a performance value correction unit 34 , a position calculation unit 36 , a power amount calculation unit 38 , and an output unit 40 . Furthermore, this program may cause the information processing device to function as the position corrector 52 . The prediction device 10 includes a receiving unit 22, an input unit 26, a target sample acquiring unit 28, a first sample group acquiring unit 30, a second sample group acquiring unit 32, a performance value correcting unit 34, a position calculating unit 36, an electric energy At least part of the calculation unit 38, the output unit 40, and the position correction unit 52 may be realized by a hardware circuit (for example, a semiconductor integrated circuit).

In addition, the program executed by the prediction device 10 of the present embodiment is a file in a format that can be installed in a computer or in a format that can be executed, and is stored on a CD-ROM, a flexible disk, a CD-R, a DVD (Digital Versatile Disk), etc. It is recorded on a computer-readable recording medium and provided.

Alternatively, the program executed by the prediction device 10 of this embodiment may be stored in a computer connected to a network such as the Internet, and may be provided by being downloaded via the network. Also, the program executed by the prediction device 10 of this embodiment may be configured to be provided or distributed via a network such as the Internet. Alternatively, the program to be executed by the prediction device 10 may be configured so as to be pre-installed in the ROM 203 or the like and provided.

While several embodiments of the invention have been described, these embodiments have been presented by way of example and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, replacements, and modifications can be made without departing from the scope of the invention. These embodiments and their modifications are included in the scope and gist of the invention, and are included in the scope of the invention described in the claims and equivalents thereof.

10 prediction device 22 reception unit 24 storage unit 26 input unit 28 target sample acquisition unit 30 first sample group acquisition unit 32 second sample group acquisition unit 34 performance value correction unit 36 position calculation unit 38 power amount calculation unit 40 output unit 52 position corrector

Sokol Dervishi and Ardeshir Mahdavi, "COMPARISON OF MODELS FOR THE DERIVATION OF DIFFUSE FRACTION OF GLOBAL IRRADIANCE DATA FOR VIENNA, AUSTRIA", Proceedings of Building Simulation 2011: 12th Conference of International Building Performance Simulation Association, Sydney, 14-16 November，P. 765-771

Claims (20)

subjecting the amount of electricity generated by the one or more power plants to a first sample set comprising a plurality of actual values for a first time period on multiple reference days of the amount of electricity generated by the one or more power plants. a position calculation unit that calculates a target position for the actual value of the first time period in a day;
Electric energy for calculating the electric energy corresponding to the target position in a second sample group including a plurality of actual values of the electric energy generated by the one or more power plants in the prediction target time period on the plurality of reference days. a calculation unit;
an output unit that outputs the amount of power corresponding to the target position as a predicted amount of power that is predicted to be generated by the one or more power plants in the prediction target time period;
Information processing device. The target position is the actual value of the first time period on the target date in a data string in which the multiple actual values included in the first sample group are arranged in the first order of either ascending order or descending order. is the order of the corresponding electric energy,
The power amount corresponding to the target position is the power amount corresponding to the target position in a data string in which a plurality of actual values included in the second sample group are arranged in the first order. Information processing equipment. Each of the plurality of actual values included in the first sample group and each of the plurality of actual values included in the second sample group are calculated based on a predetermined power plant model, and the Further includes an actual value correction unit that corrects according to the ratio with the reference date calculation value calculated based on the power plant model,
The power plant model is a model that calculates the power generated by the one or more power plants or a value proportional to the power generated by the one or more power plants,
The target day calculated value is the target day power generated by the one or more power plants in the corresponding time period on the target day or a value proportional to the target day power calculated based on the power plant model. can be,
The reference day calculated value is the reference day power generated by the one or more power plants in the corresponding time period on the corresponding reference date of the plurality of reference days, calculated based on the power plant model, or The information processing apparatus according to claim 1 or 2, wherein the value is proportional to the reference daily power. The actual value correction unit,
Each of the plurality of actual values included in the first sample group is calculated as the target date calculated value for the first time period on the target date and the reference date calculated value for the first time period on the corresponding reference date. corrected by multiplying by the ratio of
Each of the plurality of actual values included in the second sample group is calculated as the target date calculated value for the first time period on the target date and the reference date calculated value for the first time period on the corresponding reference date. The information processing apparatus according to claim 3, wherein the correction is performed by multiplying a ratio of . Each of the one or more power plants is a solar power plant capable of estimating the power generated using the same power plant model,
The power plant model is the solar radiation intensity with respect to the inclined surface of the solar cell module based on the layout angle indicating the normal direction of the solar cell module, the solar radiation intensity on the all-sky horizontal plane, the solar radiation intensity on the top horizontal plane of the atmosphere, the solar direction, and the ground surface albedo. The information processing apparatus according to claim 4, wherein the inclined plane solar radiation intensity is calculated. The information processing apparatus according to claim 5, wherein, when the ground surface albedo cannot be acquired, the power plant model calculates the inclined plane solar radiation intensity using a value preset as the ground surface albedo. The power plant model, when the global solar radiation intensity or the upper atmosphere horizontal surface solar radiation intensity cannot be obtained, uses a preset value as the global solar radiation intensity or the upper atmosphere horizontal surface solar radiation intensity to calculate the inclined surface solar radiation intensity. The information processing apparatus according to claim 5 or 6, wherein the intensity is calculated. Each of the one or more power plants is a solar power plant capable of estimating the power generated using the same power plant model,
The power plant model is based on the layout angle indicating the normal direction of the solar cell module, the all-sky solar radiation intensity, the top horizontal solar radiation intensity, the solar direction, the ground surface albedo, the temperature and the wind speed, and the one or more power plants The information processing apparatus according to claim 4, wherein the power to be generated is calculated. 9. The information processing according to claim 8, wherein when the ground surface albedo cannot be obtained, the power plant model calculates the electric power generated by the one or more power plants using a preset value as the ground surface albedo. Device. When the global solar radiation intensity or the upper atmospheric horizontal solar radiation intensity cannot be obtained, the power plant model uses a preset value as the global horizontal solar radiation intensity or the upper atmospheric horizontal solar radiation intensity to generate the one or more The information processing apparatus according to claim 8 or 9, which calculates the electric power generated by the power plant. 11. The power plant model according to any one of claims 8 to 10, wherein, when the air temperature or the wind speed cannot be obtained, the power generated by the one or more power plants is calculated using past average values. information processing equipment. The first time period is the most recent time period in which power was generated by the one or more power plants;
The information processing apparatus according to any one of claims 2 to 11, wherein the prediction target time period is a time period later than the first time period on the target date. 13. The information processing according to claim 12, wherein, when the first time period on the target date is earlier than a preset time from the current time, the position calculation unit sets a preset value as the target position. Device. When the first time period on the target date is earlier than the preset time from the current time, the position calculation unit calculates the position with the highest frequency among the plurality of actual values included in the first sample group. , is the target position. The position calculation unit calculates a predetermined number of target positions, each of a predetermined number of two or more time periods before the latest time period in which power was generated by the one or more power plants as the first time period. 12. The information processing apparatus according to any one of claims 2 to 11, wherein the average value of the calculated predetermined number of target positions is output as the target position of the actual value of the first time period on the target date. . 16. The method according to any one of claims 2 to 15, further comprising a position correction unit that corrects the target position based on a predicted change rate of the position in the prediction target time zone with respect to the position in the first time zone on the target date. The information processing device described. 17. The information according to claim 16, wherein the position correction unit calculates the predicted rate of change based on weather information for the first time slot on the target date and predicted weather information for the prediction target time slot on the prediction date. processing equipment. An information processing method executed by an information processing device,
The information processing device performs power generation by the one or more power plants for a first sample group including a plurality of actual values of the amount of electric power generated by the one or more power plants in a first time period on a plurality of reference days. Calculate the target position of the actual value of the first time period on the target day of the electric energy,
The power corresponding to the target position in a second sample group including a plurality of actual values of the amount of power generated by the one or more power plants in the prediction target time period on the plurality of reference dates. calculate the amount of
The information processing method, wherein the information processing device outputs the electric energy corresponding to the target position as a predicted electric energy expected to be generated by the one or more power plants in the prediction target time period. A program for causing an information processing device to function as a prediction device that predicts power consumption,
the information processing device,
subjecting the amount of electricity generated by the one or more power plants to a first sample set comprising a plurality of actual values for a first time period on multiple reference days of the amount of electricity generated by the one or more power plants. a position calculation unit that calculates a target position for the actual value of the first time period in a day;
Electric energy for calculating the electric energy corresponding to the target position in a second sample group including a plurality of actual values of the electric energy generated by the one or more power plants in the prediction target time period on the plurality of reference days. a calculation unit;
an output unit that outputs the amount of power corresponding to the target position as a predicted amount of power that is predicted to be generated by the one or more power plants in the prediction target time period;
A program that works as one or more power plants;
an information processing device that predicts the amount of power generated by the one or more power plants;
with
The information processing device is
subjecting the amount of electricity generated by the one or more power plants to a first sample set comprising a plurality of actual values for a first time period on multiple reference days of the amount of electricity generated by the one or more power plants. a position calculation unit that calculates a target position for the actual value of the first time period in a day;
Electric energy for calculating the electric energy corresponding to the target position in a second sample group including a plurality of actual values of the electric energy generated by the one or more power plants in the prediction target time period on the plurality of reference days. a calculation unit;
an output unit that outputs the amount of power corresponding to the target position as a predicted amount of power that is predicted to be generated by the one or more power plants in the prediction target time period;
A power generation system with

Images