JP2017011780A - POWER GENERATION PREDICTION CONTROL DEVICE, POWER GENERATION PREDICTION METHOD, AND POWER GENERATION PREDICTION PROGRAM - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a power generation amount prediction control device capable of highly accurately predicting a power generation amount of a photovoltaic power plant.SOLUTION: The power generation amount prediction control device comprises: a storage part 2 in which an attenuation rate of a power generation amount with respect to a cloud amount or the like is stored; a power generation amount correction part 13 by which, a periodically stored past power generation amount X is corrected into a power generation amount Xin the case of a cloud amount 0% based on the attenuation rate of the power generation amount corresponding to the cloud amount at the power generation time thereof, and stored in the storage part 2 as a new corrected power generation amount Xat every power generation time; a first power generation amount prediction part 14 by which the stored corrected power generation amount Xis sorted for the unit of a power generation time, an average and a standard deviation at every power generation time are calculated and based on the obtained average and standard deviation, a power generation amount Xin the case of cloud amount 0% on a prediction target date is predicted; and a second power generation amount prediction part 15 by which, based on the power generation amount Xand the attenuation rate of the power generation amount corresponding to a cloud amount on the prediction target date, a power generation amount Xon the prediction target date is predicted.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a power generation amount prediction control device, a power generation amount prediction method, and a power generation amount prediction program for predicting a power generation amount of a solar power plant.

Photovoltaic power generation is an infinite energy that does not depend on fossil fuels such as oil, and has attracted attention as clean energy that does not emit CO ₂ during power generation unlike fossil fuels such as oil. On the other hand, solar power generation is a very unstable power generation facility whose power generation amount varies depending on weather conditions. Therefore, if the power generation amount can be predicted, the scope of application will be further expanded.

  As a conventional method for predicting the power generation amount of a solar power plant, for example, there is a technique described in Patent Document 1 below. In the following Patent Document 1, in a system for predicting the power generation amount, the power generation amount is based on the weather phenomenon observed in the past in the installation area of the solar power plant and the power generation amount of the solar power plant measured in the past. Deriving a prediction formula is described.

  Patent Document 1 listed below includes a weather forecast for a prediction target date or a prediction target time zone in a solar power plant installation area, and a power generation amount of the solar power plant measured before the prediction execution time of the prediction target date. Is input to the power generation amount prediction formula, and the power generation amount of the solar power plant is predicted. In particular, regarding the power generation amount measured before the prediction execution time, the following Patent Document 1 describes the actual power generation amount in the immediately preceding time zone (actual power generation amount in the 9 o'clock time zone when the prediction calculation is performed at 10:00). Is used to predict power generation on the target day.

Japanese Patent No. 3984604

  However, although the technique described in the above cited document uses a weather phenomenon observed in the past in the installation area of the solar power plant in a system for predicting the amount of power generation, the weather phenomenon is sunny: 1, cloudy: 2, the rain: 3 and the cloud: 4. The degree of correlation between the amount of cloud and the amount of power generation is not yet grasped, and there is room for improvement in the prediction accuracy of the amount of power generation.

  The technique described in the above cited document predicts the current power generation amount using the power generation amount immediately before the prediction target time (for example, one hour ago). It is not a specification that predicts the power generation amount by inputting the amount into the power generation amount prediction formula. In other words, it is a system that predicts an accurate power generation amount by inputting the power generation amount immediately before the prediction target time as an important parameter in the power generation amount prediction formula. In such a case, the power generation amount on the target day is predicted and the retailer (the person engaged in the business of selling or buying power, such as an electric power company) is notified of the predicted power generation amount.

  The present invention has been made in view of the above, and is a power generation amount prediction control apparatus capable of predicting the power generation amount on the prediction target day with high accuracy based on the relationship between the cloud amount and the power generation amount of solar power generation. An object is to provide a power generation amount prediction method and a power generation amount prediction program.

  In order to solve the above-described problems and achieve the object, the power generation amount prediction control device according to the present invention is a power generation amount prediction control device that predicts the power generation amount of a solar power plant, and is a correlation between the cloud amount and the power generation amount. Based on the storage means storing the information indicating the relationship and the past power generation amount periodically accumulated based on the cloud amount included in the weather forecast data at the power generation time and the information indicating the correlation, the cloud amount of 0% Power generation amount correction means for correcting the generated power generation amount to a new power generation amount at each power generation time and generating the power by classifying the stored corrected power generation amount into power generation time units. A first power generation amount predicting unit that calculates an average and standard deviation for each time and predicts a first predicted power generation amount that is a power generation amount when the cloud amount is 0% on the prediction target day based on the average and standard deviation. And the first predicted power generation amount Second power generation amount prediction means for predicting a second predicted power generation amount that is the power generation amount of the prediction target day based on the cloud amount included in the weather forecast data of the prediction target day and information indicating the correlation; It is characterized by providing.

  A power generation amount prediction method according to the next invention is a power generation amount prediction method for predicting a power generation amount of a solar power plant, and stores information indicating a correlation between a cloud amount and a power generation amount in a storage unit. Step, and periodically read the past power generation amount stored in the storage unit at each power generation time, and the past power generation amount is converted into information indicating the cloud amount and the correlation included in the weather forecast data at the power generation time. Based on this, the power generation amount when the cloud amount is 0% is corrected, and the corrected power generation amount is stored in the storage unit as a new corrected power generation amount at each power generation time, and is stored in the storage unit. The corrected power generation amount is read out, the corrected power generation amount is classified into power generation time units, the average and standard deviation for each power generation time are calculated, and based on the average and standard deviation, the cloud amount on the prediction target day is 0% In the case of A first power generation amount prediction step of predicting the amount of electricity, storing the predicted power generation amount as a first predicted power generation amount in the storage unit, and reading out the first predicted power generation amount stored in the storage unit; Based on the predicted power generation amount of 1, the cloud amount included in the weather forecast data of the prediction target day, and the information indicating the correlation, the power generation amount of the prediction target day is predicted, and the predicted power generation amount is the second predicted power generation And a second power generation amount prediction step stored in the storage unit as a quantity.

  A power generation amount prediction program according to the next invention is a power generation amount prediction program that is executed by a computer that operates as a power generation amount prediction control device that predicts a power generation amount of a solar power plant, and a correlation between a cloud amount and a power generation amount The storage procedure for storing the information indicating in the storage unit, the past power generation amount periodically stored in the storage unit is read at each power generation time, and the past power generation amount is included in the weather forecast data at the power generation time Based on the cloud amount and the information indicating the correlation, the power generation amount is corrected to the power generation amount when the cloud amount is 0%, and the corrected power generation amount is accumulated in the storage unit as a new corrected power generation amount at each power generation time. Correction procedure, read the corrected power generation amount stored in the storage unit, classify the corrected power generation amount into power generation time units, calculate the average and standard deviation for each power generation time, A first power generation amount prediction procedure for predicting a power generation amount when the cloud amount is 0% on the prediction target date based on the quasi-deviation, and storing the predicted power generation amount as a first predicted power generation amount in the storage unit, the storage The first predicted power generation amount stored in the unit, and based on the information indicating the first predicted power generation amount, the cloud amount included in the weather forecast data of the prediction target day, and the correlation, A power generation amount is predicted, and the computer is caused to execute a second power generation amount prediction procedure for storing the predicted power generation amount as a second predicted power generation amount in the storage unit.

  According to the present invention, there is an effect that the power generation amount of the solar power plant on the prediction target date can be predicted with high accuracy.

FIG. 1 is a diagram illustrating a hardware configuration example of a computer that operates as a power generation amount prediction control apparatus. FIG. 2 is a diagram illustrating an example of a functional block configuration of the power generation amount prediction control apparatus. FIG. 3 is a flowchart showing a process of correcting the actual power generation amount to the power generation amount when the cloud amount is 0%. FIG. 4 is a diagram illustrating an image in a case where “a process of correcting the actual power generation amount to the power generation amount when the cloud amount is 0%” is executed. FIG. 5 is a flowchart illustrating a process of predicting the power generation amount when the cloud amount is 0% on the prediction target day. FIG. 6A is a diagram illustrating an image when the “process for predicting the power generation amount when the cloud amount is 0% on the prediction target day” is executed. FIG. 6B is a diagram illustrating an image when the “process for predicting the power generation amount when the cloud amount is 0% on the prediction target day” is executed. FIG. 7 is a flowchart illustrating a process of predicting the power generation amount on the prediction target day. FIG. 8 is a diagram illustrating an image when the “process for predicting the power generation amount on the prediction target day” is executed.

  Embodiments of a power generation amount prediction control device, a power generation amount prediction method, and a power generation amount prediction program according to the present invention will be described below in detail with reference to the drawings. Note that the present invention is not limited to the embodiments.

  FIG. 1 is a diagram illustrating a hardware configuration example of a computer that operates as a power generation amount prediction control apparatus according to the present embodiment. In FIG. 1, a computer according to the present embodiment includes a control unit 1 including a CPU (Central Processing Unit) and an FPGA (Field Programmable Gate Array), a ROM (Read Only Memory), a RAM (Random Access Memory), and the like. A storage unit 2 including various memories, an input unit 3 including a user interface such as a keyboard 8 and a mouse 9, an output unit 4 that performs output processing such as printing, a display unit 5 that is a display, and a predetermined network And a communication unit 6 that communicates with the outside. In FIG. 1, the input unit 3 including the user interface such as the keyboard 8 and the mouse 9 is provided. However, the computer of the present embodiment is not limited to this, and the display unit 5 is provided with a touch panel function. The input unit 3 may not be provided, or the input unit 3 may be used in combination.

  In FIG. 1, the control unit 1 executes a power generation amount prediction program of the present embodiment. The storage unit 2 includes an internal memory such as a ROM or a RAM, and stores a power generation amount prediction program and various input information of this embodiment, data obtained in the course of processing, and the like. The control unit 1 executes the power generation amount prediction control of the present embodiment by reading the program stored in the storage unit 2. The storage unit 2 is not limited to the internal memory, and may be an external storage medium such as a DVD (Digital Versatile Disc) or an SD memory, or may be an internal memory or an external storage medium (DVD or SD). (Memory etc.).

  In addition, the power generation amount prediction control device of the present embodiment predicts the power generation amount described later in addition to a company that operates solar power generation such as a power plant, a power company, a retailer (power purchase, power sale, etc.). It can be easily installed in an environment where an environment for obtaining parameters (information) for performing is available.

  Here, as a premise for the power generation amount prediction control apparatus of the present embodiment to predict the power generation amount by the solar power plant, the correlation between the cloud amount and the power generation amount by the solar power generation will be described.

For example, the attenuation rate α (0 ≦ α ≦ 1) of the power generation amount with respect to the cloud amount can be generalized as in the following equation (1-1). As an example of the primary equation, the following (1-2) It can be expressed as: In the following formulas (1-1) and (1-2), X ₁ is the upper cloud cover (%), X ₂ is the intermediate cloud cover (%), and X ₃ is the lower cloud cover (%). X ₄ is the total cloud amount (%), and is a general parameter included in the weather forecast data obtained every certain time (such as one hour). Further, a, b, c, and d are coefficients representing individual attenuation rates with respect to the cloud amount of each layer. For example, when the cloud amount of each layer is maximum (100%), the attenuation rate α becomes 0, which reduces the cloud amount. Accordingly, the attenuation rate α changes to 0.1, 0.2, 0.3..., And is a coefficient defined so that the attenuation rate α becomes 1 when the cloud amount of each layer is minimum (0%).

Decay rate α = f (X ₁ , X ₂ , X ₃ , X ₄ ) (1-1)
= AX ₁ + bX ₂ + cX ₃ + dX ₄ (1-2)

Therefore, the power generation amount (for example, the predicted power generation amount) X is the power generation amount X ₀ when the cloud amount is 0% (all cloud amounts in the upper layer, middle layer, and lower layer are 0%) as shown in the following equation (2). The value obtained by multiplying the attenuation rate α (0 ≦ α ≦ 1). Further, from the relationship shown in the following formula (2), the power generation amount X ₀ when the cloud amount is 0% is obtained by dividing the power generation amount (for example, the actual power generation amount) X by the attenuation rate α as shown in the following formula (3). Value.

X = X ₀ × α (2)

X ₀ = X ÷ α (3)

In the present embodiment, as a precondition process for obtaining X and X ₀ by the above formula, the control unit 1 calculates coefficients a, b, c, and d in advance by multivariate analysis based on past power generation result data. I will keep it.

  FIG. 2 is a diagram illustrating an example of a functional block configuration of the power generation amount prediction control apparatus according to the present embodiment. Specifically, the control unit 1 reads the power generation amount prediction program from the storage unit 2 and executes it. The function block to be shown is shown. In FIG. 2, the control unit 1 includes a transmission / reception control unit 11, a date / time management unit 12, a power generation amount correction unit 13, and a first power generation amount prediction unit 14 as functional blocks for executing the power generation amount prediction control of the present embodiment. And a second power generation amount prediction unit 15. Note that the hardware configuration and functional block configuration of the power generation amount predictive control device of this embodiment are the configurations related to the processing of this embodiment for convenience of explanation, and all functions of the power generation amount prediction control device are listed. It is not a representation.

  Further, the power generation amount prediction program of this embodiment can be distributed via the communication unit 6 and a network such as the Internet. The program may be recorded on a computer-readable recording medium such as a hard disk, a flexible disk (FD), a CD-ROM, an MO, or a DVD. In this case, the program is read from the recording medium by the computer. Is executed by.

  Subsequently, the process according to the power generation amount prediction control of the present embodiment will be described in detail using a flowchart. Below, the process concerning the electric power generation prediction control of a present Example is divided into the following A, B, and C for convenience, and is demonstrated in order.

A. A process of correcting the past power generation amount (hereinafter referred to as power generation amount) to the power generation amount when the cloud amount is 0%. A process for predicting the power generation amount when the cloud amount is 0% on the prediction target day C. Process to predict the power generation amount on the forecast target day

  First, the above-mentioned “A. Process for correcting the actual power generation amount to the power generation amount when the cloud amount is 0%” will be described. FIG. 3 is a flowchart showing a process of correcting the actual power generation amount to the power generation amount when the cloud amount is 0%. The storage unit 2 stores attenuation rate data as information indicating the correlation between the cloud amount and the power generation amount in advance. As the attenuation rate data, for example, α ( = Attenuation rate of power generation) is stored together with cloud amount (upper cloud amount, middle cloud amount, lower cloud amount, total cloud amount). Further, the transmission / reception control unit 11 cooperates with the date and time management unit 12 and periodically generates power generation result data (power generation date, power generation time, power generation amount, etc.) from a power plant through an external line such as the Internet and the communication unit 6. (For example, every 30 minutes) and store the data in the storage unit 2. Also, the transmission / reception control unit 11 cooperates with the date / time management unit 12 to provide weather forecast data (including position, forecast date / time, cloud cover, etc.) via an external line such as the Internet and the communication unit 6, for example, every hour. The data is received and stored in the storage unit 2.

  In a state in which the various data are accumulated in the storage unit 2, the power generation amount correction unit 13 of the control unit 1 monitors a notification that is periodically output (for example, every 30 minutes) from the date / time management unit 12 (step 30). When the notification is received (step S1, Yes), the latest power generation result data at the notified time is read from the storage unit 2 (step S2). Further, the power generation amount correction unit 13 reads the weather forecast data corresponding to the power generation time included in the power generation amount actual data read in step S2 from the storage unit 2 (step S3). Furthermore, the power generation amount correction unit 13 reads attenuation rate data corresponding to the cloud amount included in the weather forecast data read in step S3 from the storage unit 2 (step S4).

The power generation amount correction unit 13 generates power when the power generation amount X is a cloud amount of 0% based on the actual power generation amount data (power generation amount X) and attenuation rate data (attenuation rate α) obtained by the processes of steps S2 to S4. correcting the amount (= X ₀₎ (step S5). Specifically, the power generation amount X ₀ is obtained by dividing the power generation amount X by the attenuation rate α (see the above formula (3)). Then, the power generation amount correction unit 13 stores the power generation amount X ₀ in the storage unit 2 as corrected power generation amount data (including power generation date and time, power generation amount X, cloud amount, attenuation rate α, power generation amount X _0, etc.). To do.

  Thereafter, the power generation amount correction unit 13 periodically executes the processes of steps S1 to S5 each time notification is received from the date management unit 12 (in this embodiment, executed by notification every 30 minutes). New corrected power generation amount data is stored in the storage unit 2. As a result, new corrected power generation amount data is accumulated in the storage unit 2 every 30 minutes triggered by the activation of the power generation amount prediction control device.

  FIG. 4 is a diagram illustrating an image in a case where “a process of correcting the actual power generation amount to the power generation amount when the cloud amount is 0%” is executed. Here, it is assumed that sunrise is 6 o'clock and sunset is 18 o'clock. For example, the cloud cover from 10:00 to 12:00 is the upper cloud cover: 50%, middle cloud cover: 70%, lower cloud cover 30%, total cloud cover : Shows the actual power generation amount and the corrected power generation amount when the cloud amount in other time zones is 0% (total cloud amount 0%). In FIG. 4, the thin line represents the actual power generation amount, and the thick line represents the corrected power generation amount.

  Next, the above-mentioned “B. Process for predicting the power generation amount when the cloud amount is 0% on the prediction target day” will be described. FIG. 5 is a flowchart illustrating a process of predicting the power generation amount when the cloud amount is 0% on the prediction target day. It is assumed that the storage unit 2 stores in advance, as date / time data, a state in which a prediction date (calculation date), which is a date on which the power generation amount prediction calculation is performed, and a prediction target date are associated with each other. The predicted date is usually in units of one day (daily), but it is possible to set a day of the week, a designated date, or the like. In the present embodiment, as an example, it is assumed that two days after the prediction date are stored as prediction target days.

  In the control unit 1, the first power generation amount prediction unit 14 performs a process of confirming date data in the storage unit 2 at the time specified every day in cooperation with the date management unit 12 (No in step S <b> 11). When today's date coincides with the predicted date (step S11, Yes), the prediction target date associated with the predicted date is confirmed (step S12). Then, the first power generation amount prediction unit 14 checks whether or not there is corrected power generation amount data for a predetermined number of days in the storage unit 2 (step S13). In this embodiment, in order to improve the accuracy of the “B. Prediction of power generation amount when the cloud amount is 0% on the prediction target day”, as an example, whether or not there is corrected power generation amount data for seven days or more. Check. In the present embodiment, as will be described later, since the process B is performed based on the average and standard deviation of the corrected power generation amount, the power generation amount prediction is taken into consideration after correction for seven days or more. Although the power generation amount data is requested, the number of corrected power generation amount data necessary for the prediction of the power generation amount is variable according to the required prediction accuracy, and is not limited thereto.

  For example, when it is determined that there is corrected power generation amount data for seven days or more in the process of step S13 (step S13, Yes), the first power generation amount prediction unit 14 corrects the latest 30 days from the storage unit 2. The post power generation amount data is read (step S14). Here, the first day of the last 30 days is the past data start date, and the 30th day is the past data end date. As an exception to step S14, when the corrected power generation amount data is accumulated for 7 days or more but has not reached 30 days, the corrected power generation amount data for the number of days stored in the storage unit 2 is stored. Will be read. The following description will be made assuming that the corrected power generation amount data reaches 30 days, which is a general case. However, the number of corrected power generation amount data is small even in the above exceptional cases. Except for the above, the same processing as in the above general case is performed.

The first power generation amount prediction unit 14 classifies the read 30-day corrected power generation amount data for each power generation time (in this embodiment, the power generation time in increments of 30 minutes is stored), and for each power generation time. The average and standard deviation of power generation amount X ₀ from the start date of past data to the end date of past data are calculated (steps S15 and S16), and the calculation results are respectively calculated as past average data (power generation time, average of X ₀ , etc.). including), past the standard deviation data (generation time, as includes the standard deviation of X _0, etc.), the storage unit 2 for each generation time.

First power generation amount prediction unit 14, based on the average and the standard deviation of the power generation amount X ₀ for each generation time obtained by the execution of step S15 and S16, the power generation amount in the case of 0% cloud cover in the prediction target day X _{0_predict} is predicted for each power generation time (step S17). More specifically, for example, the average of the power generation amount X _{0 at} each power generation time obtained by executing step S15 is the average of the past 30 days. This corresponds to the power generation amount in the middle between the data start date and the past data end date, that is, when the cloud amount per day is 0% (referred to as intermediate power generation amount). Therefore, in the present embodiment, for example, the trend of solar radiation over the past 20 years is obtained as a seasonal coefficient using a general solar radiation database (for example, NEDO data that can be obtained using a communication line). The intermediate power generation amount obtained based on the average and standard deviation of the power generation amount X ₀ is corrected with a seasonal coefficient, and is _{defined as} the power generation amount X _{0_predict} when the cloud amount is 0% on the prediction target day. The first power generation amount prediction unit 14 stores the power generation amount X _{0_predict} obtained here in the storage unit 2 as the first predicted power generation amount data together with the prediction target date and the prediction target time (corresponding to the power generation time). To do. The correction process using the seasonal coefficient is one of the processes for increasing the accuracy of power generation prediction. For example, the power generation amount when the cloud amount per day is 0% according to the required prediction accuracy. as it may be a X _{0_Predict,} also, those corrected by other correction factors may be _X 0_predict.

FIGS. 6A and 6B are diagrams illustrating an image when the “process for predicting the power generation amount when the cloud amount is 0% on the prediction target day” is executed. Specifically, FIG. 6-1 is a diagram illustrating an image of the corrected power generation amount data (power generation amount X ₀ ) for the latest 30 days read out in step S14, and FIG. 6-2 is predicted in step S17. It is a figure which shows the image of "the electric power generation amount (power generation amount _{X0_predict} ) in the case of the cloud amount 0% in the prediction object day".

In this embodiment, the corrected power generation amount data for the last 30 days of the forecast date is read, and the average and standard deviation of the power generation amount X ₀ for 30 days are calculated for each power generation time. The number of days for reading the corrected power generation amount data can be appropriately changed according to the required accuracy of prediction. For example, it is possible to predict the power generation amount X _{0_predict} with higher accuracy by using the corrected power generation amount data exceeding 30 days, and when it is desired to reduce the calculation amount by reducing the prediction accuracy. It is also possible to predict the power generation amount X _{0_predict} using corrected power generation amount data that is less than a day.

On the other hand, if it is determined in step S13 that there is no corrected power generation amount data for seven days or more (step S13, No), the first power generation amount prediction unit 14 uses a general solar radiation amount database (for example, communication). Prediction based on parameters such as the amount of insolation between the forecasting date and the same day in the past 20 years, and the orientation and angle of the solar panel, etc. The power generation amount on the target day is calculated (step S18), and the calculation result is set as the power generation amount X _{0_predict} when the cloud amount is 0% on the prediction target day. The first power generation amount prediction unit 14 stores the power generation amount X _{0_predict} obtained here in the storage unit 2 as the first predicted power generation amount data together with the prediction target date and the prediction target time (corresponding to the power generation time). To do.

In this embodiment, the case where there is no corrected power generation amount data for 7 days or more is a very rare case that occurs only for a few days immediately after startup. % of the power generation quantity X _{0_Predict} if not priority is to predict with high accuracy, in step S18, the X _{0_Predict,} prioritized to estimate smaller than when predicting the high accuracy method in step S14~S17 . That is, the power generation amount of the prediction target day is predicted based on the amount of solar radiation on the same day as the prediction target date for the past 20 years including the day with cloud, and the result is X _{0_predict} .

  As a result, if there is no corrected power generation amount data for a predetermined number of days (for 7 days) or more, the power generation amount for the prediction target day cannot be predicted with high accuracy, but the power generation for the prediction target day is calculated by the calculation in step S18. The amount can be estimated small. Therefore, for example, even if a power supply plan is planned in advance, such as when selling electricity, the supply plan is made with a small amount of predicted power generation in advance, so that it is unexpected It becomes possible to avoid the situation.

  Next, the “C. Process for predicting the power generation amount on the prediction target day” will be described. FIG. 7 is a flowchart illustrating a process of predicting the power generation amount on the prediction target day.

After executing the process B, the second power generation amount prediction unit 15 cooperates with the first power generation amount prediction unit 14 to store the first predicted power generation amount data (prediction target date) from the storage unit 2 on the prediction target date. , Prediction time, power generation amount X _{0_predict,} etc.) are all read out (step S21). Further, the second power generation amount prediction unit 15 corresponds to the prediction target time included in the first predicted power generation amount data read in step S21 (in this embodiment, the prediction target time in increments of 30 minutes is stored). The weather forecast data (including position, forecast date and time, cloud cover, etc.) to be read is read from the storage unit 2 (step S22). Further, the second power generation amount prediction unit 15 reads attenuation rate data corresponding to the cloud amount included in the weather forecast data read in step S22 from the storage unit 2 (step S23). Here, the corresponding attenuation rate data is read for the weather forecast data unit read in step S22.

The second power generation amount prediction unit 15 calculates the prediction target date based on the first predicted power generation amount data (power generation amount X _{0_predict} ) and attenuation rate data (attenuation rate α) obtained by the processes of steps S21 to S23. A power generation amount (= X _predict ) is predicted (step S24). Specifically, the power generation amount X _predict is calculated by multiplying the power generation amount X _{0_predict} by the attenuation rate α at each prediction target time (see the above formula (2)). Then, the second power generation amount prediction unit 15 _converts the power generation amount X _predict obtained by the calculation into second predicted power generation amount data (prediction target date and prediction target time, power generation amount X _{0_predict} , cloud amount, attenuation rate α, power generation _{Stored in the} storage unit 2 as a quantity X _predict ).

  FIG. 8 is a diagram illustrating an image when the “process for predicting the power generation amount on the prediction target day” is executed. Here, sunrise is 6 o'clock and sunset is 18 o'clock. For example, the amount of clouds from 10 o'clock to 12 o'clock is 40% upper cloud cover, 60% middle cloud cover, 80% lower cloud cover, and 90% total cloud cover. The first predicted power generation amount data and the second predicted power generation amount data when the cloud amount in other time zones is 0% (total cloud amount 0%) are shown. In FIG. 8, the thin line represents the first predicted power generation amount data (power generation amount when the cloud amount is 0% on the prediction target day), and the thick line represents the second predicted power generation amount data (power generation amount on the prediction target day). Represents.

As described above, in the present embodiment, the power generation amount correction unit 13 includes the storage unit 2 in which various parameters such as the attenuation rate α of the power generation amount with respect to the cloud amount are stored. X is corrected to the power generation amount X ₀ when the cloud amount is 0% using the attenuation rate α of the power generation amount corresponding to the cloud amount at the power generation time, and is stored as a new corrected power generation amount X _{0 at} each power generation time. Accumulate in 2. Further, the first power generation amount prediction unit 14 classifies the accumulated corrected power generation amount X ₀ into power generation time units, calculates the average and standard deviation for each power generation time, and based on the obtained average and standard deviation Thus, the power generation amount X _{0_predict} when the cloud amount on the prediction target day is 0% is predicted. The second power generation amount prediction unit 15, based on the attenuation factor of the power generation amount corresponding to the cloud cover of the power generation quantity X _{0_Predict} and prediction target day alpha, was to predict the power generation amount X _predict the prediction target day. As a result, even if the forecast date (calculation date) is set to the day before the forecast date, the day before the forecast date, etc., the power generation amount of the solar power plant on the forecast date is predicted with high accuracy. It becomes possible to do.

DESCRIPTION OF SYMBOLS 1 Control part 2 Memory | storage part 3 Input part 4 Output part 5 Display part 6 Communication part 8 Keyboard 9 Mouse 11 Transmission / reception control part 12 Date management part 13 Electric power generation amount correction | amendment part 14 1st electric power generation amount prediction part 15 2nd electric power generation amount prediction Part

In order to solve the above-described problems and achieve the object, a power generation amount predictive control device according to the present invention is a power generation amount prediction control device that predicts the power generation amount of a solar power plant, and includes a cloud amount and a power generation amount. The storage means in which the information indicating the relationship is stored, and the past power generation amount accumulated periodically, the power generation when the cloud amount is 0% based on the information indicating the relationship between the cloud amount and the power generation amount at each power generation time correcting the amount, and the power generation amount correction means for storing each power generation amount after correction as compensation after power generation amount for each power generation time of each classified in accordance with the stored the corrected power generation amount of the power generation time, classification are the mean and standard deviation of the corrected power generation amount per power generation time was calculated and, on the basis of the mean and standard deviation, the first prediction prediction target day is power generation amount on the assumption that 0% cloud cover A first power generation amount prediction means for predicting a power generation amount, Based on the information indicating the relationship between the cloud cover and the power generation amount of the first prospective power generation amount and the prediction target day, the second power generation amount prediction for predicting a second predicted power generation amount of power generation of the prediction target day And means.

A power generation amount prediction method according to the next invention is a power generation amount prediction method for predicting a power generation amount of a solar power plant, and stores information indicating a relationship between a cloud amount and a power generation amount in a storage unit. a step, out read regularly power generation amount of the past accumulated in the storage unit, the past power generation amount, cloudiness 0 based on the information indicating the relationship between the cloud cover and the power generation amount at each of the power generation time % of corrected power generation amount when the power generation amount correction step of storing in each said storage unit amount of power generation corrected as compensation after power generation amount for each power generation time of each corrected accumulated in the storage unit reads the power generation amount, the post-correction power generation amount is classified according to the power generation time, the mean and standard deviation of the corrected power generation amount for each classified generator time calculated, based on the mean and standard deviation, the predicted target If the day is assumed to be 0% cloud cover The power generation amount predicted, reading a first power generation amount prediction step of storing in the storage unit a generation amount predicted as a first predicted power generation amount, the first prospective power generation amount stored in the storage unit, the based on the information indicating the relationship between the cloud cover and the power generation amount of the first prospective power generation amount and the prediction target day, to predict the power generation amount of the prediction target day, the power generation amount predicted as a second predicted power generation amount And a second power generation amount prediction step stored in the storage unit.

A power generation amount prediction program according to the next invention is a power generation amount prediction program executed by a computer that operates as a power generation amount prediction control device that predicts the power generation amount of a solar power plant, and a relationship between a cloud amount and a power generation amount. storing instructions to be stored in the storage unit information indicating, out read power generation amount of the past accumulated in regularly the storage unit, the past power generation amount, and the cloud cover and the power generation amount at each of the power generation time based on the information indicating the relationship is corrected to the power generation amount in the case of 0% cloud cover, the power generation amount correction procedures accumulated in each of the storage unit to the power generation amount of the corrected as compensation after power generation amount for each power generation time of each said reading the corrected power generation amount accumulated in the storage unit, the post-correction power generation amount is classified according to the power generation time, to calculate the mean and standard deviation of the corrected power generation amount for each classified generator time, the mean and Standard deviation Based on, it predicts the power generation amount when the prediction target day is assumed to be 0% cloud cover, the first power generation amount prediction procedure stored in the storage unit a generation amount predicted as the first prospective power generation amount, It reads the first prospective power generation amount stored in the storage unit, based on the information indicating the relationship between the cloud cover and the power generation amount of the first prospective power generation amount and the prediction target day, the power generation amount of the prediction target day The computer is caused to execute a second power generation amount prediction procedure for storing the predicted power generation amount in the storage unit as a second predicted power generation amount.

Claims (14)

In the power generation amount prediction control device that predicts the power generation amount of the solar power plant,
Storage means for storing correlation information indicating a correlation between the cloud amount and the power generation amount;
Based on the correlation information at the power generation time, the past power generation accumulated periodically is corrected to a power generation amount when the cloud amount is 0%, and the corrected power generation amount is newly corrected at each power generation time. Power generation amount correcting means for accumulating as a quantity;
The accumulated power generation amount after correction is classified into power generation time units, and the average and standard deviation for each power generation time are calculated. Based on the average and standard deviation, this is the power generation amount when the cloud amount is 0% on the prediction target day. First power generation amount predicting means for predicting the first predicted power generation amount;
Based on the first predicted power generation amount and the correlation information on the prediction target date, second power generation amount prediction means for predicting a second predicted power generation amount that is the power generation amount of the prediction target date;
A power generation amount predictive control device comprising: As the correlation information, when the attenuation rate of the power generation amount with respect to the cloud amount is stored in the storage means,
The power generation amount correcting means includes
Reading the attenuation rate corresponding to the cloud amount of the past power generation amount at the power generation time from the storage unit, and dividing the past power generation amount by the read attenuation rate, to obtain a corrected power generation amount,
The power generation amount prediction control apparatus according to claim 1. The first power generation amount prediction means includes:
A trend of past solar radiation obtained using a general solar radiation database is obtained as a seasonal coefficient, and the first predicted power generation is obtained by correcting the power generation obtained based on the average and standard deviation by the seasonal coefficient. And
The power generation amount prediction control apparatus according to claim 1 or 2, The first power generation amount prediction means includes:
When the corrected power generation amount for a predetermined number of days necessary for the prediction of the first predicted power generation amount is accumulated, the process of calculating the average and standard deviation is executed.
The power generation amount prediction control apparatus according to claim 1, 2, or 3. The first power generation amount prediction means includes:
When the corrected power generation amount for the predetermined number of days is not accumulated, the power generation amount for the prediction target day is calculated based on the general solar radiation amount database and the direction and angle of the solar panel, and the calculation result is The first predicted power generation amount
The power generation amount prediction control apparatus according to claim 4. As the correlation information, when the attenuation rate of the power generation amount with respect to the cloud amount is stored in the storage means,
The second power generation amount prediction means includes
A second predicted power generation amount is obtained by reading the attenuation rate corresponding to the cloud amount of the prediction target day from the storage unit and multiplying the first predicted power generation amount by the attenuation rate.
The power generation amount prediction control apparatus according to any one of claims 1 to 5, wherein The correlation information is defined based on multivariate analysis using past power generation amount,
The power generation amount prediction control apparatus according to any one of claims 1 to 6, A power generation amount prediction method for predicting the power generation amount of a solar power plant,
A storage step of storing, in the storage unit, correlation information indicating a correlation between a cloud amount and a power generation amount;
The past power generation amount periodically stored in the storage unit is read for each power generation time, and the past power generation amount is corrected to the power generation amount when the cloud amount is 0% based on the correlation information at the power generation time, A power generation amount correcting step for storing the corrected power generation amount in the storage unit as a new corrected power generation amount for each power generation time; and
Read the corrected power generation amount stored in the storage unit, classify the corrected power generation amount into power generation time units, calculate the average and standard deviation for each power generation time, and based on the average and standard deviation the prediction target date A first power generation amount prediction step of predicting a power generation amount when the cloud amount is 0% in the storage unit, and storing the predicted power generation amount as a first predicted power generation amount in the storage unit;
Reading the first predicted power generation amount stored in the storage unit, predicting the power generation amount on the prediction target day based on the first predicted power generation amount and the correlation information on the prediction target date, and calculating the predicted power generation amount A second power generation amount predicting step for storing the second predicted power generation amount in the storage unit;
A method for predicting the amount of power generation, comprising: As the correlation information, when the attenuation rate of the power generation amount with respect to the cloud amount is stored in the storage unit,
In the power generation amount correcting step,
Read the attenuation rate corresponding to the cloud amount at the time of power generation of the past power generation amount from the storage unit, and obtain the corrected power generation amount by dividing the past power generation amount by the read attenuation rate.
The power generation amount prediction method according to claim 8. In the first power generation amount prediction step,
When the corrected power generation amount for a predetermined number of days necessary for the prediction of the first predicted power generation amount is accumulated in the storage unit, the process of calculating the average and standard deviation is executed,
A trend of past solar radiation obtained by using a general solar radiation database is obtained as a seasonal coefficient, and the first predicted power generation is obtained by correcting the power generation obtained based on the average and standard deviation by the seasonal coefficient. The amount,
The power generation amount prediction method according to claim 8 or 9, characterized in that. In the first power generation amount prediction step,
When the corrected power generation amount for the predetermined number of days is not accumulated in the storage unit, based on the general solar radiation amount database and the azimuth and angle of the solar panel, calculate the power generation amount on the prediction target day, The calculation result is stored in the storage unit as the first predicted power generation amount,
The power generation amount prediction method according to claim 10. As the correlation information, when the attenuation rate of the power generation amount with respect to the cloud amount is stored in the storage unit,
In the second power generation amount prediction step,
Reading the attenuation rate corresponding to the cloud amount of the prediction target day from the storage unit, and applying the attenuation rate to the first predicted power generation amount, to obtain a second predicted power generation amount,
The power generation amount prediction method according to any one of claims 8 to 11, characterized in that: A defining step for defining the correlation information based on multivariate analysis using past power generation amount,
The power generation amount prediction method according to any one of claims 8 to 12, further comprising: A power generation amount prediction program that is executed by a computer that operates as a power generation amount prediction control device that predicts the power generation amount of a solar power plant,
A storage procedure for storing correlation information indicating a correlation between cloud amount and power generation amount in a storage unit,
The past power generation amount periodically stored in the storage unit is read for each power generation time, and the past power generation amount is corrected to the power generation amount when the cloud amount is 0% based on the correlation information at the power generation time, A power generation amount correction procedure for storing the corrected power generation amount in the storage unit as a new corrected power generation amount for each power generation time,
Read the corrected power generation amount stored in the storage unit, classify the corrected power generation amount into power generation time units, calculate the average and standard deviation for each power generation time, and based on the average and standard deviation the prediction target date A first power generation amount prediction procedure for predicting a power generation amount in a case where the cloud amount is 0% and storing the predicted power generation amount as a first predicted power generation amount in the storage unit,
Reading the first predicted power generation amount stored in the storage unit, predicting the power generation amount on the prediction target day based on the first predicted power generation amount and the correlation information on the prediction target date, and calculating the predicted power generation amount A second power generation amount prediction procedure stored in the storage unit as a second predicted power generation amount;
A power generation amount prediction program characterized by causing a computer to execute.

Images