KR102568489B1 - Method, server and computer program for predicting solar radiation - Google Patents

Abstract

The method of predicting insolation includes obtaining a weather model including weather information based on a grid of latitude and longitude standards; extracting at least one variable related to insolation through correlation analysis of the meteorological information; generating a multiple linear regression model based on the at least one extracted variable; generating a virtual lattice by adjusting a lattice of the weather model based on the multiple linear regression model; and arranging the weather information on the virtual grid and estimating the amount of solar radiation based on the arranged weather information.

Description

Solar radiation prediction method, server and computer program {METHOD, SERVER AND COMPUTER PROGRAM FOR PREDICTING SOLAR RADIATION}

The present invention relates to a method for predicting insolation, a server and a computer program.

When a power generation company that owns a solar power plant submits a plan for generation amount for the next 36 hours to the Korea Power Exchange, the reliability of the electricity supplied by the solar power plant is determined by the prediction error rate. Therefore, since the photovoltaic power plant needs to submit an accurate power generation plan in order to increase the reliability of supply power, it is necessary to accurately predict the amount of insolation of the photovoltaic power plant.

As a method of predicting the insolation of a solar power plant, the insolation of the power plant can be predicted from the insolation information of a weather station close to the photovoltaic power plant using triangulation, but this conventional insolation prediction method has significantly poor prediction accuracy.

In addition, as other methods of predicting insolation of a solar power plant, there are a linear regression method using past power generation data of the corresponding power plant and a method of directly installing a solar radiation meter. However, the linear regression method using past power generation data can cause a large difference from the actual solar radiation in the process of converting the data into actual power, and the method of directly installing the solar radiation meter uses the solar radiation meter used in the weather station for each solar power plant. There is a problem that the cost burden is high to install one by one.

In addition, in the method of directly installing the pyranometer, when the pyranometer has a low resolution or a short lifespan, reliability of irradiance prediction may be deteriorated, and maintenance costs may increase.

Korean Patent Publication No. 10-2017-0056837 (published on May 24, 2017) Korean Registered Patent Publication No. 10-2093796 (registered on March 20, 2020)

The present invention is to solve the problems of the prior art, and to provide a solar radiation prediction method, a server, and a computer program capable of providing a solar radiation map by predicting solar radiation.

In addition, the present invention is to provide a solar radiation prediction method, a server, and a computer program capable of predicting solar radiation of a solar power plant using a solar radiation map.

However, the technical problem to be achieved by the present embodiment is not limited to the technical problems described above, and other technical problems may exist.

As a means for achieving the above-described technical problem, an embodiment of the present invention provides a method for predicting insolation in a target region, comprising: obtaining a weather model including weather information based on a grid of latitude and longitude standards; extracting at least one variable related to insolation through correlation analysis of the meteorological information; generating a multiple linear regression model based on the at least one extracted variable; generating a virtual lattice by adjusting a lattice of the weather model based on the multiple linear regression model; and arranging the weather information on the virtual grid and estimating the solar radiation based on the arranged weather information.

Another embodiment of the present invention provides a solar radiation prediction server for a target region, comprising: a weather model unit acquiring a weather model including weather information based on a grid of latitude and longitude standards; a variable extraction unit extracting at least one variable related to insolation through a correlation analysis with respect to the meteorological information; a multi-linear regression model generation unit generating a multi-linear regression model based on the at least one or more extracted variables; a virtual lattice generation unit generating a virtual lattice by adjusting a lattice of the weather model based on the multiple linear regression model; and a solar radiation prediction unit that arranges the weather information on the virtual grid and predicts the solar radiation based on the disposed weather information.

Another embodiment of the present invention is a computer program stored in a computer readable recording medium including a sequence of instructions for predicting the amount of solar radiation in a target area, wherein the computer program, when executed by a computing device, determines the latitude and longitude reference A meteorological model including meteorological information based on a lattice is obtained, at least one or more variables related to insolation are extracted through a correlation analysis of the meteorological information, and a multiple linear regression model is based on the at least one or more extracted variables. To generate a virtual grid by adjusting a grid of the weather model based on the multiple linear regression model, to place the weather information on the virtual grid, and to predict the amount of solar radiation based on the arranged weather information. It is possible to provide a computer program stored on a computer readable recording medium, including a sequence of instructions to do.

The above-described means for solving the problems is only illustrative and should not be construed as limiting the present invention. In addition to the exemplary embodiments described above, there may be additional embodiments described in the drawings and detailed description.

According to any one of the above-described problem solving means of the present invention, the amount of solar radiation can be predicted more accurately by predicting the amount of solar radiation by extracting and utilizing variables highly correlated with the amount of solar radiation through correlation analysis.

In addition, by readjusting the existing meteorological model lattice based on latitude and longitude into a virtual lattice at equal intervals, insolation prediction information predicted per unit area may be more accurately displayed.

In addition, solar radiation prediction information for predicting power generation of the power plant may be provided using the solar radiation map in which the solar radiation prediction information is constructed.

1 is a configuration diagram of a solar radiation prediction server according to an embodiment of the present invention.
2 is an exemplary diagram for explaining a candidate multiple linear regression model according to an embodiment of the present invention.
3 is an exemplary diagram for calculating regression coefficients for each grid based on Thiessen weights according to an embodiment of the present invention.
4(a) is a lattice of a weather model according to an embodiment of the present invention, and (b) is an exemplary diagram for explaining a virtual lattice according to an embodiment of the present invention.
5 is an exemplary diagram for explaining a virtual grid generated by adjusting a grid of a weather model according to an embodiment of the present invention.
6 is a flowchart of a solar radiation prediction method according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail so that those skilled in the art can easily practice the present invention with reference to the accompanying drawings. However, the present invention may be embodied in many different forms and is not limited to the embodiments described herein. And in order to clearly explain the present invention in the drawings, parts irrelevant to the description are omitted, and similar reference numerals are attached to similar parts throughout the specification.

Throughout the specification, when a part is said to be "connected" to another part, this includes not only the case where it is "directly connected" but also the case where it is "electrically connected" with another element interposed therebetween. . In addition, when a part "includes" a certain component, this means that it may further include other components, not excluding other components, unless otherwise stated, and one or more other characteristics. However, it should be understood that it does not preclude the possibility of existence or addition of numbers, steps, operations, components, parts, or combinations thereof.

In this specification, a "unit" includes a unit realized by hardware, a unit realized by software, and a unit realized using both. Further, one unit may be realized using two or more hardware, and two or more units may be realized by one hardware.

In this specification, some of the operations or functions described as being performed by a terminal or device may be performed instead by a server connected to the terminal or device. Likewise, some of the operations or functions described as being performed by the server may also be performed in a terminal or device connected to the corresponding server.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

1 is a configuration diagram of a solar radiation prediction server according to an embodiment of the present invention.

Referring to FIG. 1 , the solar radiation prediction server 100 includes a weather model unit 110, a variable extractor 120, a multiple linear regression model generator 130, a virtual grid generator 140, and a solar radiation predictor 150. ) may be included. The variable extraction unit 120 may include a candidate model generation unit 121, a candidate prediction unit 122, and a comparison unit 123, and the multiple linear regression model generation unit 130 may include a grid-by-lattice model generation unit 131 ) may be included. However, the above components 110 to 150 are merely examples of components that can be controlled by the solar radiation prediction server 100 .

Each component of the solar radiation prediction server 100 of FIG. 1 is generally connected through a network. For example, as shown in FIG. 1, a virtual model unit 110, a variable extraction unit 120, a multiple linear regression model generation unit 130, a virtual grid generation unit 140, and a solar radiation prediction unit 150 may be connected simultaneously or at intervals of time.

A network refers to a connection structure capable of exchanging information between nodes such as terminals and servers, such as a local area network (LAN), a wide area network (WAN), and the Internet (WWW: World Wide Web), wired and wireless data communication network, telephone network, and wired and wireless television communication network. Examples of wireless data communication networks include 3G, 4G, 5G, 3rd Generation Partnership Project (3GPP), Long Term Evolution (LTE), World Interoperability for Microwave Access (WIMAX), Wi-Fi, Bluetooth communication, infrared communication, ultrasonic communication, visible light communication (VLC: Visible Light Communication), LiFi, and the like, but are not limited thereto.

Referring to FIG. 1, the insolation prediction server 100 obtains a weather model including meteorological information based on a grid of latitude and longitude standards, and analyzes the correlation of the obtained meteorological information to determine variables highly correlated with insolation. Insolation can be predicted more accurately by extracting and utilizing it to predict insolation.

In addition, the solar radiation prediction server 100 readjusts the existing meteorological model lattice based on latitude and longitude into a virtual lattice at equal intervals, and uses the solar radiation map in which the solar radiation prediction information per unit area is built to predict the amount of power generated by the power plant. can be provided more accurately.

Hereinafter, each configuration of the solar radiation prediction server 100 will be described in more detail.

The weather model unit 110 according to an embodiment of the present invention may obtain a weather model including weather information based on a grid based on latitude and longitude. For example, the weather model unit 110 may obtain a weather model including 78 types of weather information from the Korea Meteorological Administration.

For example, the weather model unit 110 may obtain a single-sided numerical forecast model as a weather model from the Korea Meteorological Administration and extract, for example, 78 variables from it. In the present application, it is described that the weather model unit 110 extracts 78 variables as an example, but the weather model unit 110 is not limited thereto, and the weather model unit 110 may extract the number of variables capable of accurately predicting insolation.

The weather model unit 110 may store 78 variable values for each date and time for each grid based on latitude and longitude.

The 78 types of weather information included in the weather model include weather prediction information for the future 48 hours (+0:00 ~ +48:00). The 78 types of meteorological information may include surface shortwave radiation, direct shortwave radiation, and scattered shortwave radiation in addition to insolation information.

The variable extractor 120 according to an embodiment of the present invention may extract at least one or more variables related to insolation through correlation analysis of weather information. For example, the variable extractor 120 may extract variables having a great influence on insolation from among 78 kinds of weather information.

The variable extractor 120 may extract at least one variable related to insolation through correlation analysis with weather information.

For example, the variable extractor 120 may extract variables related to insolation from among 78 types of weather prediction information included in a weather model based on correlation analysis and multiple linear regression on weather information. Variables extracted by the variable extractor 120 may be used as input variables of a solar radiation prediction model for solar radiation prediction.

The variable extractor 120 may perform correlation analysis for each grid of latitude and longitude standards included in the weather model. Here, the correlation analysis may be Pearson's correlation analysis. The variable extractor 120 may compare Pearson's correlation coefficients through Pearson's correlation analysis. Here, the Pearson correlation coefficient is a numerical value obtained by quantifying the linear correlation between variables.

The candidate model generator 121 according to an embodiment of the present invention may generate a plurality of candidate multiple linear regression models based on a plurality of variable sets including at least one different variable.

2 is an exemplary diagram for explaining a candidate multiple linear regression model according to an embodiment of the present invention. Referring to FIG. 2 , the candidate model generation unit 121 may generate a plurality of variable sets including at least one variable different from one or more extracted variables. The candidate model generation unit 121 may generate, for example, six candidate multiple linear regression models based on the generated variable sets.

For example, the first candidate multiple linear regression model (211, TEST 0) is predicted insolation, precipitation, east-west wind, north-south wind, temperature, relative humidity, visibility, dew point temperature, lowest cloud cover, lower cloud cover extracted through correlation analysis. , middle and upper clouds can be used as input variables.

The second candidate multiple linear regression model (212, TEST 1) inputs predicted insolation, precipitation, east-west wind, north-south wind, temperature, relative humidity, dew point temperature, lowest cloud, lower cloud, middle cloud, and upper cloud extracted through correlation analysis. can be used as a variable.

The third candidate multiple linear regression model (213, TEST2) may use predicted insolation extracted through correlation analysis, lowest cloud, lower cloud, middle cloud, and upper cloud as input variables.

The fourth candidate multiple linear regression model (214, TEST3) may use predicted insolation, lowest cloud, lower cloud, middle cloud, upper cloud, and precipitation extracted through correlation analysis as input variables.

The fifth candidate multiple linear regression model (215, TEST4) may use predicted insolation, lowest cloud, lower cloud, middle cloud, upper cloud, and visibility extracted through correlation analysis as input variables.

The sixth candidate multiple linear regression model (216, TEST5) can use predicted insolation, precipitation, east-west wind, north-south wind, air temperature, relative humidity, visibility and dew point temperature extracted through correlation analysis as input variables.

The candidate predictor 122 according to an embodiment of the present invention may predict solar radiation based on a plurality of candidate multiple linear regression models. For example, the candidate predictor 122 may predict insolation through multiple linear regression analysis for each of the plurality of candidate multiple linear regression models 211 to 216 . In addition, the candidate prediction unit 122 may predict hourly solar radiation through hourly linear regression analysis for each of the plurality of candidate multiple linear regression models 211 to 216 .

The comparator 123 according to an embodiment of the present invention may compare solar radiation predicted based on a plurality of candidate multiple linear regression models with actual solar radiation based on a meteorological model.

For example, the comparator 123 may compare the actual amount of solar radiation 221 (X_RAW) based on a meteorological model and the amount of solar radiation predicted based on a candidate multiple linear regression model (222, X_NEW).

The multiple linear regression model generation unit 130 according to an embodiment of the present invention may generate a multiple linear regression model based on at least one extracted variable. For example, the multi-linear regression model generation unit 130 may generate a candidate multi-linear regression model having the most similar amount of solar radiation to the actual amount of solar radiation as the multi-linear regression model. In this way, more accurate solar radiation prediction information can be generated by generating a multiple linear regression model by extracting and utilizing variables highly correlated with solar radiation.

For example, in the multiple linear regression model generation unit 130, the comparison unit 123 compares the solar radiation predicted based on the plurality of candidate multiple linear regression models 211 to 216 with the solar radiation based on the meteorological model, and the solar radiation prediction rate. The fourth candidate multiple linear regression model (214, TEST3) confirmed to have the highest value may be selected as the multiple linear regression model. Here, the set of variables used in the selected multiple linear regression model may include, for example, at least one of predicted insolation, lowest cloud, lower cloud, middle cloud, upper cloud, and precipitation data. Equation 1 below represents the selected multiple linear regression model.

<Equation 1>

Here, 'n' means the grid number based on latitude and longitude, and 'ß' means the regression coefficient of each variable.

The lattice-by-lattice model generator 131 calculates the regression coefficient of each of at least one or more variables for each lattice-by-lattice based on the distance weight based on the distance between the weather station and the target area and the Thiessen weight for the meteorological model, thereby generating a grid-by-lattice multiple linear regression model. can create

For example, the model generation unit 131 for each grid may calculate a regression coefficient for each of at least one variable for each grid based on latitude and longitude in Equation 1 selected based on the multiple linear regression model.

For example, the model generator 131 for each grid may calculate the regression coefficient ß of each variable for each grid n based on latitude and longitude based on Equation 2 below.

<Equation 2>

Here, when 'distance weight + Thyssen weight' is '1', the corresponding grid means a weather station.

3 is an exemplary diagram for calculating regression coefficients for each grid based on Thiessen weights according to an embodiment of the present invention.

The Thyssen method is a method of calculating the average value by assigning a weight to the area ratio of Thyssen polygons constructed around a reference point. The lattice-by-lattice model generator 131 may construct a Thyssen polygon based on ß _n (300) to calculate ß _n (300) using the Thyssen method.

For example, the lattice-specific model generation unit 2131 draws an inner triangle based on ß _n (300) in the Thyssen polygon, and divides the Thyssen polygon into three areas inside the Thyssen polygon and four areas outside the Thyssen polygon based on the inner triangle. can be divided into areas. The lattice-specific model generation unit 131, for example, converts the Thyssen polygon into A _{1 inside} (311), A _{1 outside} (321), A _{2 inside} (312), A _{2 outside} (322), A _{3 inside} ( 313) and A _{3 outside} (323).

At this time, the grid-by-lattice model generation unit 131 may derive β _{n Thyssen} in which only the Thyssen weight for the specific grid 'n' is reflected based on Equation 3 below, for example.

<Equation 3>

The lattice-by-lattice model generator 131 may derive ß _n in which both the Thyssen weight (W ₁ ) and the distance weight (W ₂ ) for a specific lattice 'n' are reflected, based on Equation 4 below, for example.

<Equation 4>

The virtual grid generator 140 according to an embodiment of the present invention may generate a virtual grid by adjusting the grid of a weather model based on the selected multiple linear regression model. For example, the virtual grid generator 140 may adjust the grid of the weather model based on a multiple linear regression model for each grid based on latitude and longitude.

4(a) is a lattice of a weather model according to an embodiment of the present invention, and (b) is an exemplary diagram for explaining a virtual lattice according to an embodiment of the present invention.

Referring to (a) of FIG. 4 , the lattice of the weather model is formed based on general latitude and longitude standards used in the earth model. However, the unit of insolation is expressed per unit area as W/m ² , and the unit area of the meteorological model needs to be constant. That is, when the meteorological model shown in (a) of FIG. 4 is used as it is in deriving the amount of solar radiation, the unit area of the grid becomes wider as the latitude approaches 0 degrees (A), so the amount of solar radiation may be displayed inaccurately.

Accordingly, the virtual grid generator 140 may generate a virtual grid obtained by adjusting the grid of the weather model, as shown in (b) of FIG. 4 . Here, the virtual lattice may be equally spaced. For example, the virtual grid generator 140 may generate a virtual grid at regular intervals based on 1.5 km based on a multiple linear regression model for each grid.

5 is an exemplary diagram for explaining a virtual grid generated by adjusting a grid of a weather model according to an embodiment of the present invention. Referring to FIG. 5 , the virtual lattice generator 140 may adjust the lattice of the weather model at regular intervals based on the regression coefficients ß _n of the multiple linear regression model for each lattice, for example.

For example, the virtual grid generator 140 is based on the areas of A1 (511), A2 (512), A4 (513), and A5 (514), which are grids of a weather model overlapping the equally spaced grid N1 (520). Thus, the regression coefficient for N1 520 can be calculated. For example, the virtual lattice generator 140 may calculate a regression coefficient for N1 520 based on Equation 5 below.

<Equation 5>

Here, ß _N1 is Regression coefficients for N1 (520), ß _A1 to ß _A4 are regression coefficients for A1 (511), A2 (512), A4 (513), and A5 (514), respectively, and W ₁ to W ₄ are A1 ( 511), A2 (512), A4 (513), A5 (514) is a weight based on the ratio of the area overlapped with N1 (520) to the total area.

The solar radiation predictor 150 according to an embodiment of the present invention may arrange weather information on a virtual grid and predict solar radiation based on the disposed weather information.

The solar radiation predictor 150 may map the grid of the weather model and the generated virtual grid based on latitude and longitude, and rearrange weather information of the weather model on the virtual grid. The solar radiation predictor 150 may extract variables highly correlated with solar radiation from the rearranged meteorological information through correlation analysis, and construct solar radiation prediction information per virtual grid unit based on the extracted variables.

For example, the insolation prediction unit 150 provides insolation power generation amount prediction information (eg, 48-hour prediction) per virtual grid unit at 6-hour intervals (eg, UTC standard, 00:00, 06:00, 12:00, 18:00) four times a day. , +0 hour ~ +48 hour) can be built. For another example, the solar radiation prediction unit 150 may match the latitude and longitude of the power plant with a virtual grid to establish solar radiation prediction information of the power plant. Insolation prediction information from power plants can be used as data when bidding for power generation at power exchanges.

According to the present invention, solar radiation between power plants in different locations can be accurately compared by constructing prediction information on solar radiation of a power plant based on a virtual grid at regular intervals.

6 is a flowchart of a solar radiation prediction method according to an embodiment of the present invention. The solar radiation prediction method shown in FIG. 6 includes steps processed time-sequentially according to the embodiment shown in FIGS. 1 to 5 . Therefore, even if the content is omitted below, it is also applied to the method of predicting solar radiation in the solar radiation prediction server 100 according to the embodiment shown in FIGS. 1 to 5 .

In step S610, the insolation prediction server 100 may acquire a weather model including weather information based on a grid based on latitude and longitude.

In step S620, the insolation prediction server 100 may extract at least one or more variables related to insolation through correlation analysis with weather information.

In step S630, the solar radiation prediction server 100 may generate a multiple linear regression model based on at least one extracted variable.

In step S640, the solar radiation prediction server 100 may generate a virtual grid by adjusting the grid of the meteorological model based on the multiple linear regression model (finally determined). The solar radiation prediction server 100 may include a virtual grid generator 140 .

In step S650, the solar radiation prediction server 100 may arrange weather information on the virtual grid and predict solar radiation based on the arranged weather information. The solar radiation prediction server 100 may include a solar radiation prediction unit 150 .

In the above description, steps S610 to S650 may be further divided into additional steps or combined into fewer steps, depending on an implementation example of the present invention. Also, some steps may be omitted as needed, and the order of steps may be switched.

The method of predicting solar radiation in the solar radiation prediction server 100 described with reference to FIGS. 1 to 6 is in the form of a computer program stored in a computer-readable recording medium executed by a computer or a recording medium including instructions executable by a computer. can also be implemented. In addition, the method of predicting solar radiation in the solar radiation prediction server 100 described with reference to FIGS. 1 to 6 may be implemented in the form of a computer program stored in a computer-readable recording medium executed by a computer.

Computer readable media can be any available media that can be accessed by a computer, and includes both volatile and nonvolatile media, removable and non-removable media. Also, computer readable recording media may include computer storage media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data.

The above description of the present invention is for illustrative purposes, and those skilled in the art can understand that it can be easily modified into other specific forms without changing the technical spirit or essential features of the present invention. will be. Therefore, the embodiments described above should be understood as illustrative in all respects and not limiting. For example, each component described as a single type may be implemented in a distributed manner, and similarly, components described as distributed may be implemented in a combined form.

The scope of the present invention is indicated by the following claims rather than the detailed description above, and all changes or modifications derived from the meaning and scope of the claims and equivalent concepts should be construed as being included in the scope of the present invention. do.

100: solar radiation prediction server
110: weather model unit
120: variable extraction unit
130: multiple linear regression model generation unit
140: virtual grid generation unit
150: insolation prediction unit

Claims (16)

In the solar radiation prediction method of the target area,
obtaining a weather model including weather information based on a grid of latitude and longitude standards;
extracting at least one variable related to insolation through correlation analysis of the meteorological information;
generating a multiple linear regression model based on the at least one extracted variable;
generating a virtual lattice by adjusting a lattice of the weather model based on the multiple linear regression model; and
arranging the meteorological information on the virtual grid and estimating the amount of solar radiation based on the arranged meteorological information;
Including, insolation prediction method.
According to claim 1,
The correlation analysis is a Pearson correlation analysis, insolation prediction method.
According to claim 1,
The step of extracting at least one or more variables,
generating a plurality of candidate multiple linear regression models based on a plurality of variable sets including at least one or more different variables;
predicting solar radiation based on the plurality of candidate multiple linear regression models; and
Comparing the predicted solar radiation based on the plurality of candidate multiple linear regression models and the actual solar radiation based on the weather model, the solar radiation prediction method.
According to claim 3,
Generating the multiple linear regression model,
Further comprising generating a candidate multiple linear regression model having the smallest numerical difference between the actual solar radiation and solar radiation among the plurality of candidate multiple linear regression models as the multiple linear regression model, the solar radiation prediction method.
According to claim 4,
The set of variables used in the multiple linear regression model includes at least one of predicted insolation, lowest cloud, lower cloud, middle cloud, upper cloud, and precipitation data.
According to claim 1,
Generating the multiple linear regression model,
generating a multiple linear regression model for each grid by calculating a regression coefficient of each of the at least one or more variables for each grid based on a distance weight based on a distance between a weather station and a target region of the weather model and a Thiessen weight; Including, insolation prediction method.
According to claim 1,
The virtual grid is equally spaced, insolation prediction method.
In the solar radiation prediction server of the target area,
a weather model unit acquiring a weather model including weather information based on a grid of latitude and longitude standards;
a variable extraction unit extracting at least one variable related to insolation through a correlation analysis with respect to the meteorological information;
a multi-linear regression model generation unit generating a multi-linear regression model based on the at least one or more extracted variables;
a virtual lattice generation unit generating a virtual lattice by adjusting a lattice of the weather model based on the multiple linear regression model; and
A solar radiation prediction unit that arranges the weather information on the virtual grid and predicts the solar radiation based on the arranged weather information.
Including, insolation prediction server.
According to claim 8,
The correlation analysis is a Pearson correlation analysis, solar radiation prediction server.
According to claim 8,
The variable extraction unit,
a candidate model generating unit generating a plurality of candidate multiple linear regression models based on a plurality of variable sets including at least one or more different variables;
a candidate predictor predicting solar radiation based on the plurality of candidate multiple linear regression models; and
And a comparison unit for comparing the predicted solar radiation based on the plurality of candidate multiple linear regression models and the actual solar radiation based on the weather model, solar radiation prediction server.
According to claim 10,
The multiple linear regression model generation unit,
Of the plurality of candidate multiple linear regression models, a candidate multiple linear regression model having the smallest numerical difference between the actual solar radiation and solar radiation is generated as the multiple linear regression model, the solar radiation prediction server.
According to claim 11,
The set of variables used in the multiple linear regression model includes at least one of predicted insolation, lower cloud, lower cloud, middle cloud, upper cloud, and precipitation data, solar radiation prediction server.
According to claim 8,
The multiple linear regression model generation unit,
A grid-by-lattice model for generating the grid-by-lattice multiple linear regression model by calculating a regression coefficient of each of the at least one variable for each grid based on a distance weight based on a distance between a weather station and a target region of the weather model and a Thiessen weight. Solar radiation prediction server further comprising a generation unit.
According to claim 8,
The virtual grid is equally spaced, solar radiation prediction server.
A computer program stored in a computer readable recording medium including a sequence of instructions for predicting the amount of solar radiation in a target area,
When the computer program is executed by a computing device,
Obtaining a meteorological model including meteorological information based on a grid of latitude and longitude standards;
Extracting at least one variable related to insolation through correlation analysis of the meteorological information;
Generating a multiple linear regression model based on the at least one or more extracted variables;
Adjusting the lattice of the weather model based on the multiple linear regression model to generate a virtual lattice;
A computer program stored on a computer-readable recording medium comprising a sequence of instructions for disposing the weather information on the virtual grid and predicting the amount of solar radiation based on the arranged weather information. In the solar radiation prediction method of the target area,
obtaining a weather model including weather information based on a grid of latitude and longitude standards;
extracting at least one variable related to solar radiation through correlation analysis between the weather information and solar radiation;
generating a multiple linear regression model for each lattice based on the at least one extracted variable;
generating a virtual lattice and generating a multiple linear regression model for the virtual lattice based on the multiple linear regression model for each lattice; and
arranging the meteorological information on the virtual grid and predicting insolation based on the arranged meteorological information and a multiple linear regression model for the virtual grid;
Including, insolation prediction method.

Images