阅读说明：本技术 一种太阳辐射量出现概率分布解析方法、太阳辐射量出现概率分布解析系统、太阳辐射量出现概率分布解析程序、太阳辐射量归一化统计解析系统、太阳辐射量归一化统计解析方法及太阳辐射量归一化统计解析程序 (Solar radiation amount occurrence probability distribution analysis method, solar radiation amount occurrence probability distribution analysis system, solar radiation amount occurrence probability di) 是由 高桥秀幸 堀内良雄 于 2019-09-13 设计创作，主要内容包括：本发明要解决的问题是：简化太阳辐射量数据的统计处理,或基于随机性解释来解析直方图化的太阳辐射量出现概率分布并将太阳辐射量数学建模为计算式,所述太阳辐射量数据用于计算预测,所述计算预测需要将太阳辐射量作为变量的情况计算。本发明使计算机的处理器执行以下步骤：接受太阳辐射量数据的输入并存储在数据库中；基于所述太阳辐射量数据来确定太阳辐射量出现概率分布；基于所述太阳辐射量出现概率分布、和第一概率密度函数及第二概率密度函数,来确定太阳辐射量估计模型并存储在所述数据库中,所述太阳辐射量估计模型由与所述第一概率密度函数关联的第一众数及第一偏差、与所述第二概率密度函数关联的第二众数及第二偏差、所述第一概率密度函数的分配系数、和所述第二概率密度函数的分配系数组成；将积分值作为重叠度存储在所述数据库中,所述积分值表示基于所述太阳辐射量估计模型的概率分布的重叠。本发明使计算机的处理器执行以下步骤：管理步骤,接受太阳辐射量数据的输入；分析步骤,基于太阳辐射量数据来确定太阳辐射量出现概率分布；其中,太阳辐射量出现概率分布将无量纲量作为随机变量。(The invention aims to solve the problems that: the statistical processing of the solar radiation amount data is simplified, or the histogram-formed solar radiation amount occurrence probability distribution is resolved based on stochastic interpretation and the solar radiation amount is mathematically modeled as a calculation formula, the solar radiation amount data being used for calculation prediction that requires calculation with the solar radiation amount as a variable. The invention causes a processor of a computer to perform the steps of: receiving the input of solar radiation data and storing the solar radiation data in a database; determining a solar radiation amount occurrence probability distribution based on the solar radiation amount data; determining a solar radiation amount estimation model based on the solar radiation amount occurrence probability distribution, and a first probability density function and a second probability density function, the solar radiation amount estimation model being composed of a first mode and a first deviation associated with the first probability density function, a second mode and a second deviation associated with the second probability density function, a distribution coefficient of the first probability density function, and a distribution coefficient of the second probability density function, and storing in the database; storing an integral value representing an overlap of probability distributions based on the solar radiation amount estimation model as an overlap degree in the database. The invention causes a processor of a computer to perform the steps of: a management step of receiving input of solar radiation amount data; an analysis step of determining a solar radiation amount occurrence probability distribution based on the solar radiation amount data; wherein, the occurrence probability distribution of the solar radiation quantity takes dimensionless quantity as a random variable.)

1. A method for analyzing the probability distribution of occurrence of solar radiation,

causing a processor of a computer to perform the steps of:

receiving the input of solar radiation data and storing the solar radiation data in a database;

determining a solar radiation amount occurrence probability distribution based on the solar radiation amount data;

determining a solar radiation amount estimation model based on the solar radiation amount occurrence probability distribution, and a first probability density function and a second probability density function, the solar radiation amount estimation model being composed of a first mode and a first deviation associated with the first probability density function, a second mode and a second deviation associated with the second probability density function, a distribution coefficient of the first probability density function, and a distribution coefficient of the second probability density function, and storing in the database;

storing an integral value representing an overlap of probability distributions based on the solar radiation amount estimation model as an overlap degree in the database;

wherein the first probability density function and the second probability density function are Gaussian functions.

2. The method for resolving a probability distribution of occurrence of solar radiation according to claim 1,

causing a processor of a computer to perform the steps of: determining a weighted average of solar radiation amounts associated with the solar radiation amount data based on at least one of the first mode, the first deviation, the second mode, the second deviation, a distribution coefficient of the first probability density function and a distribution coefficient of the second probability density function included in the solar radiation amount estimation model, and the first probability density function and the second probability density function.

3. The method for resolving a solar radiation amount occurrence probability distribution according to claim 1 or 2,

causing a processor of a computer to perform the steps of: and performing overlapping display processing on the probability distribution of the occurrence of the solar radiation quantity and the probability distribution based on the solar radiation quantity estimation model.

4. A system for analyzing a probability distribution of occurrence of solar radiation, comprising:

a unit for accepting input of solar radiation amount data and storing in a database;

means for determining a solar radiation amount occurrence probability distribution based on the solar radiation amount data;

means for determining a solar radiation amount estimation model based on the solar radiation amount occurrence probability distribution and a first probability density function and a second probability density function and storing the solar radiation amount estimation model in the database, the solar radiation amount estimation model being composed of a first mode and a first deviation associated with the first probability density function, a second mode and a second deviation associated with the second probability density function, a distribution coefficient of the first probability density function, and a distribution coefficient of the second probability density function;

a unit for storing an integrated value representing an overlap of probability distributions based on the solar radiation amount estimation model as a degree of overlap in the database;

wherein the first probability density function and the second probability density function are Gaussian functions.

5. A program for analyzing a probability distribution of occurrence of solar radiation, wherein,

the computer is used as the following unit:

a unit for accepting input of solar radiation amount data and storing in a database;

means for determining a solar radiation amount occurrence probability distribution based on the solar radiation amount data;

means for determining a solar radiation amount estimation model based on the solar radiation amount occurrence probability distribution and a first probability density function and a second probability density function and storing the solar radiation amount estimation model in the database, the solar radiation amount estimation model being composed of a first mode and a first deviation associated with the first probability density function, a second mode and a second deviation associated with the second probability density function, a distribution coefficient of the first probability density function, and a distribution coefficient of the second probability density function;

a unit for storing an integrated value representing an overlap of probability distributions based on the solar radiation amount estimation model as a degree of overlap in the database;

wherein the first probability density function and the second probability density function are Gaussian functions.

6. A solar radiation quantity normalization statistical analysis method, wherein,

causing a processor of a computer to perform the steps of:

a management step of receiving input of solar radiation amount data;

an analysis step of extracting a reference solar radiation amount by interpolating the solar radiation amount data based on a polynomial approximation,

and determining a dimensionless quantity, which is a ratio of the amount of solar radiation to the reference amount of solar radiation,

and determining a solar radiation amount occurrence probability distribution having the dimensionless amount as a random variable based on the solar radiation amount data.

7. The normalized statistical analysis method for solar radiation amount according to claim 6, wherein,

the reference solar radiation amount is a daily or monthly average solar radiation amount.

8. The normalized statistical analysis method for solar radiation amount according to claim 6 or 7, wherein,

causing a processor of a computer to perform the following displaying steps: and displaying the probability distribution of the occurrence of the solar radiation.

9. The normalized statistical analysis method for solar radiation amount according to any one of claims 6 to 8, wherein,

the solar radiation amount occurrence probability distribution corresponds to an estimated solar radiation amount occurrence probability distribution determined by a curve fitting process based on a mixed gaussian function composed of a first gaussian function and a second gaussian function.

10. A solar radiation amount normalization statistical analysis system is provided with:

the management unit is used for receiving the input of solar radiation amount data;

an analysis unit for extracting a reference solar radiation amount by interpolating the solar radiation amount data based on polynomial approximation,

and determining a dimensionless quantity, which is a ratio of the amount of solar radiation to the reference amount of solar radiation,

and determining a solar radiation amount occurrence probability distribution having the dimensionless amount as a random variable based on the solar radiation amount data.

11. A solar radiation quantity normalization statistical analysis program, wherein,

causing a computer to function as a management unit for accepting input of solar radiation amount data and an analysis unit,

the analysis unit is configured to extract a reference solar radiation amount by interpolating the solar radiation amount data based on a polynomial approximation,

and determining a dimensionless quantity, which is a ratio of the amount of solar radiation to the reference amount of solar radiation,

and determining a solar radiation amount occurrence probability distribution having the dimensionless amount as a random variable based on the solar radiation amount data.

Technical Field

The present invention relates to a solar radiation amount occurrence probability distribution analysis method, a solar radiation amount occurrence probability distribution analysis system, a solar radiation amount occurrence probability distribution analysis program, a solar radiation amount normalization statistical analysis method, a solar radiation amount normalization statistical analysis system, and a solar radiation amount normalization statistical analysis program.

Background

Sunlight is an energy source in the earth's environment and is also one of the important biological, meteorological and engineering factors. The utilization of solar energy in the engineering field is implemented in the form of photovoltaic power generation. In photovoltaic power generation, in order to calculate the predicted power generation amount with high accuracy, establishing a solar radiation amount evaluation method is positioned as one of important issues.

The photovoltaic power generation capacity is determined by meteorological factors and electronic engineering factors. Meteorological factors here include the influence of weather on the amount of solar radiation and the influence of aerosol on the amount of solar radiation. In addition, the electronic engineering factors include power generation characteristics corresponding to a solar photovoltaic panel or the like and power conversion characteristics corresponding to an inverter device or the like.

Disclosure of Invention

The accurate prediction of the photovoltaic power generation amount is realized by comprehensively considering meteorological factors and electronic engineering factors. Specifically, when generating a direct current that exceeds the orthogonal conversion efficiency that the power generation system has, it is necessary to perform a case calculation that takes the orthogonal conversion efficiency into consideration. Therefore, the amount of solar radiation related to the power generation amount needs to be quantified in the form of a probability distribution as in patent document 1, rather than in the form of an average value per unit time.

Since the occurrence probability distribution of the amount of solar radiation is affected by meteorological factors consisting of several factors, a simplified method capable of clearly evaluating each of the several factors is desired.

The present invention has been made in view of the above circumstances, and the problems to be solved by the present invention are: the statistical processing of the solar radiation amount data is simplified, or the histogram-formed solar radiation amount occurrence probability distribution is resolved based on stochastic interpretation and the solar radiation amount is mathematically modeled as a calculation formula, the solar radiation amount data being used for calculation prediction that requires calculation with the solar radiation amount as a variable.

In order to solve the above problem, the present invention provides a method for analyzing a solar radiation amount occurrence probability distribution, the method comprising the steps of: receiving the input of solar radiation data and storing the solar radiation data in a database; determining a solar radiation amount occurrence probability distribution based on the solar radiation amount data; determining a solar radiation amount estimation model based on the solar radiation amount occurrence probability distribution and the first probability density function and the second probability density function, the solar radiation amount estimation model being composed of a first mode and a first deviation associated with the first probability density function, a second mode and a second deviation associated with the second probability density function, an allocation coefficient of the first probability density function, and an allocation coefficient of the second probability density function, and being stored in a database; storing an integrated value representing an overlap of probability distributions based on the solar radiation amount estimation model as an overlap degree in a database; wherein the first probability density function and the second probability density function are Gaussian functions.

By being set to such a configuration, the present invention can mathematically model and simplify the amount of solar radiation based on stochastic interpretation. In addition, by setting to such a configuration, the present invention has a further technical effect that it is possible to efficiently calculate the amount of solar radiation based on the mathematical model and the expected value of the calculated value having the amount of solar radiation as a variable. By setting to such a configuration, the present invention can express the statistical data of the solar radiation amount as a simple mathematical model as a mixed gaussian distribution based on a gaussian function which is a probability distribution according to the central limit theorem. In addition, by setting the arrangement as described above, the present invention has a further technical effect of enabling efficient approximation processing of a probability distribution of occurrence of solar radiation amount as a simple mathematical model which does not include a target variable representing asymmetry in a distribution curve.

In a preferred aspect of the present invention, a method for analyzing a solar radiation amount occurrence probability distribution is characterized in that a processor of a computer executes the following evaluation steps: based on at least one of a first mode, a first deviation, a second mode, a second deviation, a distribution coefficient of a first probability density function and a distribution coefficient of a second probability density function included in a solar radiation amount estimation model stored in the database, and the first probability density function and the second probability density function, a weighted average of solar radiation amounts associated with the solar radiation amount data is determined by the arithmetic device and stored in the database. By setting to such a configuration, the present invention can easily evaluate the expected value of the amount of solar radiation based on the solar radiation amount estimation model. In addition, by setting to such a configuration, the present invention has a further technical effect that the frequency of reference processing relating to statistical data of the amount of solar radiation in the database can be reduced, and the database load can be reduced.

In a preferred aspect of the present invention, a method for analyzing a solar radiation amount occurrence probability distribution is characterized in that a processor of a computer executes the following display steps: the probability distribution of occurrence of the solar radiation amount stored in the database and the probability distribution based on the solar radiation amount estimation model are subjected to an overlay display process by an output device. By setting to such a configuration, the present invention can intuitively transmit the statistical data information of the solar radiation amount to the user. In addition, by being set to such a configuration, the present invention has a further technical effect that it is possible to reduce the database load by referring to data in the form of probability density distribution without referring to all data tables relating to statistical data of the amount of solar radiation.

The present invention is a solar radiation amount occurrence probability distribution analysis system, including: a unit for accepting input of solar radiation amount data and storing in a database; means for determining a solar radiation amount occurrence probability distribution based on the solar radiation amount data; a unit for determining a solar radiation amount estimation model based on the solar radiation amount occurrence probability distribution and the first and second probability density functions, the solar radiation amount estimation model being composed of a first mode and a first deviation associated with the first probability density function, a second mode and a second deviation associated with the second probability density function, an allocation coefficient of the first probability density function, and an allocation coefficient of the second probability density function, and being stored in a database; a unit for storing an integrated value representing an overlap of probability distributions based on the solar radiation amount estimation model as an overlap degree in a database; wherein the first probability density function and the second probability density function are Gaussian functions.

The present invention is a program for analyzing a solar radiation occurrence probability distribution, characterized in that a computer is used as: a unit for accepting input of solar radiation amount data and storing in a database; means for determining a solar radiation amount occurrence probability distribution based on the solar radiation amount data; a unit for determining a solar radiation amount estimation model based on the solar radiation amount occurrence probability distribution and the first and second probability density functions, the solar radiation amount estimation model being composed of a first mode and a first deviation associated with the first probability density function, a second mode and a second deviation associated with the second probability density function, an allocation coefficient of the first probability density function, and an allocation coefficient of the second probability density function, and being stored in a database; a unit for storing an integrated value representing an overlap of probability distributions based on the solar radiation amount estimation model as an overlap degree in a database; wherein the first probability density function and the second probability density function are Gaussian functions.

In order to solve the above problem, the present invention provides a method for normalized statistical analysis of solar radiation amount, the method comprising the steps of: a management step of receiving input of solar radiation amount data; an analysis step of extracting a reference solar radiation amount by interpolating the solar radiation amount data based on polynomial approximation, and determining a dimensionless amount which is a ratio of the solar radiation amount and the reference solar radiation amount, and determining a solar radiation amount occurrence probability distribution having the dimensionless amount as a random variable based on the solar radiation amount data.

By being set to such a configuration, the present invention can estimate the amount of solar radiation based on stochastic interpretation. In addition, by being set to such a configuration, the present invention has a further technical effect that it is possible to simplify the statistical data of a large amount of solar radiation amount and to reduce the load of the database including data reference and data storage. In addition, by setting to such a configuration, the present invention can estimate the solar radiation amount in the form of a relative value including a deviation from the reference solar radiation amount.

In a preferred aspect of the present invention, the reference solar radiation amount is a daily or monthly average solar radiation amount.

In a preferred aspect of the present invention, a method for normalized statistical analysis of solar radiation amount is characterized in that a processor of a computer executes the following display steps: and displaying the probability distribution of the occurrence of the solar radiation.

In a preferred mode of the present invention, the solar radiation amount occurrence probability distribution corresponds to an estimated solar radiation amount occurrence probability distribution determined by a curve fitting process based on a mixed gaussian function composed of a first gaussian function and a second gaussian function.

The invention relates to a solar radiation quantity normalization statistical analysis system, which is characterized by comprising the following components: the management unit is used for receiving the input of solar radiation amount data; an analysis unit for extracting a reference solar radiation amount by interpolating the solar radiation amount data based on polynomial approximation, and determining a dimensionless amount which is a ratio of the solar radiation amount and the reference solar radiation amount, and determining a solar radiation amount occurrence probability distribution having the dimensionless amount as a random variable based on the solar radiation amount data.

The present invention is a solar radiation amount normalization statistical analysis program characterized by causing a computer to function as a management unit for accepting input of solar radiation amount data and an analysis unit for extracting a reference solar radiation amount by interpolating the solar radiation amount data based on polynomial approximation and determining a dimensionless amount which is a ratio of the solar radiation amount and the reference solar radiation amount and determining a solar radiation amount occurrence probability distribution which has the dimensionless amount as a random variable based on the solar radiation amount data.

Effects of the invention

The present invention can look at the occurrence probability distribution of the solar radiation amount without looking at the change of the solar radiation amount on the time axis. In addition, the present invention can calculate the deviation between the solar radiation amount and the average line of the solar radiation amount as a relative value, and can histogram the occurrence probability distribution of the solar radiation amount.

Drawings

FIG. 1 is a functional block diagram according to an embodiment of the present invention;

FIG. 2 is a diagram of a hardware configuration according to an embodiment of the present invention;

FIG. 3 is a process flow diagram according to an embodiment of the invention;

fig. 4 is a diagram showing solar radiant quantity data and an approximation example thereof according to an embodiment of the present invention;

fig. 5 is a diagram showing a display example of a solar radiation amount occurrence probability distribution according to the embodiment of the present invention.

Detailed Description

In the following, the system 1 according to the invention is described with reference to the accompanying drawings. The embodiment described below is an example of the present invention, and the present invention is not limited to the embodiment described below, and various configurations may be adopted.

In the present embodiment, the configuration, operation, and the like of the system 1 are described, but the same configuration method, program, recording medium, and the like have the same operational advantages.

The solar radiation amount normalization statistical analysis program 2001 in the present embodiment is preferably recorded in a non-transitory recording medium. The solar radiation amount normalization statistical analysis program 2001 can be introduced into the computer device 2 by using the recording medium.

The functions of the system 1 are implemented by the computer resources of the computer device 2 in an analysis application program that includes a solar radiation amount normalization statistical analysis program 2001 stored in the computer device 2. The application is implemented in the form of installation-type software or cloud-type software.

Fig. 1 shows a functional block diagram of a system 1 according to an embodiment of the present invention.

The computer device 2 in the system 1 has an input unit 21, a measurement unit 22, an analysis unit 23, an analysis unit 24, an estimation unit 25, a display unit 26 and a management unit 27.

The input unit 21 is used for input processing of the system 1. As one example, the input unit 21 is used for input processing of solar radiation amount data. In the present embodiment, the data input for processing by the input unit 21 is stored in the database DB in the computer device 2.

The input unit 21 may be configured to perform input processing via the network NW and the application interface in the plurality of computer apparatuses 2.

The measuring unit 22 is used for measuring the amount of solar radiation throughout the day. The amount of solar radiation is measured every second, every minute, every hour, every day, every month, every season, or every year. In this case, the solar radiation amount may be an instantaneous solar radiation amount at the time of measurement, may be a total solar radiation amount since the last measurement, or may be an average solar radiation amount per unit time.

The analyzing unit 23 is configured to determine a solar radiation amount occurrence probability distribution D1. The solar radiation amount occurrence probability distribution D1 is determined based on the solar radiation amount data. At this time, the solar radiation amount data may be the solar radiation amount measured by the measurement unit 22, or may be an average solar radiation amount included in statistical data of the solar radiation amount disclosed in the external service. At this time, the solar radiation amount data includes time-series data of the solar radiation amount or the average solar radiation amount.

The analysis unit 23 is preferably configured to store data relating to the solar radiation amount occurrence probability distribution D1 in the form of a data table having a dimensionless quantity representing the amount of solar radiation normalized based on the reference amount of solar radiation (standard amount of solar radiation) and a solar radiation amount occurrence probability associated with the amount of solar radiation in the database DB.

The reference solar radiation amount is preferably an average value of the solar radiation amounts of all the days included in the solar radiation amount data. The reference solar radiation amount may be configured based on statistical data of solar radiation amounts disclosed in external services.

The analysis unit 23 is preferably configured to determine solar radiation amount interpolation data R2 by polynomial approximation of the solar radiation amount data R1, and extract a daily reference solar radiation amount. The polynomial approximation is preferably configured to be performed based on a polynomial comprising a plurality of trigonometric functions. The interpolation of the solar radiation amount data in the present embodiment may be performed in the form of spline interpolation, and the method thereof is not limited.

The analysis unit 24 performs a curve fitting process on data relating to the solar radiation amount occurrence probability distribution D1 based on a gaussian mixture function (p (x)) composed of a first gaussian function and a second gaussian function represented by the following expression 1. The random variable (x) in the mixed gaussian function refers to a dimensionless quantity that is a normalized amount of solar radiation.

The analysis unit 24 performs curve fitting processing on a target variable as an unknown target variable, where the target variable includes: a first mode (μ _ a) and a first deviation (σ _ a) associated with the first gaussian function, a second mode (μ _ b) and a second deviation (σ _ b) associated with the second gaussian function, and a distribution coefficient (α) of the first gaussian function and a distribution coefficient (β) of the second gaussian function. At this time, the analysis unit 24 stores the first mode, the first deviation, and the distribution coefficient of the first gaussian function as a first target variable set, and stores the second mode, the second deviation, and the distribution coefficient of the second gaussian function as a second target variable set in the database DB. Note that the distribution coefficient and the mixing coefficient in the present embodiment are synonymous.

The resolving unit 24 preferably performs a curve fitting process based on a non-linear least squares method. Solution exploration in the curve fitting process is performed based on a generalized reduced gradient method or a simplex method.

In the present embodiment, it may be configured such that: background removal is performed in the curve fitting process by the parsing unit 24. At this time, the analysis unit 24 performs background removal on the solar radiation amount occurrence probability distribution D1 based on the straight line method or the Shirley method to extract a solar radiation amount occurrence probability distribution in the vicinity of any dimensionless amount.

The probability density function in this embodiment may be a lorentzian function or a bifurcation function. The probability density function in the present embodiment may be the solar radiation amount estimation model in the present embodiment if it is a target variable corresponding to a peak shape function that can express a continuous probability distribution.

The analysis unit 24 may be configured to perform a curve fitting process with one of a first mode (μ _ a) and a first deviation (σ _ a) associated with the first gaussian function, a second mode (μ _ b) and a second deviation (σ _ b) associated with the second gaussian function, and distribution coefficients (α) and (β) of the first and second gaussian functions as a known target variable.

The estimation unit 25 determines an estimated solar radiation amount occurrence probability distribution D2 based on at least one of the target variables in the solar radiation amount estimation model determined by the parsing unit 24 and the gaussian mixture function (p (x)). At this time, the estimation unit 25 determines a weighted average of the solar radiation amount based on the estimated solar radiation amount occurrence probability distribution D2, and stores in the database DB.

The display unit 26 is used for display processing of the system 1. As one example, the display unit 26 performs display processing based on the solar radiation amount occurrence probability distribution D1 or the estimated solar radiation amount occurrence probability distribution D2. The probability distribution in the present embodiment is displayed as a graph, a chart, or a plot. Note that the display processing may be configured to be executed cooperatively by the plurality of computer devices 2 based on an object-oriented script language such as a JavaScript (registered trademark) language.

The management unit 27 is used to manage the database DB of the system 1. The management unit 27 accepts input processing of solar radiation amount data and stores it in the database DB. The management unit 27 associates region data indicating information on a measurement place of the solar radiation amount, time data indicating information on a measurement time of the solar radiation amount, and the solar radiation amount data, the solar radiation amount occurrence probability distribution D1, the solar radiation amount estimation model, the estimated solar radiation amount occurrence probability distribution D2, and the weighted average, and updates the database DB. The database DB is constructed in the form of a relational database, a column-type database, or a key-value store. Note that, the data included in the database DB in the present embodiment may be encrypted based on a one-way function.

The managing unit 27 may be configured to store the degree of overlap based on the first probability density function and the second probability density function in the database DB. The degree of overlap refers to a value in which the overlap of the probability distribution based on the first gaussian function and the probability distribution based on the second gaussian function in the estimated solar radiation amount occurrence probability distribution D2 is represented in the form of an integrated value. The degree of overlap is preferably associated with regional data and/or time data.

The managing unit 27 may be configured to store the degrees of separation based on the first probability density function and the second probability density function in the database DB. The degree of separation is a value indicating a difference from the first mode and the second mode in the estimated solar radiation amount occurrence probability distribution D2. The degree of separation is preferably associated with regional data and/or temporal data.

Fig. 2 shows a hardware configuration diagram of the system 1 according to an embodiment of the present invention.

The computer device 2 includes: a calculation device 201, a main storage device 202, an auxiliary storage device 203, an input device 204, an output device 205, a communication device 206, a measurement device 207, and a bus interface for interconnection.

Calculation device 201 includes a processor that can execute a set of instructions. The main memory device 202 includes volatile memory, such as ram (random Access memory). In the present embodiment, an SoC (System-on-Chip) including the operation device 201 and the main storage device 202 may be included in the computer device 2. The SoC may be configured to include a coprocessor comprising an integrated circuit that is optimized for specific uses including encoding and machine learning.

The auxiliary storage device 203 serves as a database DB, which is a storage location of various data related to the system 1. The secondary storage device 203 includes a non-volatile memory. The nonvolatile memory may be a hard disk drive, which may be used in addition to a semiconductor memory such as a flash memory, and the recording method is not limited.

The auxiliary storage device 203 stores therein, in addition to the solar radiation amount normalization statistical analysis program 2001, device drivers corresponding to the operating system (OS2002), the input device 204, the output device 205, the communication device 206, or the measurement device 207.

The secondary storage device 203 may be configured to store a browser application for normalizing a web page of the statistical analysis system with the amount of solar radiation and a mail application for confirming a notification related to the system 1.

The input device 204 is used for the user to perform input processing performed by the input unit 21. The input device 204 includes a keyboard or a touch screen, and its input manner is not limited. The output device 205 is used for display processing performed by the display unit 26. The output device 205 includes a video memory for storing a frame buffer associated with the display process and a display. The driving method of the display is not limited.

The communication device 206 is configured to perform communication processing in the network NW via at least one of a wireless wan (wide Area network), a lan (local Area network), and a pan (personal Area network). The communication process is based on a wired communication standard or a wireless communication standard. As one example, Ethernet (registered trademark) is used in the wired communication standard. Wi-Fi (registered trademark) is used as the wireless communication standard, and radio waves, sound waves, or light waves may be used as the communication medium of the wireless communication standard.

The network NW in the present embodiment uses, for example, TCP/IP (Transmission Control Protocol/Internet Protocol) as a communication Protocol. The network NW may use a catv (community antenna television) line, a mobile communication network, an aviation communication network, or a satellite communication network, and the type thereof is not limited.

The measurement device 207 comprises an all-day solar watch comprising a pyroelectric element or a photoelectric element. In this embodiment, the measurement device 207 may be configured to be remotely controlled by the computer device 2 via the network NW. At this time, data on the amount of solar radiation throughout the day obtained by the measuring device 207 is transmitted via the network NW and stored in the database DB. The data may be configured to be included in solar radiation amount data.

It should be noted that the computer device 2 in this embodiment may take the form of a mobile terminal, including a smartphone and a tablet computer, in addition to the form of a workstation. The computer device 2 may be a single board type device such as a Raspberry Pi (registered trademark).

Fig. 3 shows a process flow diagram in the system 1.

As shown in fig. 3(a), in the present embodiment, first, the processed solar radiation amount data is input by the input unit 21 or the input unit 21 and the measurement unit 22, and the solar radiation amount data is received by the management unit 27 (first step S10). Next, the solar radiation amount occurrence probability distribution D1 is determined by the analysis unit 23 (second step S20). Then, the solar radiation amount estimation model is determined by the interpretation unit 24 (third step S30). Finally, the estimated solar radiation amount occurrence probability distribution D2 and the weighted average of the solar radiation amounts are determined by the estimating unit 25 (fourth step S40). It should be noted that the display unit 26 and the management unit 27 are preferably configured to sequentially exert their functions in the respective steps.

As shown in fig. 3(b), in the third step S30, solution exploration for an unknown target variable is performed based on the solar radiation amount occurrence probability distribution D1 and the first probability density function and the second probability density function (first analysis step S31). When the sum of the squares of the residuals of the probability distribution of the candidate solution obtained during the solution search and the solar radiation amount occurrence probability distribution D1 is lower than a predetermined value (yes (y) in the second analysis step S32), the candidate solution is stored in the database DB as a solar radiation amount estimation model (third analysis step S33). When the sum of squared residuals is not lower than the predetermined value (no (n) in the second parsing step S32), a transition is made to a state immediately before the second parsing step S32.

Fig. 4 shows solar radiation amount data and an approximation thereof.

As described previously, the analysis unit 23 performs polynomial interpolation based on the solar radiation amount data R1, and determines the solar radiation amount interpolation data R2. At this time, it is preferably configured to extract the daily reference solar radiation amount from the solar radiation amount interpolation data R2.

Fig. 5 shows a display example of the solar radiation amount occurrence probability distribution D1 and the estimated solar radiation amount occurrence probability distribution D2.

As described above, the estimated solar radiation amount occurrence probability distribution D2 is determined by a curve fitting process to the solar radiation amount occurrence probability distribution D1. Estimating the solar radiation occurrence probability distribution D2 comprises a linear combination of a probability distribution based on the first probability density function and a probability distribution based on the second probability density function. At this time, the first mode (μ _ a) and the first deviation (σ _ a) associated with the first probability density function (gaussian function), the second mode (μ _ b) and the second deviation (σ _ b) associated with the second probability density function (gaussian function), and the distribution coefficient (α) of the first probability density function (gaussian function) and the distribution coefficient (β) of the second probability density function (gaussian function) in fig. 5 are-50.3, 38.6, 1.1, 42.8, 21.3, and 1.0, respectively, and the distribution coefficients are appropriately doubled based on the absolute value of the number of times of occurrence of the solar radiation amount. The integrated value relating to the overlapped portion D2A in fig. 5 corresponds to the aforementioned degree of overlap.

Note that the solar radiation amount occurrence probability distribution D1 and the estimated solar radiation amount occurrence probability distribution D2 in fig. 5 have an overlap D2A in which the ratio to the entire estimated solar radiation amount occurrence probability distribution D2 is 6.1% and a standard error in which the ratio to the average value of the estimated solar radiation amount occurrence probability distribution D2 is 14.5%. In addition, the error rate based on the total expected value of the number of occurrences of the amount of solar radiation having a dimensionless amount of 1.0 or more and the total expected value of the number of occurrences of the amount of solar radiation having a dimensionless amount of 1.0 or less was 0.075%. At this time, it is estimated that the solar radiation amount occurrence probability distribution D2 does not include the divisors corresponding to the occurrence probability in fig. 5 in which the dimensionless amount is less than 0.0.

The probability distribution based on the first probability density function and the probability distribution based on the second probability density function in the present embodiment may be interpreted as the influence of different factors on the amount of solar radiation, including the influence of weather on the amount of solar radiation or the influence of aerosol on the amount of solar radiation, respectively, for example. At this time, the first target variable set, the second target variable set, and the degree of overlap are stored in the database DB as information quantifying the factor.

The probability distribution in the present embodiment may be configured to be stored in the database DB as a correspondence relation of a random variable, the number of occurrences (the number of occurrences), or the occurrence probability, and to perform display processing. At this time, of course, the occurrence probability is determined based on the total number of occurrences.

The invention causes a processor of a computer to perform the steps of: a management step of receiving input of solar radiation amount data; an analysis step of determining a solar radiation amount occurrence probability distribution based on the solar radiation amount data; wherein, the occurrence probability distribution of the solar radiation quantity takes dimensionless quantity as a random variable. Thus, the present invention can focus on the occurrence probability distribution of the solar radiation amount without focusing on the change of the solar radiation amount on the time axis. In addition, the present invention can calculate the deviation between the solar radiation amount and the average line of the solar radiation amount as a relative value, and histogram the occurrence probability distribution of the solar radiation amount.

Description of the symbols

1 System

2 computer device

21 input unit

22 measuring cell

23 analysis unit

24 analysis unit

25 estimation unit

26 display unit

27 administration unit

201 calculation device

202 primary storage device

203 auxiliary storage device

204 input device

205 output device

206 communication device

207 measuring device

2001 solar radiation amount normalization statistical analysis program

2002 OS

DB database

Probability distribution of occurrence of D1 solar radiation amount

D2 estimating the probability distribution of occurrence of solar radiation

D2A overlap

Outside of D2B

NW network

R1 solar radiation data

R2 solar radiation quantity interpolation data

S10 first step

S20 second step

S30 third step

S40 fourth step

S31 first analyzing step

S32 second analysis step

S33 third analysis step