일사량 추정을 위해 전천사진으로부터 운량을 계산하는 방법, 상기 계산한 운량을 이용한 태양광 발전량 예측 장치

The present invention refers to rolling observation device and solar electric power generation system for predicting the received signals. However, the present invention refers to without limited, the technical idea of the present invention is applicable to all multimedia system and method to be applied to device number may be filled.

The present invention refers to various modification example can apply various embodiment which may have bar, the detailed description and specific examples embodiment example drawing broadcast receiver. However, the present invention is defined with respect to a particular embodiment form which are not included, all changing range of idea and techniques of the present invention, including the water to replacement should understood to evenly.

Number 1, number 2 including various components such as ordinal number signifies a describes an can be used but the term, defined by said terms of said components are not. Components are mounted to one of said terms are used only distinguished from other components of the object. For example, number 1 number 2 outside of the range of the present invention rights without components can be termed components, similarly number 2 of the elements can be termed component number 1. That term and/or a plurality of related items or the combination of a plurality of associated substrate monolithic comprising any of the described items.

Any component and other components "connected" or folder "connected" when referred to that, or the other components connected directly or may be connected, other components may be filled intermediate present. While, any component and other components referred to as "directly connected" or folder that when "directly connected to", should be understood to does not exist in the middle of other components will.

The term used in application specific embodiment is only used to account for example, defining the present invention intending to be is endured. It is apparent that a single representation of the differently in order not providing language translators, comprising plurality of representation. In the application, the term "comprising" or "having disclosed" specification of articles feature, number, step, operation, components, component or a combination of these can hardly present designated not included, another aspect of one or more moveable, number, step, operation, component, component or a combination of these is understood to presence of or additionally pre-times those possibility should not number.

Not defined differently, scientific or technical terms so that all terms in the present invention thus is provided to the person with skill in the art will generally have the meanings etc. by same. Dictionary used for such as generally defined on the context of respective technical terms have the meanings must be consistent semantics and having interprets, the application will not become manifest in defining, or overly formal sense interpreted not ideal.

Hereinafter, the present invention preferred embodiment detailed drawing objects in reference to where for example, the damaged regions are described with reference to appended drawing is connected to the drawing code corresponding components impart the same reference number or local description dispensed the on-sensors other.

2 To 8 of the present invention number 1 is also shown in basic general outline also: 1) fish-eye lens or the like through a calculation taking into account the distortion of the picked-up in regions of the lens weight value calculates more accurate ; when a calculated based on predicting future , by correcting an error of the prediction forecast compared with; in addition said value (i.e., information) is calculated using embodiment predicting solar electric power generation are disclosed.

9 To 11 of the present invention number 2 is also shown in basic general outline also: V) based on said number 1 basic general outline; a) a calculation from solar , local characteristic as well as, seasonal properties, such variations as in the embodiment according to the installation environment and contamination of the panel to improve the accuracy of applying energy loss between the texts, a certain) measured over time and measuring a measured photovoltaic (GHI) each time history data with reference to the error (i.e., solar photovoltaic measured as to guide) having dynamic coefficient to minimize'D'and'E ' and remove an obtained via a loop (loop); the) solar to perform the estimation calculation, determining a weight calculating solar around are disclosed.

Figure 2 shows a one embodiment of the present invention also as an example, the present invention according to indicating flow calculating method are disclosed.

Figure 3 of the present invention one embodiment example, the present invention according to a deposit to calculate region are disclosed. Examples of Figure 3 embodiment, (i.e., area 1 - area n) regions in a ceiling (zenith) by concentric region is divided into a origin, each regions of different weight value by calculating the , calculate more accurate micrographs.

Also as shown in 2 and 3 also, examples of the present invention embodiment lens (e.g., fish-eye lens) therefore photograph taken by considering distortion (i.e., 3 also such as concentric regions, region 1 - n region) areas into the PCM, after setting a different weight for each region, in addition each region calculating calculates the entire are disclosed. Wherein, a fish-eye lens (or viewing angle) is 180° and angle of view, the entire sky photographing lens properties disclosed. Only, a fish-eye lens is a wide angle of view picked-up relationship object distortion occurs. I.e., the fish-eye lens therefore, negative distortion occurs is rolling one end of the picked-up photograph. As shown in fig. 3, in particular near the ceiling (zenith) carrier (descriptions pixel number 1) and, in particular near the horizon carrier the same size (width) (pixel number 2 description) corresponding to other (i.e., Image one pixel occupies scope lens distortion regardless of the number chamber is wire width of difference from object occurs). However, fish-eye lens therefore, pixel number 1 number 2 number less than the number of pixels the thread width chamber width. However, the inner edge (e.g., region n) most lens distortion, chamber number edges also distorted object's oldest photographing is set.

Thus, the present invention refers to, such corrects the distortion from the Image measuring an error occurrence, the plummet used for measuring accurate which in addition, the user edits the picked-up according to implement distortion in the degree to which regions (e.g., such as dividing the region also 3) correcting distortion of a lens can be inputted in each of the different weight calculating method (measuring) a number each cylinder regions of a substrate.

Hereinafter, with reference to the present invention according to method 2 and 3 that are directionally from calculating also also described therein.

A fish-eye lens for photographing other (S21) with rolling observation device. Said rolling region to separate it from the Image processing work combustion chamber (S22) [chwal it became sky part. The, empty sky blue color portion in the picked-up (i.e., rolling-free region) to recognize, white portion in a rolling recognizing and substrate.

The picked-up fish-eye lens therefore, the distorted Image, shown sky and rolling region of size and shape to an Image distortion also is under or over. The, such as also 3, two (i.e., 1 - n) N centers in reference concentric region divided into each other (S23). Each area is lens distortion rate like part corresponding to the other.

N in relation to the area of each of said S23 process apparatus and calculates (S24). The, calculating each region of 1 expressions such as disclosed.

[Mathematical equation 1]

C_i Of all the pixels in one pixel pitch/i=i be in rolling

Wherein, i=1, 2, .. , N

I.e., i expressions in any of said n 1 is divided regions specific recorded on the recording medium substrate.

In said S23, each region of the fish-eye lens distortion calculated through a mathematical equation 1 reflecting property of Image extracted from equipment, each regions of the ball cannot accurately measure the amount of rolling distributed there. For example, the centrally located within a Image distortion from an accident within a n 1 of Figure 3. For example, the number of pixels measured at n region (region) rolling to the common, measured at 1 region (region) that is adjacent to the one pixel when rolling to the common, region 1 and region of n measured at the seal number such as to amount of rolling it psychology. Thus, each region (i.e., region 1 - n region) weighted by applying, after inputting the expressions 2 calculates (S25).

[Equations 2]

Wherein, i=1, 2, .. , N

I.e., each areas (Ci) angles and expressions 2 is 1 weight (i.e., α_i ) After the product, detector are disclosed. I.e., each region obtained by using expressions (i.e., Ci) 1 weight each areas (i.e., α_i ) By multiplying, each areas correcting distortion of the lens (i.e., α_i * Ci) 2042 obtain. The total '1' in ' n' value except region to region are disclosed. I.e., is α (C)₁ * C₁ + Α₂ * C₂ +.. + Α_n * C_n Are disclosed.

On the other hand weight α_i Is, as can be obtained by experimental data, of Figure 3 is set to 1 at a lower distortion is the lowest is most small value, most severe distortion region 'n' region in the largest value. I.e., weight α₁ + Α₂ +.. + Α_n And=1, α₁ Α<₂ <.. Α<_n Such as properties disclosed. I.e., weight α_i Is, '1' in 'n' region such as region in range also 4 properties.

Than, 2 and 3 are also shown through method, according to the present invention refers to a more accurate value to an angle (Zenith angle) calculating can be different.

On the other hand, weight setting another embodiment example, also such as 2 and 3 also, the mathematical calculations and regions via the step M28, value to an electrical signal continuously disapproval.

Figure 5 shows a one embodiment of the present invention also example, to predict the future based on the calculated from , indicating the flow parameter prediction method are disclosed.

Examples of Figure 5 embodiment of the present invention according to predicted values based on a measure of a forecast is obtained to predicted values of bigger differs movements correcting are disclosed.

In particular, time-series regression analysis to predict future S. through examples of Figure 5 embodiment. Wherein, ' time-series regression analysis' RM, changes over time (e.g., 2 and of Figure 3 embodiment example also calculated through ) calculating the Image are disclosed. In particular, examples of Figure 5 embodiment when an excessively large value and the immediately preceding value scheme is appears to be future side being formed by, in this case forecast value products on the predicted values to particular wider, stable prediction or hypermetropia.

Hereinafter, with reference to the more detailed also 5 to each other.

A fish-eye lens for photographing other (S51) with rolling observation device. Said rolling region to separate it from the picked-up Image processing work combustion chamber (S52) sky part. The, empty sky blue color portion in the picked-up (i.e., rolling-free region) to recognize, white portion in a rolling recognizing and substrate.

Image processing tasks has been made from said server calculates (S53). The, in one example, when in calculating said S53 , action is taken of Figure 2 S23 and S24 disapproval.

And, the present invention according to preset time intervals based on said calculated rolling observation equipment predicting future S. (S54). In said S54, based on of prediction is provided, this specific area to set the time (e.g., 2 time hereinafter interval forecast) suitable way prediction are disclosed. And, now photograph aspect includes a motion vector (motion vector) reflect past rolling horizontal movement by extracting information predicting future 2000. And, according to the moving speed rolling of predictable time-flow prevention. The temperature detector, the time series analysis information obtained in a photograph such numerical prediction, senses sudden weather change to tame. The temperature, time interval between weather in a number of predicting future stations are each station 3 3 substrate. The, interrupted a number to forecast (e.g., 3 hours interval forecast) using the forecast , the correction value prediction can be in a photograph. On the other hand, Figure 6 shows a time-series regression analysis shown prediction method through surface also are disclosed. I.e., also 6 as shown in, e.g. prior to current time (t) as well as future play times (t + 1) based on the certification of measuring future change can be predicted.

And, according to the prediction values in time series regression analysis is the number through said S54 (S55) calculates error between the forecast . The, calculated error is exceeds the threshold value, the correction value of the last (S56 and S57) and determines a prediction. However, the computed error hereinafter said S55 calculated threshold value in one case, the wiring apparatus of error estimate determined (S57) to estimate noise in S54 final prediction value. Wherein, said threshold is, in one example, the computed error calculated in S55 of normal distribution can be determined. I.e., the distribution of the entire error value when the error value specified in range (e.g., 95% or more higher distribution of encrypted error) to a preset can.

Distilled, to calculate and predict through when, n estimate of error is generated when an outline vapor but, on one of Figure 5 embodiment, forecast value of error using error calculating apparatus is, predicting can be more accurate.

Figure 7 shows a one embodiment of the present invention also example, calculated from predicting solar electric power generation method using shown flow are disclosed.

The above may power generation system characteristics affecting solar electric power generation, panel temperature, which ventilate, a photovoltaic solar electric power generation core part is essentially larger than the prediction variation among and occupies substrate.

For example solar electric power generation method and device according to prediction of Figure 7 embodiment is, for example between 2 and of Figure 3 embodiment also calculated through using solar electric power generation can be embodiment to predict.

The reference also 7, of Figure 7 S71 to S74 of Figure 5 S51 to S54 is the same disclosed. Thus, in Figure 7 S71 to S74 corresponding to operation of each step of each stage of the corresponding operation corresponding description of Figure 5 description is as received signals. I.e., the abstract, imaging, Image processing treated calculates area as is a camera. And, at some future time-series regression analysis to predict other (S71 - S74) (C).

Future predicted values obtained through said steps using a photovoltaic (solar irradiance) (or, prediction information) can be estimating future (S77). I.e., using expressions 3

Measuring future solar can be.

[Equations 3]

In expressions 3 'Sky Cover' which is indicating , 2 expressions of 'C' corresponding to other.

' GHI_Clear ' Is a photovoltaic (i.e., fair weather or clear sky) of clear sky indicating other.

3 A B applications is in on position and resistance coefficient coefficient expressions each other along other optimum constant value or a variable value received signals. For example, the constant value may be Kasten F. And G. By Czeplak A=0. 75, B=3. 4 Are disclosed. The present invention is by determining the coefficient in accordance with coefficient B A fluidly value, can be more accurately predicting future solar.

On the other hand, Figure 8 shows a measured estimated solar and also through, comparing measured solar inclusion content tables. Figure 8 shows a blade shape also, the analogy blade, having the lowest priority is calculated estimated solar charge 2 change measures a timing measured solar and expressions are disclosed.

Said S77 through expressions in estimating future solar 3,

In longitude and dates and clear sky (fair weather ball) and taking into account a first computing a photovoltaic (i.e., ' GHI_Clear '), Based on the same coefficient in said S77 A B on coefficient information needed, through photovoltaic expressions 3 future estimates.

On the other hand, Image processing after said S72, rather than calculating pixel number , motion via a server (i.e., photographing area) to predict the future rolling Image calculating future disapproval (S75 and S76) process selectively performed.

Said estimating future solar based on information, power generation system using information of the surface, can be predicting solar electric power generation (S78).

One embodiment of the present invention also Figure 9 shows a example, when estimating the present invention according to from solar, photovoltaic (dynamic coefficient) measured by applying dynamic coefficient for correcting a difference between a voltage indicating flow method are disclosed.

In a mathematical equation 4 A and B in Figure 9 dynamic coefficient indicative of the other.

[Mathematical equation 4]

GHI/GHI_Clear =1 - A (C)^B

The GHI=GHI expressions 4_Clear (1 - A (C)^B ) Can be represented. 4 Expressions is differently angles 3 expression is security related. I.e., in expressions 4 'C'is indicating and, expressions of 3' Sky Cover ' such as disclosed.

Recognizing a prior art fixed coefficient applied but, this seasonal characteristics, panel contamination, according to the installation environment to reduce such variations as filtered by solar and photovoltaic energy loss measured point estimates in the last 20 years.

The, for reducing a difference between a measured solar photovoltaic of Figure 9 embodiment examples in the prior art estimates of the plummet used are disclosed.

I.e., a calculation example of Figure 9 embodiment from solar, local characteristic as well as, seasonal properties, such variations as in the embodiment according to the installation environment between energy loss and contamination of the panel applied to improve the accuracy of the micrographs. To this end, measured over time equal to a measured over time with reference to the history data (GHI) photovoltaic (i.e., solar photovoltaic measured as to guide) to minimize error having dynamic coefficient 'A' and 'B' represented Feeaback loop (loop) generates a flow of Figure 9 obtained via are disclosed.

9 Also refers to data transmissions take, 'expressions 2' using ' C'is determined substrate (S91). As above-mentioned, is 2 value to an electrical signal for correcting distortion of the lens at the expressions are disclosed. On the other hand, 'C'is' expressions 2' without using, determined by means of a conventional method may be disclosed.

A calculated by said S91 'C'and solar photovoltaic through estimation formulas (i.e., a mathematical equation 3) (hereinafter, ' GHI_Cal ' Is combined with a load) 2042 obtain (S92). The, 'expressions 3' the count 'A' and 'B' estimated is a photovoltaic (' GHI_Cal ') Photovoltaic (hereinafter, ' GHI ' is combined with a load) such as architecture guide measured and determined.

Said coefficient 'A' and 'B' transmissions take a determining method as follows. I.e., measured in said S92 ' C'on photovoltaic (GHI) measured by the recording of measured data are stored in database (or storage) (S93). Stored in said database 'C' recording with each 'C' measured using photovoltaic (GHI) respectively corresponding to measuring time, measured at specific point estimated solar photovoltaic (i.e., these calculated at S92) (so-called, 'average estimation errors' is equal to) the difference between such as architecture, said coefficient'A 'and' B ' determines a.

I.e., i.e., ' average estimation errors (i.e., GHI - GHI_Cal ) 'Is urging the minimum, ' 3 'or' expressions and expressions 4 'in'A 'and' B ' lead the coefficient (S94).

The new coefficients obtained through said S93 and S94 'A' and 'B' is a photovoltaic estimation when calculating said S92, coefficients from the received signals (S95) expressions of 3.

Than, examples of Figure 9 embodiment measured 'C'is calculated using the estimated solar (' GHI_Cal ') When derive, such as architecture (GHI) and error measured solar dynamic coefficient'A 'and' B ' can be determined. Dynamic coefficient 'A' and 'B' is determined, as well as local characteristic straight, seasonal characteristics, such variations as in the embodiment according to the installation environment of sheet panel between the results of an reflected energy loss are disclosed.

Figure 10 of the present invention one embodiment example, the present invention according to when estimating from solar, solar around a weight measuring method by applying solar are disclosed.

Figure 11 of the present invention one embodiment example, for measuring a weight value according to the photovoltaic solar around deposit region are disclosed. As shown in fig. 11, aspects of the LC is changed clouds around a missing occurs the presence aspect thereof can areas. A calculation in Figure 10, for applying weights to the solar around , region segmentation can be variously embodiment. For example, such as also 11, center, i.e. around region (i.e., region 3) and ceiling (zenith), fish-eye lens distortion in the rear region (i.e., region 2) and, solar (i.e., region 1) calculating an error amount of rolling due to a region can be divided into.

Also shown in the present invention refers to 10 and 11 also, accurately calculated for estimating a photovoltaic solar such peripheral rolling amount are disclosed. To this end, in an example of the present invention embodiment accurately to calculate, at least one divided into; each division area areas where each value to an electrical signal; a lead the entire ; and, estimating and using solar derived overall are disclosed.

In one embodiment of Figure 11 embodiment 10 and also, through a mathematical equation 5 extracted from each extracted solar around calculation can be achieved.

[Mathematical equation 5]

C=d + e (C1) (C2 + C3)

In expressions 5 'C'of Figure 11 is shown the entire , ' C1 'is of Figure 11' region 1 'of , ' C2 'is of Figure 11' region 2 'of , ' C3 'is of Figure 11' region 3' of among others.

I.e., examples of Figure 11 embodiment 10 and also, through a calculation from photovoltaic, solar peripheral area for calculating correction method controlling the rolling distributed are disclosed.

10 Also reference surface, with a fish-eye lens to 2000 (S101) rolling observation device for photographing. Said rolling region to separate it from the Image processing work combustion chamber (S102) [chwal it became sky part. The, empty sky blue color portion in the picked-up (i.e., rolling-free region) to recognize, white portion in a rolling recognizing and substrate.

Said Image processing tasks from the window substrate (S103) solar peripheral has been made. , the latitudes and longitudes, hardness and a date and time achieved by deriving position calculated using aspects of the fastening, the derived aspect using references to the positions, e.g., such as region (i.e., region 1) setting 1 also solar around 2000. In addition, 'region 1' in addition, 'region 2' on 'region 3' by dividing the switchboard.

And, each region of the calculates (S104). The, each region using expressions 2 weight (this calculating , lens for correcting distortion of the weight are disclosed) is applied so that areas to compute a disapproval.

5 Calculates a total using expressions (S105). I.e., 'region 1' weights of 'A' on 'region 1' is applied so that a region than the weighting of 'B', calculates a total from .

Overall calculated in said S105, formulas using solar photovoltaic (i.e., a mathematical equation 3) estimates.

On the other hand, solar position of '1' in 'region region 3' movement when, from the entire , 'C=d + e (C1 + C2) (C3)' calculated are disclosed. I.e., accurately aspect of a peripheral region adapted for calculating weights' d'is, ' region 3' will be applied to.

Than, by applying a weight around solar, solar peripheral rolling amount calculation can be correcting error may occur. The, around a weight value according to aspect , in estimating ensure accurate solar can be.

Hereinafter, the present invention according to solar electric power generation prediction device is described substrate.

Solar electric power generation prediction device of the present invention according to the shock and the cloud sky, photographed Image processing of Image processor, storing , storing data memory (storage or database) on measured solar and, in addition to basic form of the device such as display for displaying , for example to a processing section of Figure 11 embodiment 2 also perform (or, 'number control unit', 'algorithm performing module' or 'module' or the like can be called) is a including features.

I.e., in aspects of the present invention according to said position detection processing section photographed , calculated based on the position of aspects (e.g., in Figure 11 'region 1') and other than peripheral area of solar in dividing screen by, setting the area of the area (i.e., in a mathematical equation 5'd') weights to be applied to a peripheral aspect determines, said expressions using the weights of the determined area by performing calculating total 5 from other processing operation. In addition, said processing unit being the latitudes and longitudes, hardness, and calculate the position of the solar using the data of a temporary, calculated based on the position of aspects in said peripheral area of switchboard to switch of aspect.

In addition, such as for example said processing section of Figure 9 embodiment efficiently processing substrate. I.e., said processing section, a calculation from solar , local characteristic as well as seasonal characteristics, panel contamination, according to the installation environment for measuring such variations as in the embodiment applied to a photovoltaic energy loss between the, a photovoltaic (GHI) and stored in memory (or database) history of measured with reference to the '3' and 'expressions and expressions 4' for dynamically coefficient 'A' and 'B' derive other. The, said processing section ' average estimation errors (i.e., GHI - GHI_Cal ) 'Is urging the minimum, ' 3 'or' expressions and expressions 4 'in'A 'and' B ' lead the coefficient.

In addition said processing unit by applying a mathematical equation 3 future solar is estimated, the coefficients of the adaptive filter 3 performs operations and counting B A expressions. In addition said estimated future solar gasification and applying said processing section, predicting solar electric power generation operation has a plurality of hierarchies.

On the other hand, the present invention according to method described herein to software, hardware, or a combination of them can be implemented. For example, the present invention according to method storage medium (e.g., mobile terminal internal memory, flash memory, hard disk, and others) can be stored in, processor (e.g., mobile terminal internal microprocessor) in codes or instructions that can be executed by the software program can be implemented.

Or more, the present invention refers to drawing shown in example embodiment described with reference to but, this exemplary to avoid a and, if various deformation and equally to the examples of the present invention art therefrom person with skill in the art will understand enabling other embodiment. The, technical idea of the present invention defined by appended claim of true technology protection range generated by the will.