Solar radiation amount estimation device, solar radiation amount estimation system, method for estimating solar radiation amount, and solar radiation amount estimation program

Abstract

To improve the accuracy of estimating a solar radiation amount.SOLUTION: The present invention includes: an image acquisition unit for acquiring an empty image; a solar specification unit for specifying the position of the sun in the image; a region setting unit for setting a target region based on the position of the sun in the image, the target region being used to estimate a solar radiation amount including a diffused solar radiation amount based on clouds around the sun; a calculation unit for calculating a feature amount based on a pixel in the target region; and an estimation unit for performing estimation processing of estimating the solar radiation amount including the diffused solar radiation amount on the basis of the position of the sun and the feature amount.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a solar radiation estimation device, a solar radiation estimation system, a solar radiation estimation method, and a solar radiation estimation program. [Background technology]

[0002] Conventionally, techniques for estimating solar insolation have been developed to predict the amount of power generated in a photovoltaic power generation system. For example, Patent Document 1 (JP Patent Publication No. 2015-059821) discloses the following insolation amount estimation device. That is, the insolation amount estimation device includes a cutout unit that extracts a first cloud mass image which is an image area of ​​at least one first cloud mass included in a first image of the sky photographed from a measurement point where the amount of insolation is measured, a position identification unit that identifies the spatial position of the first cloud mass from the first cloud mass image, an influence calculation unit that calculates the influence degree of the first cloud mass on the insolation based on the insolation at the measurement point when the spatial position of the first cloud mass is an influence position which is a spatial position that influences the insolation at the measurement point, and a detection unit that detects the influence degree of the first cloud mass on the insolation based on the amount of insolation at the measurement point. The system comprises a feature extraction unit that extracts a first image feature that indicates an image feature of one cloud mass; a storage unit that stores the correspondence between the first image feature and the degree of influence for each first cloud mass at the influenced position; a movement prediction unit that predicts the movement path of a second cloud mass having a spatial position other than the influenced position and predicts the time when the second cloud mass will reach the influenced position; and a solar radiation prediction unit that predicts the degree of influence that the second cloud mass will have on the solar radiation at the measurement point at the time by referring to the correspondence and the second image feature of the second cloud mass. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2015-059821 A [Patent Document 2] JP 2007-184354 A [Patent Document 3] JP 2015-141438 A Summary of the Invention [Problem to be solved by the invention]

[0004] When clouds are present around the sun, in addition to direct sunlight, scattered light due to the clouds may occur. In this case, the amount of solar radiation becomes larger than that in a clear sky, and the accuracy of estimating the amount of solar radiation decreases. There is a need for a technology that can improve the accuracy of estimating the amount of solar radiation.

[0005] The present invention has been made to solve the above-mentioned problems, and its object is to provide a solar radiation estimation device, a solar radiation estimation system, a solar radiation estimation method, and a solar radiation estimation program that can improve the accuracy of estimating solar radiation. [Means for solving the problem]

[0006] (1) In order to solve the above problem, a solar radiation estimation device according to one aspect of the present invention includes an image acquisition unit that acquires an image of the sky, a sun identification unit that identifies a position of the sun in the image, a region setting unit that sets a target region on the image based on the position of the sun, the target region being used for estimating solar radiation including scattered solar radiation due to clouds around the sun, a calculation unit that calculates features based on pixels included in the target region, and an estimation unit that performs an estimation process to estimate the amount of solar radiation including the scattered solar radiation based on the position of the sun and the features.

[0007] In this way, in addition to the position of the sun in the image, a target area in the image for estimating the amount of solar radiation is set, and the feature amount in the target area is used to estimate the amount of solar radiation including the amount of scattered solar radiation. In the case where there are clouds around the sun, the amount of solar radiation can be estimated taking into account the effect of scattered light generated by the clouds. This improves the accuracy of estimating the amount of solar radiation.

[0008] (2) In the above (1), the solar radiation estimation device may further include a cloud identification unit that identifies a cloud area in the image that contains the clouds, and the features may include features based on the pixels that constitute the cloud area included in the target area.

[0009] With this configuration, the solar radiation estimation process can take into account the influence of scattered light depending on the brightness in the cloud areas included in the target area, i.e., the size and thickness of the clouds around the sun, etc., thereby making it possible to more accurately estimate the solar radiation when clouds are present around the sun.

[0010] (3) In the above (2), the feature may include a statistical value of brightness of a plurality of the pixels that constitute the cloud region included in the target region.

[0011] With this configuration, even if there is variation in brightness in a cloud region included in the target area, the statistical values ​​of brightness in the cloud region can be used in the process of estimating the amount of solar radiation, including diffuse solar radiation, so that the amount of solar radiation can be estimated more accurately.

[0012] (4) In any of (1) to (3) above, the solar radiation estimation device may further include a cloud identification unit that identifies a cloud area in the image that contains the clouds, and the feature may include the proportion of the cloud area that is occupied by the cloud area in the target area.

[0013] With this configuration, the solar radiation estimation process can take into account the influence of scattered light depending on the proportion of cloud areas in the target area, i.e., the amount of clouds around the sun, etc., making it possible to more accurately estimate the solar radiation when clouds are present around the sun.

[0014] (5) In any one of (1) to (4) above, the region setting unit may set an area in the image that does not include the sun as the target region.

[0015] With this configuration, when estimating the amount of solar radiation including diffuse solar radiation, it is possible to use features in areas of the image other than those areas where whiteout is likely to occur, i.e., areas with excessively high brightness, thereby making it possible to more accurately estimate the amount of solar radiation.

[0016] (6) In the above (5), the shape of the target area may be a circular ring shape centered on the position of the sun.

[0017] With this configuration, when estimating the amount of solar radiation including diffuse solar radiation, it is possible to use features in an area that does not include the sun and that is around the sun, thereby making it possible to more accurately estimate the amount of solar radiation.

[0018] (7) In any of (1) to (6) above, the sun identification unit may identify the position of the sun based on optical axis information indicating the direction of an optical axis of an imaging unit that captures the image, and orbit information indicating the orbit of the sun.

[0019] With this configuration, the position of the sun in the image can be more accurately identified according to the direction of the optical axis of the imaging unit that captures the image and the actual solar orbit.

[0020] (8) In any of (1) to (7) above, the solar radiation estimation device may further include an occlusion prediction unit that performs an occlusion prediction process to predict occlusion of the sun by the clouds based on the image, and the estimation unit may estimate the solar radiation using a predetermined calculation formula in the estimation process, and the estimation unit may select the calculation formula to be used in the estimation process depending on the prediction result by the occlusion prediction unit.

[0021] The amount of solar radiation when the sun is blocked by clouds is smaller than the amount of solar radiation on a clear day. On the other hand, when the sun is not blocked by clouds and clouds are present around the sun, the amount of solar radiation may be larger than the amount of solar radiation on a clear day. As described above, by selecting a formula to be used for estimating the amount of solar radiation according to the predicted result of blocking the sun by clouds, the amount of solar radiation can be estimated using an appropriate formula, thereby improving the accuracy of estimating the amount of solar radiation.

[0022] (9) In the above (8), the image acquisition unit may acquire a time series of images including the cloud, and the calculation unit may further create cloud movement information including at least one of the movement direction of the cloud and the movement amount of the cloud per unit time based on the time series of images, and the occlusion prediction unit may set a prediction area that is a target area for the occlusion prediction process based on the cloud movement information, and the occlusion prediction unit may predict that the sun will be occluded by the cloud in the occlusion prediction process if the cloud is present in the prediction area.

[0023] In this way, by performing shading prediction processing based on the presence or absence of clouds in a prediction area set according to the calculation results of at least one of the cloud movement direction and the cloud movement amount per unit time, it is possible to easily predict whether the sun will be shading by clouds at the time for which the solar radiation is to be estimated.

[0024] (10) In the above (9), the shape of the prediction region may be a sector having a predetermined central angle.

[0025] For example, when calculating the direction of movement of clouds, an error may occur in the calculation result. With the above configuration, the shape of the prediction area used in the occlusion prediction process can be set to a sector shape having a central angle that takes into account the error, so that the prediction accuracy in the occlusion prediction process can be prevented from decreasing.

[0026] (11) In the above (9) or (10), the area setting unit may determine a position of the target area in the image based on the cloud movement information.

[0027] With this configuration, for example, areas of clouds predicted to reach the vicinity of the sun at the time for which the amount of solar radiation is to be estimated can be included in the target area, thereby making it possible to more accurately estimate the amount of solar radiation, including diffuse solar radiation.

[0028] (12) In any of (8) to (11) above, the solar radiation estimation device may further include a cloud identification unit that identifies a cloud region in the image that contains the clouds, and the calculation unit may calculate the proportion of the cloud region in the target region and a statistical value of brightness in the cloud region included in the target region as the feature amount, and the estimation unit may use the following equation (1) as the calculation formula when the occlusion prediction unit predicts that the sun will be occupying the cloud in the occlusion prediction process. Q=(a1×F+b1×K+c1)×sin(h) ···(1) where Q is the estimated value of solar radiation, F is the ratio, K is the statistic, h is the solar elevation angle, and a1, b1, and c1 are coefficients.

[0029] With this configuration, when estimating the amount of solar radiation when there are clouds around the sun, the amount of solar radiation can be easily estimated by substituting the elevation angle of the sun and characteristic quantities corresponding to the size, thickness, amount, etc. of the clouds into a specified calculation formula.

[0030] (13) In (12) above, the estimation unit may, in the estimation process, estimate a first amount of solar radiation, which is the amount of solar radiation at a first position in the target area, and a second amount of solar radiation, which is the amount of solar radiation at a second position different from the first position, using the calculation formula, and the estimation unit may set the coefficient included in the calculation formula used to estimate the first amount of solar radiation to a value different from the coefficient included in the calculation formula used to estimate the second amount of solar radiation, depending on the distance between the first position and the position of the sun and the distance between the second position and the position of the sun.

[0031] With this configuration, when estimating the amount of solar radiation at multiple positions within an area to be estimated, it is possible to estimate the amount of solar radiation for each position that reflects the distance from the sun, thereby making it possible to more accurately estimate the amount of solar radiation at each position.

[0032] (14) In order to solve the above problem, a solar radiation estimation system according to an aspect of the present invention includes an image acquisition unit that acquires an image of the sky, a sun identification unit that identifies a position of the sun in the image, an area setting unit that sets a target area in the image based on the position of the sun, the target area being used for estimating solar radiation including scattered solar radiation due to clouds around the sun, a calculation unit that calculates features based on pixels included in the target area, and an estimation unit that performs an estimation process to estimate the amount of solar radiation including the scattered solar radiation based on the position of the sun and the features.

[0033] In this way, in addition to the position of the sun in the image, a target area in the image for estimating the amount of solar radiation is set, and the feature amount in the target area is used to estimate the amount of solar radiation including the amount of scattered solar radiation. In the case where there are clouds around the sun, the amount of solar radiation can be estimated taking into account the effect of scattered light generated by the clouds. This improves the accuracy of estimating the amount of solar radiation.

[0034] (15) In order to solve the above problem, a solar radiation estimation method according to one aspect of the present invention is a solar radiation estimation method in a solar radiation estimation device, which acquires an image of the sky, identifies a position of the sun in the image, sets a target area based on the position of the sun in the image to be used for estimating solar radiation including scattered solar radiation due to clouds around the sun, calculates features based on pixels included in the target area, and performs an estimation process to estimate the solar radiation including the scattered solar radiation based on the position of the sun and the features.

[0035] In this way, by setting a target area in the image for estimating the amount of solar radiation in addition to the position of the sun in the image, and estimating the amount of solar radiation including diffuse solar radiation using the feature amount in the target area, when there are clouds around the sun, it is possible to estimate the amount of solar radiation taking into account the effect of scattered light generated by the clouds, thereby improving the accuracy of estimating the amount of solar radiation.

[0036] (16) In order to solve the above problem, a solar radiation estimation program according to one aspect of the present invention is a solar radiation estimation program for use in a solar radiation estimation device, and causes a computer to execute the following processes: acquiring an image of the sky; identifying the position of the sun in the image; setting a target area in the image based on the position of the sun, the target area being used for estimating solar radiation including scattered solar radiation caused by clouds around the sun; calculating features based on pixels included in the target area; and estimating the amount of solar radiation including the scattered solar radiation based on the position of the sun and the features.

[0037] In this way, in addition to the position of the sun in the image, a target area in the image for estimating the amount of solar radiation is set, and the feature amount in the target area is used to estimate the amount of solar radiation including the amount of scattered solar radiation. In the case where there are clouds around the sun, the amount of solar radiation can be estimated taking into account the effect of scattered light generated by the clouds. This improves the accuracy of estimating the amount of solar radiation. Effect of the Invention

[0038] According to the present invention, it is possible to improve the accuracy of estimating the amount of solar radiation. [Brief description of the drawings]

[0039] [Figure 1] FIG. 1 is a diagram showing an example of the configuration of a solar radiation estimation system according to a first embodiment of the present invention. [Diagram 2] FIG. 2 is a diagram showing an example of time-series images acquired by the solar radiation estimation device according to the first embodiment of the present invention. [Diagram 3] FIG. 3 is a diagram for explaining the cloud identification process performed by the solar radiation estimation device according to the first embodiment of the present invention. [Figure 4] FIG. 4 is a diagram for explaining the region setting process performed by the solar radiation estimation device according to the first embodiment of the present invention. [Diagram 5] FIG. 5 is a diagram for explaining the process of calculating the cloud movement vector by the solar radiation estimation device according to the first embodiment of the present invention. [Figure 6] FIG. 6 is a diagram for explaining the shading prediction process performed by the solar radiation estimation device according to the first embodiment of the present invention. [Figure 7] FIG. 7 is a diagram showing estimated values ​​of the amount of solar radiation by a solar radiation estimation device according to a comparative example, and actual measured values ​​of the amount of solar radiation. [Figure 8] FIG. 8 is a diagram showing estimated values ​​of the amount of solar radiation by the solar radiation estimation device according to the first embodiment of the present invention, and actual measured values ​​of the amount of solar radiation. [Figure 9] FIG. 9 is a flowchart defining an operation procedure when the solar radiation estimation device according to the first embodiment of the present invention performs estimation processing. [Figure 10] FIG. 10 is a flowchart defining an operation procedure when the solar radiation estimation device according to the first embodiment of the present invention performs estimation processing. [Figure 11] FIG. 11 is a diagram for explaining the region setting process according to the first modification of the solar radiation estimation device according to the first embodiment of the present invention. [Figure 12]FIG. 12 is a diagram for explaining the estimation process according to the second modification of the solar radiation estimation device according to the first embodiment of the present invention. [Figure 13] FIG. 13 is a diagram showing an example of the configuration of a solar radiation estimation device according to the second embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0040] Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the drawings, the same or corresponding parts are denoted by the same reference characters and their description will not be repeated. In addition, at least some of the embodiments described below may be arbitrarily combined.

[0041] <First embodiment> [Solar radiation estimation system] Fig. 1 is a diagram showing an example of a configuration of a solar radiation estimation system according to a first embodiment of the present invention. Referring to Fig. 1, a solar radiation estimation system 301 includes a solar radiation estimation device 101 and a camera 201. The camera 201 is an example of an imaging unit.

[0042] The solar radiation estimation device 101 estimates solar radiation. The estimation result by the solar radiation estimation device 101 is used, for example, in a solar power generation system, for predicting the amount of power generated and for adjusting the supply and demand of power. The solar radiation estimation device 101 is, for example, a server.

[0043] The camera 201 is installed, for example, fixedly, at a predetermined location. The camera 201 captures an image of a predetermined area of ​​the sky, for example, periodically. Specifically, the camera 201 captures an image of the predetermined area of ​​the sky, for example, every 30 seconds.

[0044] The camera 201 is connected to the solar radiation estimation apparatus 101 via, for example, a wireless transmission path. Note that the camera 201 may be configured to be connected to the solar radiation estimation apparatus 101 via a wire.

[0045] [Solar radiation estimation device] The solar radiation estimation device 101 includes an image acquisition unit 11, a sun identification unit 12, a cloud identification unit 13, a region setting unit 14, a calculation unit 15, an occlusion prediction unit 16, an estimation unit 17, and a storage unit 18. The image acquisition unit 11, the sun identification unit 12, the cloud identification unit 13, the region setting unit 14, the calculation unit 15, the occlusion prediction unit 16, and the estimation unit 17 are partly or entirely realized by a processing circuit including one or more processors. The storage unit 18 is, for example, a non-volatile memory included in the processing circuit.

[0046] (Image acquisition section) The image acquiring unit 11 acquires an image G of the sky captured by the camera 201. More specifically, for example, the image acquiring unit 11 acquires images G in time series.

[0047] Specifically, for example, the camera 201 transmits image information including the captured image G and the capture time to the solar radiation estimation device 101. The camera 201 transmits the image information to the solar radiation estimation device 101, for example, periodically.

[0048] Every time the image acquisition unit 11 in the solar radiation estimation device 101 receives image information from the camera 201, the image acquisition unit 11 stores the received image information in the storage unit .

[0049] Fig. 2 is a diagram showing an example of time-series images acquired by the solar radiation estimation device according to the first embodiment of the present invention. Fig. 2 shows an image G1 that is an image G captured at an imaging time t1, and an image G2 that is an image G captured at an imaging time t2 that is later than the imaging time t1.

[0050] 2, each of the images G1 and G2 includes a sun S and a cloud C. In the example shown in FIG. 2, each of the images G1 and G2 includes clouds C1, C2, and C3 that are cloud C.

[0051] Note that camera 201 is not limited to a configuration that transmits image information to solar radiation estimation device 101, and may be configured to store image information in a recording medium such as a memory card or a USB (Universal Serial Bus) (registered trademark). In this case, image acquisition unit 11 may acquire image information from the recording medium.

[0052] Furthermore, the image acquiring unit 11 may be configured to perform a correction process for correcting the image G included in the image information received from the camera 201. Specifically, in the correction process, the image acquiring unit 11 may perform correction of distortion of the image G and correction for removing the background of the image G. In this case, the image acquiring unit 11 stores the corrected image G in the storage unit 18.

[0053] (Sun Specification Department) The sun identifying unit 12 identifies the position Ps of the sun S in the image G. More specifically, for example, the sun identifying unit 12 identifies the position Ps of the sun S based on optical axis information indicating the direction of the optical axis of the camera 201 and trajectory information indicating the trajectory of the sun S.

[0054] For example, the image acquisition unit 11 acquires a plurality of pieces of image information stored in the storage unit 18 every time a certain time has elapsed since the solar radiation estimation device 101 of the image acquisition unit 11 is started. That is, the image acquisition unit 11 acquires time-series image information including a plurality of images G and a plurality of image capture times respectively corresponding to the plurality of images G. Then, the image acquisition unit 11 outputs the acquired time-series image information to the sun identification unit 12 and the cloud identification unit 13, and deletes the time-series image information from the storage unit 18.

[0055] For example, the storage unit 18 stores orbit information. Specifically, for example, the orbit information indicates a time series change in the position Ps of the sun S in a three-dimensional orthogonal coordinate system with the x-axis in the east-west direction, the y-axis in the north-south direction, and the z-axis in the vertical direction. The optical axis information and orbit information are, for example, registered in advance in the storage unit 18 by a user. Hereinafter, the position Ps of the sun S in the three-dimensional orthogonal coordinate system indicated by the orbit information is also referred to as the "three-dimensional coordinate of the sun S." In addition, the position Ps of the sun S in the image G is also referred to as the "two-dimensional coordinate of the sun S."

[0056] When the sun identifying unit 12 receives time-series image information from the image acquiring unit 11, it confirms the three-dimensional coordinates of the sun S at the latest imaging time among the multiple imaging times included in the time-series information by referring to the orbit information in the storage unit 18. Here, it is assumed that the sun identifying unit 12 has confirmed the three-dimensional coordinates of the sun S at imaging time t2 as the three-dimensional coordinates of the sun S at the latest imaging time.

[0057] Here, the camera coordinate system is defined as follows: The camera coordinate system is expressed by Cartesian coordinates having x-axis, y-axis, and z-axis components, and is a coordinate system in which the camera 201 is located at the origin. The direction of the z-axis is the direction of the optical axis of the camera 201.

[0058] When the sun identifying unit 12 identifies the three-dimensional coordinates of the sun S at the image capturing time t2, it converts the three-dimensional coordinates into the three-dimensional coordinates of the sun S in the camera coordinate system using a predetermined rotation matrix.

[0059] For example, the storage unit 18 further stores a predetermined conversion formula for converting the three-dimensional coordinates of the sun S in the camera coordinate system into the two-dimensional coordinates of the sun S.

[0060] The sun identifying unit 12 converts the converted three-dimensional coordinates of the sun S into two-dimensional coordinates of the sun S in the image G2 captured at the imaging time t2 using a conversion formula stored in the storage unit 18. Then, the sun identifying unit 12 outputs the two-dimensional coordinates of the sun S and sun identifying information indicating the imaging time t2 of the image G2 to the area setting unit 14 and the occlusion prediction unit 16.

[0061] (Cloud identification department) For example, the cloud identifying unit 13 performs a cloud identifying process to identify an area in the image G that includes a cloud C (hereinafter, also referred to as a "cloud area CR").

[0062] More specifically, for example, when the cloud identifying unit 13 acquires time-series image information from the image acquiring unit 11, the cloud identifying unit 13 performs a cloud identifying process for each image included in the time-series image information.

[0063] Fig. 3 is a diagram for explaining the cloud identification process performed by the solar radiation estimation device according to the first embodiment of the present invention. Fig. 3 shows an example of the cloud identification process performed on an image G2 captured at an image capturing time t2.

[0064] With reference to FIG. 3, the cloud identifying unit 13 divides the image G2 into a plurality of divided images GS, and identifies, for each divided image GS, pixels P that constitute a cloud region CR from among a plurality of pixels P contained in the divided image GS.

[0065] Specifically, for example, in the cloud identification process, the cloud identifying unit 13 checks, for each pixel P, whether or not the difference between the blue luminance and the red luminance is less than a predetermined threshold value Th.

[0066] When the difference between the blue luminance and the red luminance in a pixel P is less than a threshold value Th, the cloud identifying unit 13 determines that the pixel P constitutes a cloud region CR. Then, for each piece of image information, the cloud identifying unit 13 identifies the coordinates in image G of one or more pixels P that constitute the cloud region CR (hereinafter also referred to as "cloud coordinates"). Here, it is assumed that the cloud identifying unit 13 identifies multiple cloud coordinates (hereinafter also referred to as "cloud coordinate group") for each piece of image information.

[0067] On the other hand, when the difference between the blue luminance and the red luminance in a pixel P is equal to or greater than the threshold value Th, the cloud identifying unit 13 determines that the pixel P does not constitute a cloud region CR.

[0068] When the cloud identification unit 13 completes the cloud identification process, it outputs to the calculation unit 15 cloud identification information including time-series image information from the image acquisition unit 11 and multiple cloud coordinate groups each corresponding to multiple images G contained in the time-series image information.

[0069] (Area setting part) The area setting unit 14 performs area setting processing to set a target area T based on the position of the sun S in the image G. The target area T is an area used for estimating the amount of solar radiation including the amount of scattered solar radiation caused by clouds C present around the sun S. The target area T is an area based on the position of the sun S, for example.

[0070] More specifically, for example, in the region setting process, the region setting unit 14 sets a region in the image G that does not include the sun S as the target region T. For example, the shape of the target region T is a circular ring shape centered on the position of the sun S.

[0071] Specifically, for example, the storage unit 18 stores annular information indicating an inner diameter R1 and an outer diameter R2 of the target region T. The annular information is registered in the storage unit 18 in advance by the user, for example.

[0072] When the area setting unit 14 receives the sun specifying information from the sun specifying unit 12, it acquires from the storage unit 18 the image G2 corresponding to the imaging time t2 indicated by the sun specifying information, and the annular information.

[0073] FIG. 4 is a diagram for explaining the region setting process performed by the solar radiation estimation device according to the first embodiment of the present invention.

[0074] Referring to Figure 4, when the area setting unit 14 acquires an image G2 from the memory unit 18, it sets a target area T in the image G2, centered on the two-dimensional coordinates of the sun S indicated by the sun identification information from the sun identification unit 12, and with inner and outer diameters equal to the inner diameter R1 and outer diameter R2 indicated by the annular information, respectively.

[0075] When the region setting unit 14 completes the region setting process, it outputs to the calculation unit 15 region information including an image G2 (hereinafter also referred to as the “estimation target image”) including the target region T and the imaging time t2 of the estimation target image.

[0076] The shape of the target region T is not limited to a circular ring shape centered on the position Ps of the sun S, but may be a rectangle centered on the position Ps, a star centered on the position Ps, or the like.

[0077] (Calculation section) <Calculation of feature values> The calculation unit 15 performs a feature amount calculation process to calculate feature amounts based on pixels included in the target area T set by the area setting unit 14. For example, the calculation unit 15 performs the feature amount calculation process when the target area T satisfies a predetermined condition. The predetermined condition is, for example, that the target area T set by the area setting unit 14 includes at least a part of the cloud area CR identified by the cloud identification unit 13.

[0078] More specifically, for example, when the calculation unit 15 receives region information from the region setting unit 14 and cloud identification information from the cloud identification unit 13, it checks the coordinates of each of the multiple pixels P that constitute the target region T (hereinafter also referred to as "multiple coordinates Ct") in the estimated target image included in the region information.

[0079] When the calculation unit 15 checks the multiple coordinates Ct, it identifies a cloud coordinate group from among the multiple cloud coordinate groups contained in the cloud identification information from the cloud identification unit 13 that corresponds to image information that includes the same imaging time as the imaging time t2 contained in the area information from the area setting unit 14.

[0080] Then, the calculation unit 15 checks whether or not at least one of the checked coordinates Ct is included in the identified cloud coordinate group.

[0081] If all of the confirmed coordinates Ct are not included in the identified cloud coordinate group, the calculation unit 15 does not perform the feature amount calculation process.

[0082] On the other hand, when at least one of the coordinates Ct is included in the identified cloud coordinate group, the calculation unit 15 determines that the target region T includes a cloud region CR and performs feature amount calculation processing. Hereinafter, the cloud region CR included in the target region T is also referred to as a "cloud region CR1."

[0083] For example, in the feature amount calculation process, the calculation unit 15 calculates a luminance statistical value K of the cloud region CR1 included in the target region T and a proportion F of the target region T that the cloud region CR1 occupies, as the feature amount.

[0084] Specifically, for example, when the calculation unit 15 determines that the target region T includes the cloud region CR1, the calculation unit 15 calculates an average value of the luminance of a plurality of pixels P constituting the cloud region CR1 as the statistical value K. Note that the calculation unit 15 may calculate a representative value of the luminance of the plurality of pixels P as the statistical value K.

[0085] Furthermore, for example, when the calculation unit 15 determines that the target region T includes the cloud region CR1, it calculates the area B1 of the target region T using the inner radius R1 and the outer radius R2 indicated by the annular information stored in the storage unit 18. Furthermore, when the calculation unit 15 determines that the target region T includes the cloud region CR1, it counts the number of pixels P that make up the cloud region CR1.

[0086] The storage unit 18 further stores the area of ​​one pixel P. When the calculation unit 15 counts the number of the multiple pixels P that make up the cloud region CR1, the calculation unit 15 calculates a value obtained by multiplying the area of ​​one pixel P stored in the storage unit 18 by the count value as the area B2 of the cloud region CR1. Then, the calculation unit 15 calculates the ratio of the area B2 to the area B1 as the ratio F.

[0087] After calculating the statistical value K and the proportion F, the calculation unit 15 outputs feature amount information indicating the calculation results to the estimation unit 17.

[0088] Note that the calculation unit 15 is not limited to a configuration in which it calculates both the statistical value K and the ratio F as the feature amount in the feature amount calculation process, but may be a configuration in which it calculates either one of the statistical value K or the ratio F as the feature amount.

[0089] <Calculation of cloud movement information> For example, the calculation unit 15 further creates cloud movement information including the movement direction of cloud C and the movement amount of cloud C per unit time (hereinafter also referred to as ``cloud movement amount L'') based on the time series images G acquired by the image acquisition unit 11.

[0090] More specifically, for example, when the calculation unit 15 receives region information from the region setting unit 14 and cloud identification information from the cloud identification unit 13, it identifies image information GA and image information GB from the time-series image information contained in the cloud identification information, which correspond to the imaging time t2 included in the region information and the imaging time t1 immediately preceding the imaging time t2, respectively.

[0091] Then, the calculation unit 15 calculates the cloud movement amount L using the image G1 captured at the imaging time t1, which is included in the image information GA, and the image G2 captured at the imaging time t2, which is included in the image information GB.

[0092] The calculation unit 15 is not limited to a configuration that calculates both the moving direction of the cloud C and the cloud movement amount L based on the time-series images G, but may be configured to calculate either one of the moving direction of the cloud C or the cloud movement amount L.

[0093] FIG. 5 is a diagram for explaining the process of calculating the cloud movement vector by the solar radiation estimation device according to the first embodiment of the present invention.

[0094] 2 and 5, for example, the calculation unit 15 divides the image G2 into a plurality of divided images GS. Then, the calculation unit 15 uses the image G1 and the image G2 to calculate a movement vector V of the cloud C for each divided image GS, for example, by optical flow.

[0095] Then, when the calculation unit 15 calculates the multiple movement vectors V, the calculation unit 15 calculates a statistical value of the magnitudes of the multiple movement vectors V as the cloud movement amount L. For example, the calculation unit 15 calculates a representative value of the magnitudes of the multiple movement vectors V as the cloud movement amount L. Note that the calculation unit 15 may calculate an average value of the magnitudes of the multiple movement vectors V as the cloud movement amount L.

[0096] Furthermore, when the calculation unit 15 calculates the multiple movement vectors V, the calculation unit 15 specifies the direction of one of the multiple movement vectors V as the movement direction of the cloud C.

[0097] The calculation unit 15 calculates the cloud movement amount L and, upon identifying the movement direction of the cloud C, outputs the calculated cloud movement amount L and cloud movement information indicating the movement direction to the occlusion prediction unit 16.

[0098] (Occlusal prediction section) For example, the shading prediction unit 16 performs shading prediction processing to predict shading of the sun S by a cloud C based on the image G acquired by the image acquisition unit 11 .

[0099] More specifically, for example, the occlusion prediction unit 16 sets a prediction area A, which is the area to be targeted for the occlusion prediction processing, based on the position of the sun S identified by the sun identification unit 12 and the cloud movement amount L calculated by the calculation unit 15.

[0100] FIG. 6 is a diagram for explaining the shading prediction process performed by the solar radiation estimation device according to the first embodiment of the present invention.

[0101] 6, for example, the occlusion prediction unit 16 sets a part of a circular region centered on the position of the sun S in the image G as a prediction region A based on the position Ps of the sun S identified by the sun identification unit 12, and the cloud movement amount L and the movement direction of the cloud C calculated by the calculation unit 15. For example, the shape of the prediction region A is a sector having a predetermined central angle θ.

[0102] Specifically, for example, the storage unit 18 stores central angle information indicating the central angle θ of the prediction area A. The central angle information is registered in the storage unit 18 in advance by the user, for example.

[0103] When the occlusion prediction unit 16 receives sun identification information from the sun identification unit 12 and cloud movement information from the calculation unit 15, it acquires from the memory unit 18 an image G2 corresponding to the imaging time t2 indicated by the sun identification information and central angle information.

[0104] Then, based on the cloud movement amount L and the movement direction of the cloud C indicated by the cloud movement information from the calculation unit 15, the occlusion prediction unit 16 sets, as a prediction area A, an area in the image G2 that overlaps with the position of the sun S at a target time tr for estimating the amount of solar radiation (hereinafter also referred to as the "target estimation time") and has a central angle θ indicated by the central angle information acquired from the storage unit 18. Note that the shape of the prediction area A may be any shape other than a sector.

[0105] The estimation target time tr is, for example, a time that is a predetermined time after the imaging time t2 indicated by the sun identification information from the sun identification unit 12. Specifically, for example, the estimation target time tr is a time that is one minute after the imaging time t2.

[0106] After setting the prediction area A in the image G2, the occlusion prediction unit 16 checks whether or not a cloud C exists in the prediction area A.

[0107] When a cloud C is present in the prediction area A, the occlusion prediction unit 16 predicts in the occlusion prediction process that the sun S will be occluded by the cloud C. Then, the occlusion prediction unit 16 outputs prediction result information indicating that the sun S will be occluded by the cloud C to the estimation unit 17.

[0108] On the other hand, when the cloud C does not exist in the prediction area A, the occlusion prediction unit 16 predicts in the occlusion prediction process that the sun S will not be occluded by the cloud C. Then, the occlusion prediction unit 16 outputs prediction result information indicating that the sun S will not be occluded by the cloud C to the estimation unit 17.

[0109] (Estimation Department) The estimation unit 17 performs an estimation process to estimate the amount of solar radiation including the amount of scattered solar radiation caused by clouds C present around the sun S, based on the position of the sun S identified by the sun identification unit 12 and the feature amount calculated by the calculation unit 15.

[0110] More specifically, for example, in the estimation process, the estimation unit 17 estimates the amount of solar radiation using a predetermined calculation formula.

[0111] For example, the estimation unit 17 selects a calculation formula to be used in the estimation process depending on the prediction result by the occlusion prediction unit 16.

[0112] Specifically, for example, the storage unit 18 stores the following formulas (1) and (2), which are calculation formulas. Q=(a1×F+b1×K+c1)×sin(h) ···(1) R=(a2×F+b2×K+c2)×sin(h) ···(2)

[0113] In formula (1), Q is an estimated value of the amount of solar radiation when the sun S is blocked by a cloud C, h is the elevation angle of the sun S, and a1, b1, and c1 are coefficients. In formula (2), R is an estimated value of the amount of solar radiation when the sun S is not blocked by a cloud C, and a2, b2, and c2 are coefficients. The coefficients a1, b1, and c1 are different values ​​from the coefficients a2, b2, and c2, respectively. Formulas (1) and (2) are, for example, registered in advance in the storage unit 18 by a user.

[0114] For example, when the occlusion prediction unit 16 predicts that the sun S will be occluded by a cloud C in the occlusion prediction process, the estimation unit 17 uses the following formula (1).

[0115] Specifically, for example, the storage unit 18 stores the elevation angle h of the sun S at the estimation target time tr.

[0116] When the estimation unit 17 receives prediction result information indicating that the sun S will be blocked by the cloud C from the shading prediction unit 16 and also receives feature information from the calculation unit 15, it reads out equation (1) and the elevation angle h of the sun S at the estimation target time tr from the memory unit 18.

[0117] Then, the estimation unit 17 calculates an estimated value Q of the amount of solar radiation by substituting the ratio F and the statistical value K indicated by the feature amount information from the calculation unit 15, and the elevation angle h of the sun S at the estimation target time tr into equation (1).

[0118] Furthermore, for example, when the occlusion prediction unit 16 predicts that the sun S will not be occluded by the cloud C in the occlusion prediction process, the estimation unit 17 uses the following formula (2).

[0119] Specifically, when the estimation unit 17 receives prediction result information indicating that the sun S is not blocked by the cloud C from the occlusion prediction unit 16 and also receives feature information from the calculation unit 15, it reads out equation (2) and the elevation angle h of the sun S at the estimation target time tr from the memory unit 18.

[0120] Then, the estimation unit 17 calculates an estimated value R of the amount of solar radiation by substituting the ratio F and the statistical value K indicated by the feature amount information from the calculation unit 15, and the elevation angle h of the sun S at the estimation target time tr into equation (2).

[0121] Fig. 7 is a diagram showing an estimated value of solar radiation by a solar radiation estimation device according to a comparative example, and an actual measured value of solar radiation. In Fig. 7, the horizontal axis represents time, and the vertical axis represents solar radiation. Graph N1 shows a time series change in an actual measured value U1 of solar radiation in a time period TA from 12:30 to 14:30 on a certain date. Graph N2 shows a time series change in an estimated value U2 of solar radiation in a time period TA by a solar radiation estimation device according to a comparative example. In the process of estimating solar radiation, the solar radiation estimation device according to the comparative example does not use a feature amount in a target area T, but uses a position Ps of the sun S in an estimation target image.

[0122] 7, the actual measured value U1 at 12:45 shown in graph N1 is 1155 kilowatts per square meter, and the estimated value U2 at 12:45 shown in graph N2 is 895 kilowatts per square meter. In this case, the difference between the actual measured value U1 and the estimated value U2 is 260 kilowatts per square meter.

[0123] Fig. 8 is a diagram showing the estimated value of the amount of solar radiation by the solar radiation estimation device according to the first embodiment of the present invention and the actual measured value of the amount of solar radiation. In Fig. 8, the horizontal axis represents time and the vertical axis represents the amount of solar radiation. Graph N12 shows the time series change of the estimated value U12 of the amount of solar radiation by the solar radiation estimation device 101 in the same time period as the time period TA shown in Fig. 7.

[0124] 8, the actual measured value U1 at 12:45 shown in graph N1 is 1155 kilowatts per square meter, as in FIG. 7. The estimated value U12 at 12:45 shown in graph N12 is 1048 kilowatts per square meter. In this case, the difference between the actual measured value U1 and the estimated value U12 is 107 kilowatts per square meter. Thus, the difference between the actual measured value U1 and the estimated value U12 is smaller than the difference between the actual measured value U1 and the estimated value U2. That is, in the solar radiation estimation device 101, the above-mentioned configuration can improve the estimation accuracy of the solar radiation.

[0125] Referring again to FIG. 2, for example, when the estimation unit 17 completes the estimation process, it transmits estimation result information indicating the estimation result to a prediction device that predicts the amount of power generation in the photovoltaic power generation system.

[0126] In a solar power generation system, when a prediction device receives estimation result information from a solar radiation estimation device 101, the prediction device predicts the amount of power generation at an estimation target time tr based on the estimation result indicated by the estimation result information.

[0127] The estimation unit 17 may transmit estimation result information indicating the ratio between the estimated value of the amount of solar radiation and the amount of solar radiation in a clear sky to a prediction device in the solar power generation system.

[0128] [Operation flow] The solar radiation estimation device 101 according to the embodiment of the present invention includes a computer including a memory, and a processor such as a CPU in the computer reads out from the memory a program including some or all of the steps in the following flowcharts and executes it. The program for this device can be installed from outside. The program for this device is distributed in a state stored on a recording medium or via a communication line.

[0129] 9 and 10 are flow charts defining an operation procedure when the solar radiation estimation device according to the first embodiment of the present invention performs estimation processing.

[0130] 9 and 10, first, the solar radiation estimation device 101 receives image information from the camera 201 and stores it in the storage unit 18 (step S101).

[0131] The solar radiation estimation device 101 stores new image information received from the camera 201 in the storage unit 18 (step S101) until a certain time, for example, 10 minutes, has elapsed since the solar radiation estimation device 101 was started (NO in step S102).

[0132] Then, when a certain time has elapsed since the solar radiation estimation device 101 was started (YES in step S102), the solar radiation estimation device 101 acquires the time-series image information accumulated in the storage unit 18 (step S103).

[0133] Next, the solar radiation estimation device 101 uses the acquired time-series image information, as well as the optical axis information and trajectory information stored in the memory unit 18, to identify the position Ps of the sun S in the image G captured at the most recent imaging time (step S104).

[0134] Next, the solar radiation estimation device 101 performs a cloud identification process to identify a cloud region CR for each image G included in the acquired time-series image information (step S105). Note that steps S104 and S105 may be executed in reverse order or in parallel.

[0135] Next, the solar radiation estimation device 101 sets a target area T in the image G captured at the latest imaging time based on the position of the sun S. As described above, the target area T is an area used for estimating the amount of solar radiation including the amount of scattered solar radiation caused by clouds C present around the sun S (step S106).

[0136] Next, the solar radiation estimation device 101 calculates feature amounts in the target region T. For example, as described above, the solar radiation estimation device 101 calculates, as feature amounts, a luminance statistical value K in the cloud region CR included in the target region T and a proportion F of the cloud region CR in the target region T (step S107).

[0137] Next, the solar radiation estimation device 101 creates cloud movement information indicating the cloud movement amount L and the movement direction of the cloud C based on the acquired time-series images G (step S108).

[0138] Next, the solar radiation estimation device 101 sets a prediction area A for the shading prediction process based on the specified position Ps of the sun S and the created cloud movement information. For example, as described above, the solar radiation estimation device 101 sets a part of a circular area centered on the position of the sun S as the prediction area A (step S109).

[0139] Next, the solar radiation estimation device 101 performs a shading prediction process for predicting shading of the sun S by the cloud C at the estimation target time tr, using the set prediction region A (step S110).

[0140] When the solar radiation estimation device 101 predicts that the sun S will be blocked by the cloud C (YES in step S111), it selects the formula (1) stored in the storage unit 18 as the calculation formula to be used in the estimation process (step S112).

[0141] On the other hand, if the solar radiation estimation device 101 predicts that the sun S will not be blocked by the cloud C (NO in step S111), it selects formula (2) stored in the memory unit 18 as the calculation formula to be used in the estimation process (step S113).

[0142] Next, the solar radiation estimation device 101 performs estimation processing using the selected calculation formula (step S114).

[0143] Next, the solar radiation estimation device 101 transmits estimation result information indicating the estimation result of the estimation process to a prediction device that predicts the amount of power generation in the photovoltaic power generation system (step S115).

[0144] Next, if a certain time, for example, 10 minutes, has not elapsed since the previous acquisition of time-series image information (NO in step S102), the solar radiation estimation device 101 stores new image information received from the camera 201 in the memory unit 18 (step S101).

[0145] On the other hand, if a certain time has passed since the solar radiation estimation device 101 last acquired time-series image information from the memory unit 18 (YES in step S102), the solar radiation estimation device 101 acquires time-series image information newly stored in the memory unit 18 (step S103).

[0146] [Variation 1] The solar radiation estimation device 101 may be configured to perform the following process as the region setting process.

[0147] 11 is a diagram for explaining the region setting process according to the first modification of the solar radiation estimation device according to the first embodiment of the present invention. In FIG. 11, an arrow indicates the moving direction of a cloud C.

[0148] 2 and 11, in the first modification, the region setting unit 14 determines the position of the target region T in the image G based on the moving direction of the cloud C.

[0149] More specifically, for example, when the calculation unit 15 creates the cloud movement information, the calculation unit 15 outputs the created cloud movement information to the region setting unit 14 and the occlusion prediction unit 16 .

[0150] When the area setting unit 14 receives sun identification information from the sun identification unit 12 and cloud movement information from the calculation unit 15, it acquires from the memory unit 18 an image G2 corresponding to the imaging time t2 indicated by the sun identification information, and annular information.

[0151] Then, the area setting unit 14 checks the two-dimensional coordinates of the sun S indicated by the sun identification information from the sun identification unit 12, i.e., the position Ps of the sun S, in the image G2, and sets the center position Pt of the target area T to a position shifted by the cloud movement amount L indicated by the cloud movement information from the calculation unit 15, in the opposite direction to the movement direction of the cloud C indicated by the cloud movement information from the calculation unit 15, relative to the position Ps of the sun S.

[0152] After setting the center position Pt of the target region T, the region setting unit 14 sets the inner and outer diameters of the target region T to the inner diameter R1 and outer diameter R2 indicated by the annular information acquired from the storage unit 18, respectively.

[0153] [Variation 2] The solar radiation estimation device 101 may be configured to perform the following process in the estimation process.

[0154] FIG. 12 is a diagram for explaining the estimation process according to the second modification of the solar radiation estimation device according to the first embodiment of the present invention.

[0155] 12, in the second modification, the estimation unit 17 estimates the amount of solar radiation at a position Pb1 in the target area T (hereinafter also referred to as "amount of solar radiation Q1") and the amount of solar radiation at a position Pb2 different from the position Pb1 in the target area T (hereinafter also referred to as "amount of solar radiation Q2") using the above formula (1) or (2). The position Pb1 is an example of a first position, and the position Pb2 is an example of a second position. The amount of solar radiation Q1 is an example of a first amount of solar radiation, and the amount of solar radiation Q2 is an example of a second amount of solar radiation in the estimation process.

[0156] Specifically, for example, when the occlusion prediction unit 16 completes the occlusion prediction process, it outputs to the estimation unit 17 prediction result information indicating the prediction result and the two-dimensional coordinates of the sun S indicated by the sun identification information from the sun identification unit 12.

[0157] For example, the storage unit 18 stores a coordinate list indicating the coordinates of the position Pb1 and the coordinates of the position Pb2 for each two-dimensional coordinate of the sun S.

[0158] When the estimation unit 17 receives prediction result information from the occlusion prediction unit 16, it refers to the coordinate list in the memory unit 18 to confirm the coordinates of position Pb1 and the coordinates of position Pb2 that correspond to the two-dimensional coordinates of the sun S indicated by the prediction result information.

[0159] Then, the estimation unit 17 calculates a distance D1 between the coordinates of the position Pb1 and the two-dimensional coordinates of the sun S, and a distance D2 between the coordinates of the position Pb2 and the two-dimensional coordinates of the sun S.

[0160] When the prediction result information from the shading prediction unit 16 indicates that the sun S is shaded by the cloud C, the estimation unit 17 reads out the formula (1) stored in the storage unit 18. Then, for example, the estimation unit 17 sets the coefficient b1 included in the formula (1) for predicting the amount of solar radiation Q1 to a value different from the coefficient b1 included in the formula (1) for predicting the amount of solar radiation Q2, according to the calculated distances D1 and D2.

[0161] Specifically, for example, when the distance D1 is smaller than the distance D2, the estimation unit 17 sets the coefficient b1 included in the equation (1) for predicting the amount of solar radiation Q1 to a value larger than the coefficient b1 included in the equation (1) for predicting the amount of solar radiation Q2.

[0162] Furthermore, for example, when the prediction result information from the shading prediction unit 16 indicates that the sun S is not shaded by the cloud C, the estimation unit 17 reads out the formula (2) stored in the storage unit 18. Then, for example, the estimation unit 17 sets the coefficient b2 included in the formula (2) for predicting the amount of solar radiation Q1 to a value different from the coefficient b2 included in the formula (2) for predicting the amount of solar radiation Q2, according to the calculated distances D1 and D2.

[0163] Specifically, for example, when the distance D1 is smaller than the distance D2, the estimation unit 17 sets the coefficient b1 included in the equation (2) for predicting the amount of solar radiation Q1 to a value larger than the coefficient b1 included in the equation (2) for predicting the amount of solar radiation Q2.

[0164] Next, another embodiment of the present invention will be described with reference to the drawings. In the drawings, the same or corresponding parts are designated by the same reference characters and their description will not be repeated.

[0165] <Second embodiment> In the above-described first embodiment, the solar radiation estimation device 101 performs estimation processing by identifying the position Ps of the sun S and the cloud region CR in the image G. In contrast, in the second embodiment of the present invention, the solar radiation estimation device 101A performs estimation processing by identifying the position of the sun S in the image G without identifying the cloud region CR in the image G. Contents other than those described below are the same as those of the solar radiation estimation device 101 according to the first embodiment.

[0166] 13 is a diagram showing an example of a configuration of a solar radiation estimation device according to the second embodiment of the present invention. Referring to FIG. 13, a solar radiation estimation system 301 includes a solar radiation estimation device 101A and a camera 201.

[0167] The solar radiation estimation device 101A includes an image acquisition unit 11, a sun identification unit 12, a region setting unit 14, a calculation unit 15A, an estimation unit 17A, and a storage unit 18. The image acquisition unit 11, the sun identification unit 12, the region setting unit 14, the calculation unit 15A, and the estimation unit 17A are partly or entirely realized by a processing circuit including one or more processors. The storage unit 18 is, for example, a non-volatile memory included in the processing circuit.

[0168] (Calculation section) The calculation unit 15A performs a feature amount calculation process to calculate a feature amount in the target region T. More specifically, for example, when the calculation unit 15A receives region information from the region setting unit 14, the calculation unit 15A calculates, as a statistical value K2, an average value of luminance of a plurality of pixels constituting the target region T in the estimation target image included in the region information.

[0169] Then, the calculation unit 15A outputs feature amount information indicating the calculation result to the estimation unit 17A. Note that the calculation unit 15A may calculate a representative value of the luminance of a plurality of pixels that form the target region T as the statistical value K2.

[0170] (Estimation Department) The estimation unit 17A performs an estimation process to estimate the amount of solar radiation based on the position Ps of the sun S identified by the sun identifying unit 12 and the statistical value K2 calculated by the calculation unit 15A.

[0171] More specifically, for example, the storage unit 18 stores the following formula (3), which is a calculation formula. In formula (3), Q12 is an estimated value of the amount of solar radiation. Formula (3) is, for example, registered in advance in the storage unit 18 by a user.

[0172] Q12=(b1×K2+c1)×sin(h) ···(3)

[0173] When the estimation unit 17A receives the feature amount information from the calculation unit 15A, the estimation unit 17A reads out from the storage unit 18 equation (3) and the elevation angle h of the sun S at the estimation target time tr.

[0174] Then, the estimation unit 17A calculates an estimated value Q12 of the amount of solar radiation by substituting the statistical value K2 indicated by the feature amount information from the calculation unit 15A and the elevation angle h of the sun S at the estimation target time tr into equation (3). Then, the estimation unit 17A transmits estimation result information indicating the estimation result to the prediction device in the solar power generation system.

[0175] Note that some or all of the functions of the solar radiation estimation apparatus 101, 101A according to the embodiment of the present invention may be provided by cloud computing. That is, the solar radiation estimation apparatus 101, 101A according to the embodiment of the present invention may be a cloud server composed of a plurality of servers.

[0176] The above-described embodiments should be considered as illustrative and not restrictive in all respects. The scope of the present invention is defined by the claims, not by the above description, and is intended to include all modifications within the meaning and scope of the claims. [Explanation of symbols]

[0177] 11 Image acquisition section 12 Solar Specification Department 13 Cloud identification part 14 Area setting section 15,15A calculation part 16. Obstruction prediction section 17,17A Estimation part 18 Memory section 101,101A solar radiation estimation device 201 Camera 301 Solar Radiation Estimation System A Prediction Region C,C1,C2,C3 cloud CR,CR1 Cloud region S sun T Target area

Claims

1. an image acquisition unit for acquiring an image of the sky; a sun identification unit that identifies a position of the sun in the image; a region setting unit that sets, on the image, a target region based on the position of the sun, the target region being used for estimating the amount of solar radiation including the amount of scattered solar radiation due to clouds present around the sun; A calculation unit that calculates a feature amount based on pixels included in the target region; an estimation unit that performs an estimation process to estimate the amount of solar radiation including the amount of diffuse solar radiation based on the position of the sun and the feature amount; A solar radiation estimation device comprising:

2. The solar radiation estimation device further comprises: A cloud identification unit is provided to identify a cloud region in the image, the cloud region being a region including the cloud, The solar radiation estimation device according to claim 1 , wherein the feature amount includes a feature amount based on the pixels that constitute the cloud region included in the target region.

3. The solar radiation estimation device according to claim 2 , wherein the feature amount includes a statistical value of luminance of a plurality of the pixels constituting the cloud region included in the target region.

4. The solar radiation estimation device further comprises: A cloud identification unit is provided to identify a cloud region in the image, the cloud region being a region including the cloud, The solar radiation estimation device according to claim 1 , wherein the feature amount includes a proportion of the target region that is occupied by the cloud region.

5. The solar radiation estimation device according to claim 1 , wherein the region setting unit sets an area in the image that does not include the sun as the target region.

6. The solar radiation estimation device according to claim 5 , wherein the shape of the target area is an annular shape centered on the position of the sun.

7. 3. The solar radiation estimation device according to claim 1, wherein the sun identification unit identifies the position of the sun based on optical axis information indicating a direction of an optical axis of an imaging unit that captures the image and trajectory information indicating an trajectory of the sun.

8. The solar radiation estimation device further comprises: a shading prediction unit that performs a shading prediction process to predict shading of the sun by the cloud based on the image, The estimation unit estimates the amount of solar radiation by using a predetermined calculation formula in the estimation process; The solar radiation estimation device according to claim 1 , wherein the estimation unit selects the calculation formula to be used in the estimation process in accordance with a prediction result by the occlusion prediction unit.

9. The image acquisition unit acquires the images in a time series including the cloud, The calculation unit further creates cloud movement information including at least one of a movement direction of the cloud and a movement amount of the cloud per unit time based on the time series images, The occlusion prediction unit sets a prediction area that is a target area of ​​the occlusion prediction process based on the cloud movement information, The solar radiation estimation device according to claim 8 , wherein the shading prediction unit predicts that the sun will be shaded by the cloud in the shading prediction process when the cloud is present in the prediction region.

10. The solar radiation estimation device according to claim 9 , wherein the shape of the prediction region is a sector having a predetermined central angle.

11. The solar radiation estimation device according to claim 9 , wherein the region setting unit determines a position of the target region in the image based on the cloud movement information.

12. The solar radiation estimation device further comprises: A cloud identification unit is provided to identify a cloud region in the image, the cloud region being a region including the cloud, The calculation unit calculates, as the feature amount, a ratio of the cloud region to the target region and a luminance statistic of the cloud region included in the target region; The solar radiation estimation device according to claim 8 , wherein the estimation unit uses the following formula (1) as the calculation formula when the occlusion prediction unit predicts that the sun will be occluded by the cloud in the occlusion prediction process: Q=(a1×F+b1×K+c1)×sin(h)...(1) where Q is the estimated value of the amount of solar radiation, F is the ratio, K is the statistic, h is the solar elevation angle, and a1, b1, and c1 are coefficients.

13. The estimation unit estimates, in the estimation process, a first amount of solar radiation that is the amount of solar radiation at a first position in the target area and a second amount of solar radiation that is the amount of solar radiation at a second position different from the first position, using the calculation formula; The solar radiation estimation device of claim 12, wherein the estimation unit sets the coefficient included in the calculation formula used to estimate the first solar radiation to a value different from the coefficient included in the calculation formula used to estimate the second solar radiation, depending on the distance between the first position and the position of the sun and the distance between the second position and the position of the sun.

14. an image acquisition unit for acquiring an image of the sky; a sun identification unit that identifies a position of the sun in the image; a region setting unit that sets, on the image, a target region based on the position of the sun, the target region being used for estimating the amount of solar radiation including the amount of scattered solar radiation due to clouds present around the sun; A calculation unit that calculates a feature amount based on pixels included in the target region; an estimation unit that performs an estimation process to estimate the amount of solar radiation including the amount of diffuse solar radiation based on the position of the sun and the feature amount; A solar radiation estimation system comprising:

15. A method for estimating solar radiation in a solar radiation estimation device, comprising: Get a blank image, Identifying the position of the sun in the image; A target area based on the position of the sun is set in the image to be used for estimating the amount of solar radiation including the amount of scattered solar radiation due to clouds present around the sun; Calculating a feature amount based on pixels included in the target region; A solar radiation estimation method comprising: performing an estimation process for estimating the amount of solar radiation including the amount of diffuse solar radiation based on the position of the sun and the feature amount.

16. A solar radiation estimation program for use in a solar radiation estimation device, comprising: A process of acquiring an image of the sky; determining a position of the sun in the image; A process of setting a target area based on the position of the sun in the image, the target area being used for estimating the amount of solar radiation including the amount of scattered solar radiation due to clouds around the sun; A process of calculating a feature amount based on pixels included in the target region; and estimating the amount of solar radiation including the amount of diffuse solar radiation based on the position of the sun and the feature amount.

Images