JP5752031B2 - Program, information processing apparatus, information processing method, and information processing system - Google Patents

Description

  The present invention relates to a program for performing information processing by a computer, an information processing apparatus, an information processing method, and an information processing system.

  Conventionally, a solar panel, a camera that captures an image of the whole sky at a point where the solar panel is installed, and an image processing unit that detects a cloud distribution and a cloud movement from the whole sky image A photovoltaic system is known. In this solar power generation system, cloud distribution is predicted based on the captured cloud distribution and cloud movement, and the generated power of the solar panel at a future time point is predicted based on the predicted cloud distribution ( For example, see Patent Documents 1 and 2).

JP 2003-149350 A JP 2007-184354 A

  However, the conventional technique merely calculates the velocity and direction of the movement vector using the optical flow method based on the current image and the image one cycle before, and has a problem that the prediction accuracy is low. It was.

  The present invention has been made in view of such circumstances. The purpose is to provide a program or the like capable of predicting the power generation amount with higher accuracy by positioning a solar power generation device spread around the area (more densely in the future) as a cloud detection sensor.

The program disclosed in the present application is based on a power generation amount acquired from a plurality of distributed solar power generation devices, and includes a region including a solar power generation device whose power generation amount is a threshold value or less and a reference point in the region by the control unit. One identification information having a reference point that is calculated and stored in the storage unit in association with the calculated reference point and area identification information, and whose distance from the calculated reference point is within a threshold distance, or a calculated reference One identification information having an area similar to the area related to the point is extracted, the extracted one identification information is stored in the storage unit in association with the calculated area and the reference point, and the other identification stored in the storage unit A reference point group corresponding to information is read by the control unit, and based on a past reference point group of a plurality of other identification information stored in the storage unit and a past reference point group of the one identification information Multiple other identification information with high The computer having the control unit executes the process of calculating the prediction reference point related to the one identification information by the control unit based on the reference point group of the plurality of other identification information extracted by the control unit. Let

  According to one aspect of the present application, it is possible to predict the power generation amount with higher accuracy.

It is explanatory drawing which shows the outline | summary of an information processing system. It is a block diagram which shows the hardware group of an electric power generating apparatus. It is a block diagram which shows the hardware group of a server computer. It is explanatory drawing which shows the record layout of position DB. It is explanatory drawing which shows a cloud area | region. It is explanatory drawing which shows the record layout of cloud DB. It is explanatory drawing which shows the specific method of a shape. It is explanatory drawing which shows the other area | region identification method. It is explanatory drawing which shows the correlation process of a cloud area | region. It is explanatory drawing which shows the locus | trajectory of a gravity center. It is explanatory drawing which shows the procedure of a similarity calculation process. It is explanatory drawing which shows the record layout of candidate DB. It is explanatory drawing which shows the record layout of final candidate DB. It is explanatory drawing which shows the record layout of prediction gravity center DB. It is explanatory drawing which shows a prediction gravity center locus | trajectory and a prediction cloud area | region. It is explanatory drawing which shows the record layout of track record DB. It is explanatory drawing which shows the record layout of prediction DB. It is explanatory drawing which shows the record layout of regulation value DB. It is a flowchart which shows the procedure of a prediction process. It is a flowchart which shows the procedure of a prediction process. It is a flowchart which shows the procedure of a 1st prediction process. It is a flowchart which shows the procedure of a 1st prediction process. It is a flowchart which shows the procedure of a 2nd prediction process. It is a flowchart which shows the procedure of a 2nd prediction process. It is a flowchart which shows the procedure of a matching process. It is a flowchart which shows the procedure of a matching process. It is a flowchart which shows the procedure of a matching process. It is a flowchart which shows the procedure of a 3rd prediction process. It is a flowchart which shows the procedure of a 3rd prediction process. It is a flowchart which shows the procedure of a 3rd prediction process. It is a flowchart which shows the procedure of a 3rd prediction process. FIG. 6 is an explanatory diagram showing an overview of an information processing system according to a second embodiment. 6 is a block diagram illustrating a hardware group of a server computer according to Embodiment 2. FIG. It is explanatory drawing which shows the record layout of wind direction wind speed DB. It is explanatory drawing which shows the record layout of position DB which concerns on Embodiment 2. FIG. It is explanatory drawing which shows the record layout of the cloud DB which concerns on Embodiment 2. FIG. It is explanatory drawing which shows the record layout of candidate DB which concerns on Embodiment 2. FIG. It is a flowchart which shows the memory | storage process procedure of a wind direction and a wind speed. It is a flowchart which shows the procedure of the prediction process using a wind direction and a wind speed. It is a flowchart which shows the procedure of the prediction process using a wind direction and a wind speed. It is a functional block diagram which shows operation | movement of the server computer of the form mentioned above. FIG. 9 is a block diagram illustrating a hardware group of a server computer according to a third embodiment.

Embodiment 1
Hereinafter, embodiments will be described with reference to the drawings. FIG. 1 is an explanatory diagram showing an outline of an information processing system. The information processing system includes a plurality of photovoltaic power generation apparatuses 2 and information processing apparatuses 1 that are connected to each other via a communication network N such as the Internet, a LAN (Local Area Network), and a public telephone network. The information processing apparatus 1 is, for example, a server computer or a personal computer. Hereinafter, the information processing apparatus 1 will be described as the server computer 1. The solar power generation device 2 (hereinafter referred to as the “power generation device 2”) is distributed and installed in facilities such as homes, buildings, condominiums, schools, and stations, and the amount of power generated by sunlight is transferred to the server computer 1 via the communication network N. Send.

  The server computer 1 regards the power generation devices 2 distributed as a sensor, and identifies a region where the power generation amount is less than or equal to a threshold (hereinafter referred to as a cloud region). The server computer 1 stores the movement history of the cloud area. Further, the server computer 1 refers to the history, predicts the locus of the cloud region, and predicts the power generation amount of the power generation device 2. Details will be described below.

  FIG. 2 is a block diagram illustrating a hardware group of the power generation device 2. The power generation device 2 includes a central processing unit (CPU) 21, a random access memory (RAM) 22, a solar module 23, a watt hour meter 24, an output unit 26, and a clock unit 28 as control units. The CPU 21 is connected to each part of the hardware via the bus 27. The CPU 21 controls each part of the hardware according to the control program stored in the RAM 22. The RAM 22 is, for example, SRAM (Static RAM), DRAM (Dynamic RAM), flash memory, or the like. The RAM 22 temporarily stores various data generated when the CPU 21 executes various programs. The clock unit 28 outputs date / time information to the CPU 21.

  The solar module 23 generates power by receiving sunlight. The watt-hour meter 24 measures the amount of power generated by the solar module 23. The watt-hour meter 24 outputs the measured power generation amount to the CPU 21. The CPU 21 transmits the power generation amount to the server computer 1 via the output unit 26. The CPU 21 refers to the date and time information output from the clock unit 28 and uses the power generation amount measured every predetermined time (for example, every 15 minutes) and identification information (hereinafter referred to as device ID) for specifying the power generation device 2 as a server. Send to computer 1

  In addition, although this embodiment shows the form in which the power generation amount and the device ID are directly output from each power generation device 2 to the server computer 1, the present invention is not limited to this. For example, relaying may be performed by a relay computer (not shown) provided between the power generation device 2 and the server computer 1. The relay computer is installed in a facility such as a power management company that manages the region and a municipal office. In this case, the relay computer transmits the device ID and the power generation amount received from the power generation device 2 to the server computer 1.

  FIG. 3 is a block diagram showing a hardware group of the server computer 1. The server computer 1 includes a CPU 11, a RAM 12, an input unit 13, a display unit 14, a storage unit 15, a communication unit 16, a clock unit 18, and the like. The CPU 11 is connected to each part of the hardware via the bus 17. The CPU 11 controls each part of the hardware according to the control program 15P stored in the storage unit 15. The RAM 12 is, for example, SRAM, DRAM, or flash memory. The RAM 12 also functions as a storage unit, and temporarily stores various data generated when the CPU 11 executes various programs.

  The communication unit 16 is a gateway or the like that functions as a firewall, and transmits / receives information to / from the power generation apparatus 2 using HTTP (HyperText Transfer Protocol) or the like. The display unit 14 is a liquid crystal display or an organic EL (electroluminescence) display, and displays necessary information under the control of the CPU 11. The storage unit 15 is, for example, a large-capacity flash memory or a hard disk, and stores a control program 15P, a cloud database (hereinafter referred to as DB) 151, a position DB 152, a prediction DB 153, a performance DB 154, and a candidate DB 155. The storage unit 15 stores a final candidate DB 157, a predicted center of gravity DB 158, a specified value DB 156, and the like. In the present embodiment, an example in which the cloud DB 151 and the position DB 152 are stored in the storage unit 15 will be described. However, the present invention is not limited to this. For example, the cloud DB 151 and the position DB 152 may be stored in another computer (not shown) connected to the server computer 1 via the communication network N. In this case, the CPU 11 accesses each DB as necessary, and reads or writes information.

  FIG. 4 is an explanatory diagram showing a record layout of the position DB 152. The position DB 152 includes a device ID field, a coordinate field, and the like. In the device ID field, device IDs for specifying the power generation devices 2 distributed and arranged are stored. In the coordinate field, the position of the power generation device 2 is stored in association with the device ID. The position is the latitude and longitude of the power generation device 2, and the latitude is stored in the coordinate X field and the longitude is stored in the coordinate Y field in association with the device ID. For example, coordinates (xa, ya) are stored in association with the device ID “A”.

  FIG. 5 is an explanatory diagram showing a cloud region. The horizontal axis in FIG. 5 is the x-axis corresponding to longitude, and the vertical axis is the y-axis corresponding to latitude. In the present embodiment, in order to facilitate the description, a predetermined latitude range and longitude range will be described as all areas to be measured. The entire region to be measured is divided into 20 in the x-axis direction and 20 in the y-axis direction. A region group indicated by hatching in FIG. 5A is a cloud region. For example, the coordinates (2, 15) are a cloud region, and (2, 14) is a clear region. FIG. 5B is an explanatory diagram showing a cloud region after one unit time has elapsed. It can be understood that the cloud region indicated by hatching moves to the east side while changing its shape after one unit time has passed. After one unit time has elapsed, the coordinates (2, 15) become a clear region. One unit time is, for example, 15 minutes or 1 hour. In addition, the numerical value quoted by this embodiment is an example, and is not restricted to this. In the present embodiment, a description will be given on the assumption that one power generation device 2 exists in one divided area.

  The CPU 11 of the server computer 1 receives the power generation amount from the power generation device 2 via the communication unit 16. When the power generation amount is equal to or less than the threshold value stored in the storage unit 15, the CPU 11 determines that the cloud region. On the other hand, when the received power generation amount exceeds the threshold value, the CPU 11 determines that the area is clear. For example, when the power generation amount of the power generation device 2 located at the device ID “A” and the coordinates (4, 10) is 3.05 KVA and exceeds the threshold value of 0.5 KVA, the CPU 11 determines the area (coordinates 4, 10) related to the coordinates. ) Is judged to be a clear area. Further, for example, when the power generation amount of the power generation device 2 located at the device ID “A” and the coordinates (4, 10) is 0.25 KVA and the threshold is 0.5 KVA or less, the CPU 11 determines the region coordinates (4, 4) related to the coordinates. 10) is determined as a cloud region. In the present embodiment, for ease of explanation, an example in which one power generation device 2 is installed in one area will be described. However, the present invention is not limited to this. A plurality of power generators 2 may exist in one area. In this case, the average power generation amount, the maximum power generation amount, the minimum power generation amount, the average power generation amount between the maximum power generation amount and the minimum power generation amount, and the like are compared with the threshold. That's fine.

  FIG. 6 is an explanatory diagram showing a record layout of the cloud DB 151. The cloud DB 151 includes a time field, a cloud ID field, a centroid coordinate field, a distance field from the centroid to the edge, a distance difference field from one unit time ago, and the like. Date and time are stored in the time field. It is assumed that the time is stored every unit time (for example, every 15 minutes). In the cloud ID field, unique identification information (hereinafter referred to as cloud ID) for specifying the cloud region shown in FIG. 5 is stored in association with the date and time. The barycentric coordinate field stores the coordinates of the barycenter as a reference point in the cloud area.

  The reference point of the cloud area refers to a point within the cloud area and calculated based on the shape of the cloud area. The reference point may be the center of gravity of the cloud region, for example. In the example of FIG. 5, the CPU 11 extracts the coordinates of the outer peripheral area of the cloud area. The CPU 11 calculates the x coordinate of the center of gravity by dividing the total x coordinate value of each outer peripheral area by the number of outer peripheral areas. Similarly, the CPU 11 calculates the y coordinate of the center of gravity by dividing the total y coordinate value of each outer peripheral area by the number of outer peripheral areas. The CPU 11 stores the calculated barycentric coordinates in association with the cloud ID.

  Other reference points may be calculated based on the average of the maximum value and the minimum value of the outer peripheral coordinates. Specifically, the CPU 11 extracts the maximum value and the minimum value from the x coordinates of the outer peripheral area. CPU11 calculates | requires x coordinate by taking the average of the maximum value and the minimum value. Similarly, the CPU 11 extracts the maximum value and the minimum value from the y coordinates of the outer peripheral area. The CPU 11 obtains the y coordinate by taking the average of the maximum value and the minimum value. In addition, the reference point may be the center coordinates of a circle that is inscribed or circumscribed to the cloud region. In the distance field from the center of gravity to the edge, information for specifying the shape of the cloud region in association with the cloud ID is stored. Note that when there is no cloud area in the entire area, information corresponding to time is not stored in the cloud DB 151.

  FIG. 7 is an explanatory diagram showing a method for specifying a shape. In the present embodiment, a distance group from the center-of-gravity coordinates for each predetermined angle to the area outer periphery coordinates is used as the cloud area shape. Specifically, the 12 o'clock direction is 0 degree and the 3 o'clock direction is 90 degrees. CPU11 extracts the outer periphery area | region coordinate which exists in a 0 degree direction with respect to a gravity center. The CPU 11 stores the distance between the center of gravity and the extracted outer peripheral area coordinates in association with the cloud ID and the angle (0 degree). Similarly, the CPU 11 extracts outer peripheral area coordinates existing in the direction of 10 degrees with respect to the center of gravity. The CPU 11 stores the distance between the center of gravity and the extracted outer peripheral area coordinates in association with the cloud ID and the angle (10 degrees). In the present embodiment, an example is given in which the distance to the outer peripheral area coordinates is calculated every 10 degrees, but the present invention is not limited to this, and other angles may be used. The CPU 11 repeats the process until it reaches 360 degrees.

  Note that the region calculation method is not limited to this. For example, the CPU 11 may extract the coordinates of the cloud region, and store the extracted coordinate group as the region shape in the storage unit 15 in association with the cloud ID. Further, the CPU 11 may count the number of coordinates of the region and store it in the storage unit 15 as an area in association with the cloud ID. Further, the CPU 11 may calculate a circle that is inscribed or circumscribed to the cloud region with the center of gravity as the center, and may use the circle as the shape of the region. In the distance difference field from one unit time ago, the distance between the centroid of the cloud region and the centroid of the cloud region one unit time before the cloud region is stored in association with the cloud ID. The CPU 11 calculates the distance between the centroids and stores it in association with the cloud ID.

  FIG. 8 is an explanatory diagram showing another region specifying method. There may be a case where one power generation device 2 does not exist in one region. In this case, the area and the center of gravity may be specified by the cluster method. The CPU 11 groups a plurality of power generation devices 2 whose power generation amount is equal to or less than a threshold value using a cluster method based on K-means. The CPU 11 sets the grouped power generation devices 2 as one area. The CPU 11 specifies the center of gravity based on the plurality of grouped power generators 2. In the example of FIG. 8, two cloud regions are specified. In the cloud region, the center of gravity indicated by X is determined.

  FIG. 9 is an explanatory diagram showing a cloud region association process. It is assumed that C indicated by hatching is a cloud region C currently specified based on the amount of power generation. The center of gravity of the cloud region C is indicated by G. Since the cloud moves or newly occurs, it is necessary to specify which cloud region is one hour before. In the example of FIG. 9, it is assumed that a cloud region C1, a cloud region C2, and a cloud region C3 indicated by dotted lines exist one unit time ago. The centers of gravity of the cloud region C1, the cloud region C2, and the cloud region C3 are G1, G2, and G3, respectively. The CPU 11 reads out the cloud ID and barycentric coordinates of the cloud region one unit time ago from the cloud DB 151. Further, the CPU 11 reads out the cloud ID and the barycentric coordinates of the current cloud region C from the cloud DB 151.

  The distance between the barycentric coordinates of the cloud area C and the barycentric coordinates of the cloud area one unit time ago is calculated. The CPU 11 determines whether or not the calculated distance is within the threshold distance stored in the storage unit 15. When the CPU 11 determines that the distance is within the threshold distance, the CPU 11 determines that the cloud area related to the barycentric coordinates is a cloud area one unit time before the cloud area C. In the example of FIG. 9, it is determined that the cloud area C1 is a cloud area one unit time ago. Note that the CPU 11 may determine that the cloud area having the center-of-gravity coordinates that is the closest distance among the calculated distances is a cloud area one unit time before the cloud area C.

  Alternatively, the cloud region one unit time ago, which is most similar to the cloud region shape and the cloud region shape one unit time ago, may be extracted. The CPU 11 reads out the outer peripheral coordinates of the current cloud area and the outer peripheral coordinates of the cloud area one unit time ago from the storage unit 15 and performs pattern matching. As a result of pattern matching, the cloud region having the most similar cloud region shape is defined as a cloud region one unit time ago. In addition, the CPU 11 reads the identification information stored in the cloud DB 151 and the distance from the center of gravity to the edge. The CPU 11 calculates the sum of the squares of the difference in distance between the angles related to the two cloud regions. The CPU 11 calculates the total value for each cloud region one unit time ago. In the example of FIG. 9, the total values for the cloud region C and the cloud region C1, the cloud region C and the cloud region C2, and the cloud region C and the cloud region C3 are obtained. The CPU 11 reads out the cloud ID of the cloud region related to one unit time before the calculated total value is the smallest.

  Further, the cloud areas having the closest areas may be determined as similar cloud areas. The CPU 11 counts the number of coordinates constituting the cloud area. The cloud area one unit time before the counted value closest to the current cloud area is extracted as a similar cloud area. In the present embodiment, for ease of explanation, the cloud area having the closest area is assumed to be a similar cloud area. In addition, the specification of the cloud region one unit time ago will be described using an example in which both the distance between the centers of gravity and the area are used, but only one of them may be used.

  In the example of FIG. 9, description will be made assuming that the cloud region one unit time before the cloud region C is the cloud region C1. The CPU 11 reads the cloud ID of the cloud area C1 with reference to the cloud DB 151. The CPU 11 stores the time of the cloud region C, the coordinates of the center of gravity, and the distance from the center of gravity to the edge in association with the same cloud ID as the read cloud ID. The CPU 11 calculates the distance based on the barycentric coordinates of the cloud area one unit time ago and the barycentric coordinates of the current cloud area. The CPU 11 associates the calculated distance with the cloud ID of the current cloud region and stores it in the distance difference field from one unit time ago.

  When the CPU 11 determines that the distance between the centroids exceeds a predetermined distance and the area difference is equal to or larger than the predetermined area, the CPU 11 determines that a new cloud has been generated. The CPU 11 generates a new cloud ID, and stores time, barycentric coordinates, and the distance from the barycenter to the edge in association with the generated cloud ID. Note that when a new cloud ID is generated, the CPU 11 stores 0 in the distance difference field from one unit time ago. A cloud ID is generated by repeating the above processing.

  FIG. 10 is an explanatory diagram showing the locus of the center of gravity. The solid line indicates the center of gravity locus of the cloud region to be predicted. The black circle in practice is the center of gravity of the current cloud region. Below, the process which estimates the locus | trajectory of the gravity center of the said cloud area | region is demonstrated. Dotted lines L1 to L8 indicate the center-of-gravity locus of the cloud area related to another cloud ID. It can be understood that the center-of-gravity locus of the cloud region to be predicted substantially matches the locus from L3 to L7. Conversely, the trajectories of the dotted lines L1, L2, and L8 do not match. It can be said that the trajectories from L3 to L7 will match the trajectories from L5 to L7 in the future (the time zone after the black circle), but L3 and L4 will not match.

  The CPU 11 reads the history of the center of gravity of the cloud area to be predicted. The CPU 11 calculates the similarity based on the read past centroid group and the past centroid group related to another cloud ID different from the cloud area to be predicted stored in the cloud DB 151. The CPU 11 extracts a plurality of cloud IDs with high similarity as candidates. Specifically, the CPU 11 calculates the Euclidean distance based on the read past centroid group and the past centroid group related to another cloud ID different from the cloud region to be predicted stored in the cloud DB 151. When the calculated Euclidean distance is equal to or less than the threshold stored in the storage unit 15, the CPU 11 extracts the cloud ID as a candidate.

  FIG. 11 is an explanatory diagram showing a procedure of similarity calculation processing. The CPU 11 reads out the barycentric coordinates of the cloud ID to be predicted from the cloud DB 151 by going back a predetermined unit time from the present time. For example, it is sufficient to read for 5 unit times. The table on the left side of FIG. 11 is the barycentric coordinate history of the cloud ID to be predicted. The CPU 11 reads the barycentric coordinates related to other cloud IDs with reference to the cloud DB 151. The table on the right side of FIG. 11 shows the barycentric coordinate history of other cloud IDs. A history of barycentric coordinates is stored in association with time.

  The CPU 11 calculates the Euclidean distance based on the following formula 1.

  Specifically, the square root of the difference between the x centroid coordinates at time t0 and the x centroid coordinates at time t11 and the square of the difference between the y centroid coordinates at time t0 and the y centroid coordinates at time t11 are added to obtain the square root. Ask for. The CPU 11 performs the same processing for times t1 and t12, t2 and t13, t3 and t14, and t4 and t15. The CPU 11 sets the total square root value as the Euclidean distance. When the Euclidean distance is equal to or less than the threshold value stored in the storage unit 15, the CPU 11 stores the cloud ID, time, and barycentric coordinates in the candidate DB 155 as candidates. The CPU 11 performs the above-described processing while shifting the time. That is, subsequently, the Euclidean distance with respect to the time t12, t13, t14, t15, t16 of the center of gravity of another cloud ID is calculated.

  The CPU 11 performs the process described above for all cloud ID records stored in the cloud DB 151. FIG. 12 is an explanatory diagram showing a record layout of the candidate DB 155. The candidate DB 155 includes a cloud ID field, a time field, and a barycentric coordinate field. In the cloud ID field, cloud IDs with high similarity extracted as candidates are stored. In the time field, a time zone in which the Euclidean distance is less than or equal to the threshold is stored in association with the cloud ID. In the barycentric coordinate field, barycentric coordinates corresponding to the cloud ID and time are stored. Note that when there are a plurality of time zones in which the Euclidean distance is equal to or less than the threshold in the same cloud ID, the time zone with the shortest Euclidean distance may be stored in the candidate DB 155.

  Thereby, the dotted lines L1, L2, and L8 are excluded from the candidates. In this embodiment, an example in which the Euclidean distance is used for calculating the similarity will be described. However, the present invention is not limited to this. The CPU 11 may calculate a correlation function as the similarity and may extract a cloud ID having a correlation function larger than a predetermined threshold as a candidate. For example, the CPU 11 calculates an approximate straight line based on the read barycentric coordinates. The CPU 11 uses the correlation function between the approximate straight line of the cloud region to be predicted and the approximate straight line of another cloud region as the similarity. The CPU 11 may extract time and barycentric coordinates related to an approximate line having a correlation function larger than a threshold value.

  The CPU 11 reads out the cloud ID and time stored in the candidate DB 155. The CPU 11 refers to the cloud DB 151 and reads the barycentric coordinates related to the cloud ID after the read time. The reading time may be a predetermined unit time stored in the storage unit 15 (for example, 5 unit hours).

  FIG. 13 is an explanatory diagram showing a record layout of the final candidate DB 157. The final candidate DB 157 includes a cloud ID field, a time field, and a barycentric coordinate field. In the cloud ID field, candidate cloud IDs are stored. The CPU 11 refers to the candidate DB 155 and the cloud DB 151 and reads the time after the time stored in the candidate DB 155 and the corresponding barycentric coordinates. The CPU 11 stores the barycentric coordinates read in association with the cloud ID and time in the barycentric coordinate field. In the example of FIG. 13, for the cloud ID “11JA0005”, the corresponding barycentric coordinates from t10 to t14 after time t9 are read.

  That is, the CPU 11 stores a barycentric coordinate group for a predetermined time (for example, for 5 unit hours) after the time of the cloud region determined to have high similarity in the barycentric coordinate field. In the example of FIG. 13, the times t10, t11, t12, t13, and t14 after the time t9 are stored. The CPU 11 calculates a predicted centroid based on the time and centroid coordinates stored in the final candidate DB 157. CPU11 performs the process which deletes the cloud area | region which concerns on cloud ID which does not deserve reference first. The CPU 11 calculates the variance with reference to the barycentric coordinates of the plurality of candidate cloud IDs stored in the final candidate DB 157. Specifically, the CPU 11 calculates the variance based on the difference between the average value of the barycentric coordinates for each unit time stored in the final candidate DB 157 and the barycentric coordinates.

  In the example of FIG. 13, the average coordinate is calculated based on the barycentric coordinate of cloud ID “11JA0005” at time t10, the barycentric coordinate of cloud ID “11JA0006” at time t24, and the barycentric coordinate of cloud ID “11JA0007” at time t66. calculate. The CPU 11 adds the square of the difference between the average coordinate and the centroid coordinate of t10, the square of the difference between the average coordinate and the centroid coordinate of t24, and the average coordinate and the centroid coordinate of t66. The CPU 11 performs the same processing for t11, t25 and t67, t12 and t26 and t68, t13 and t27 and 69, and t14 and t28 and t70. The CPU 11 sets the total value of the added values for all time zones as variance. When the CPU 11 determines that the total value is equal to or less than the threshold value stored in the storage unit 15, the CPU 11 calculates the predicted centroid based on the centroid coordinates of the candidate cloud ID stored in the final candidate DB 157 as having little variation. In the present embodiment, dispersion is cited as a method for calculating the degree of variation, but the present invention is not limited to this. CPU11 calculates | requires an approximate line based on the gravity center coordinate of cloud ID. The CPU 11 may calculate the degree of variation based on the variance of the inclination of the approximate straight line of each cloud ID or the difference between the maximum value and the minimum value of the inclination.

  If the CPU 11 determines that the total value is larger than the threshold stored in the storage unit 15, the CPU 11 calculates the predicted centroid based on the past centroid coordinates of the cloud ID to be predicted, assuming that the variation is large. For example, when the candidate is a cloud region from dotted lines L3 to L8, the variance increases due to the dotted lines L3 and L4. In this case, the CPU 11 calculates the predicted centroid based on the past centroid coordinates indicated by the solid line L. Specifically, the CPU 11 calculates a movement vector from the barycentric coordinate one unit time before the cloud area to be predicted to the barycentric coordinate of the current cloud area. The CPU 11 calculates the barycentric coordinates after one unit time based on the calculated movement vector. The CPU 11 adds the movement vector to the barycentric coordinates after one unit time, and calculates the barycentric coordinates after two unit times. The predicted center of gravity is calculated by repeating the above processing. On the contrary, when it is determined that the candidates are the dotted lines L5 to L7 and the variance is small, the predicted centroid is calculated based on the centroid coordinates of the candidates L5 to L7.

  Below, the process which calculates a prediction gravity center based on extracted cloud candidate ID is demonstrated. Assume that the candidate cloud IDs in FIG. 10 are dotted lines L5, L6, and L7. The CPU 11 reads the barycentric coordinates of each cloud ID from the final candidate DB 157 in chronological order. In the example of FIG. 13, the barycentric coordinates at times t10, t24, and t66 are read. CPU11 calculates | requires the average value of the gravity center coordinate read in time series order. The CPU 11 stores the average value in the predicted centroid DB 158 as a predicted centroid (hereinafter referred to as a predicted centroid).

  FIG. 14 is an explanatory diagram showing a record layout of the predicted centroid DB 158. As shown in FIG. The predicted centroid DB 158 includes a cloud ID field, a time field, a predicted centroid coordinate field, and the like. The cloud ID field stores the cloud ID of the cloud region to be predicted. In the time field, the future time is stored for each unit time. In the predicted centroid coordinate field, a centroid predicted in association with a future time is stored as a predicted centroid. The CPU 11 stores the time obtained by sequentially adding the unit time to the current time in the time field of the predicted centroid DB 158 in association with the cloud ID. For example, when the current time is 9:00 am and the unit time is 15 minutes, the predicted center-of-gravity coordinate time is 9:15 (T1), 9:30 (T2), 9:45 (T3), 10:00 (T4), 10:15 (T5), etc. are stored every 15 minutes.

  In the present embodiment, the average value of the barycentric coordinates at each time stored in the final candidate DB 157 is calculated as the predicted barycenter, but the present invention is not limited to this. For example, the CPU 11 first obtains an average value for each time. CPU11 calculates | requires the distance with each gravity center coordinate of the time corresponding to an average value. The CPU 11 deletes a predetermined number (for example, 3) of barycentric coordinates having a large distance. The CPU 11 calculates the average value of the barycentric coordinates after deletion. The CPU 11 may store the calculated average coordinates in the predicted centroid DB 158 in association with time. As a result, it is possible to prevent a decrease in prediction accuracy due to variations in barycentric coordinates.

  In addition, the CPU 11 extracts the maximum and minimum x coordinates from the candidate DB 155 every time. The CPU 11 stores the average value of the maximum x coordinate and the minimum x coordinate in the predicted centroid DB 158. Subsequently, the CPU 11 extracts the maximum and minimum y coordinates from the candidate DB 155 every time. The CPU 11 stores the average value of the maximum y coordinate and the minimum y coordinate in the predicted centroid DB 158. Thus, CPU11 is good also considering the average coordinate based on the maximum value and minimum value for every time as a prediction gravity center.

  FIG. 15 is an explanatory diagram showing a predicted gravity center locus and a predicted cloud area. The cloud area indicated by hatching is the current cloud area. A point indicated by a white circle is a predicted center of gravity. The cloud region is predicted to move according to a dotted line (predicted centroid trajectory) passing through each predicted centroid. The cloud region will be described as being the same region as the current cloud region. The CPU 11 reads the current barycentric coordinates corresponding to the cloud ID to be predicted from the cloud DB 151. The CPU 11 specifies the current cloud region shown in FIG. Specifically, the CPU 11 reads the coordinates determined to be a cloud. The CPU 11 calculates a movement vector from the current centroid coordinates to the predicted centroid coordinates stored in the predicted centroid DB 158. The CPU 11 determines the predicted cloud region by adding the movement vector to each predicted coordinate determined to be a cloud. The CPU 11 calculates a predicted cloud region for each unit time by repeating the above processing for each unit time.

  A region surrounded by a dotted line shown in FIG. 15 is a predicted cloud region for each unit time. Note that the distance from the center of gravity to the edge stored in the cloud DB 151 may be used as the predicted cloud region. That is, the CPU 11 reads the distance from the current barycentric coordinates to the edge from the cloud DB 151. The CPU 11 may calculate the predicted cloud region based on the predicted centroid coordinates stored in the predicted centroid DB 158 and the distance to the read edge. In addition, what is necessary is just to identify the outer periphery of a prediction cloud area | region by connecting the front-end | tip part of the line segment which added distance from the prediction gravity center coordinate in each angle with a straight line or a curve.

  The CPU 11 predicts the power generation amount for each power generation device 2 based on the predicted cloud region. FIG. 16 is an explanatory diagram showing a record layout of the record DB 154. As shown in FIG. The performance DB 154 includes a time field and a device ID field. The time field stores the current time and the time that goes back every unit time. t0 is the current time, t1 is the time one unit time before (for example, 15 minutes before), and t2 is the time two units before the time (for example, 30 minutes before). The power generation amount is stored in the device ID field in association with the device ID and time. When the device ID and the power generation amount are transmitted from the power generation device 2, the CPU 11 stores the power generation amount in the result DB 154 in association with the device ID and the current time t0. When the CPU 11 stores the power generation amount in the field of time t0, the CPU 11 stores the power generation amount already stored in association with the time one unit time ago. For example, the power generation amount at time t1 is stored as the power generation amount at updated time t2. The power generation amount is KVA as an example, but is not limited thereto. It is good also as a ratio (for example, 60% etc.) of the present electric power generation amount with respect to the electric power generation capability of the solar module 23 as electric power generation amount.

  FIG. 17 is an explanatory diagram showing a record layout of the prediction DB 153. The prediction DB 153 includes a time field and a device ID field. The time field stores a time one unit time after the current time. T1 is a time after one unit time (for example, 15 minutes) after the current time t0, and T2 is a time after two unit time (for example, after 30 minutes). In the device ID field, the predicted power generation amount is stored in association with the device ID and time.

  CPU11 reads the prediction cloud area | region for every unit time. The CPU 11 reads out device IDs belonging to the predicted cloud area. FIG. 18 is an explanatory diagram showing a record layout of the specified value DB 156. The specified value DB 156 includes a month field and a specified value field. The month field stores November to March, April to October, and cloudy information. In the specified value field, the predicted power generation amount is stored in association with each of November to March, April to October, and cloudy. For example, in the case of November to March, the predicted power generation amount may be set to 3.05 KVA, and from April to October when high power generation amount can be expected to be set to 3.5 KVA. Also, if it is determined that it belongs to the cloud area and it is cloudy, 0 KVA is stored because the power generation amount cannot be expected sufficiently. Note that the stored content of the specified value DB 156 is an example, and the specified values may be stored in more detailed date. The operator can change the specified value from the input unit 13 as appropriate. Note that the specified value may be provided for each device ID based on the power generation capability of each power generation device 2.

  The CPU 11 refers to the specified value DB 156 and stores the specified value (0 KVA) in the predicted DB 153 in association with the device ID belonging to the predicted cloud area. As a result, the predicted power generation amount of the device ID belonging to the predicted cloud region is 0 KVA. Next, a device ID that does not belong to the predicted cloud area will be described. The CPU 11 reads the specified value (3.05 V, etc.) with reference to the date and time output from the specified value DB 156 and the clock unit 18. The CPU 11 stores the read specified value in the prediction DB 153 in association with the device ID that does not belong to the predicted cloud region. As a result, 3.05V is stored in the device ID belonging to the area predicted to be sunny. The prediction DB 153 is generated by repeating the above processing every unit time.

  Various software processes in the above hardware will be described with reference to flowcharts. 19 and 20 are flowcharts showing the procedure of the prediction process. CPU11 receives apparatus ID and electric power generation amount output from the output part 26 of the electric power generating apparatus 2 for every unit time (step S191). In the present embodiment, the CPU 21 of the power generation device 2 is described as transmitting the power generation amount via the output unit 26 for each time stored in the storage unit 15 in advance. CPU11 reads the coordinate corresponding to apparatus ID from position DB152 (step S192). CPU11 reads a threshold value from the memory | storage part 15 (step S193). The CPU 11 determines whether or not the power generation amount exceeds a threshold value (step S194). If the CPU 11 determines that the threshold value is exceeded (YES in step S194), the CPU 11 regards the area of the coordinate as clear and moves the process to step S195. CPU11 memorize | stores apparatus ID, a coordinate, power generation amount, sunny information, and time in the memory | storage part 15 (step S195).

  On the other hand, if the CPU 11 determines that the power generation amount does not exceed the threshold value (NO in step S194), the CPU 11 regards the coordinate area as cloudy and shifts the process to step S196. The CPU 11 stores the device ID, coordinates, power generation amount, cloudiness information, and time in the storage unit 15 (step S196). After the processes in steps S195 and S196, the process proceeds to step S197. The CPU 11 determines whether or not the processing for all device IDs has been completed (step S197). If the CPU 11 determines that the process for all device IDs has not been completed (NO in step S197), the process returns to step S191. As a result, information on whether all the coordinates are clear or cloudy is stored.

  If the CPU 11 determines that the process for all the device IDs has been completed (YES in step S197), the process proceeds to step S198. The CPU 11 determines whether or not cloudy information is stored in all coordinates one unit time before (step S198). If the CPU 11 determines that cloudiness information is stored in all coordinates one unit time before (YES in step S198), the process proceeds to step S199. In step S199, a first prediction process described later is executed (step S199).

  If the CPU 11 determines that cloudiness information is not stored in all coordinates one unit time ago (NO in step S198), the process proceeds to step S201. The CPU 11 determines whether clear information is stored in all coordinates one unit time before (step S201). If the CPU 11 determines that clear information is stored in all coordinates one unit time before (YES in step S201), the process proceeds to step S202. CPU11 performs the 2nd prediction process mentioned later (step S202). When the CPU 11 determines that clear information is not stored in all coordinates one unit time before (NO in step S201), the association process (step S203) and the third prediction process (step S204) are executed. The The processing in steps S203 and S204 will be described separately.

  21 and 22 are flowcharts showing the procedure of the first prediction process. The CPU 11 reads the clear information and the cloudy information at each current coordinate (step S211). The CPU 11 determines whether or not cloudy information is stored in all current coordinates (step S212). If the CPU 11 determines that the cloudy information is stored in all the current coordinates (YES in step S212), the CPU 11 determines that it is cloudy in all the coordinates for two unit times in succession, and proceeds to step S213. . The CPU 11 reads the cloudy defined value from the defined value DB 156 (step S213). The CPU 11 stores the specified value read in association with each device ID and each unit time in the prediction DB 153 (step S214). For example, the predicted power generation amount 0 KVA is stored for all device IDs and all unit times.

  If the CPU 11 determines that cloud information is not stored in all current coordinates (NO in step S212), the process proceeds to step S215. The CPU 11 determines whether clear information is stored in all current coordinates (step S215). If the CPU 11 determines that clear information is stored in all current coordinates (YES in step S215), the process proceeds to step S216. The CPU 11 reads the date and time from the clock unit 18 (step S216). The CPU 11 reads a specified value corresponding to the date and time (step S217). The CPU 11 stores the specified value read in association with each device ID and each unit time in the prediction DB 153 (step S218). For example, the predicted power generation amount of 3.05 KVA is stored for all device IDs and all unit times.

  If the CPU 11 determines that clear information is not stored in all the current coordinates (NO in step S215), the process proceeds to step S219. The CPU 11 performs edge detection based on the read sunny information and cloudy information of each coordinate, and identifies a cloud region (step S219). Specifically, a coordinate group that changes from sunny coordinates to cloudy coordinates is extracted. CPU11 specifies the coordinate of the outer periphery of a cloud area | region based on the detected edge (step S221). The CPU 11 calculates the barycentric coordinates based on the outer peripheral coordinates (step S222).

  The CPU 11 newly generates a cloud ID (step S223). The CPU 11 stores the barycentric coordinates in the cloud DB 151 in association with the cloud ID (step S224). The CPU 11 stores the coordinates of the outer periphery and the coordinates of the outer periphery in the storage unit 15 in association with the cloud ID (step S225). Further, the CPU 11 calculates an area based on the coordinate group of the region, and stores the calculated area in the storage unit 15 in association with the cloud ID. The CPU 11 calculates the distance from the barycentric coordinate to the edge detected in step S219 for each unit angle (step S226). The CPU 11 stores the calculated distance in the cloud DB 151 in association with the cloud ID and the angle (step S227).

  The CPU 11 determines whether the processing has been completed for all cloud regions (step S228). If the CPU 11 determines that the process has not been completed for all cloud regions (NO in step S228), the process proceeds to step S221. Accordingly, when it is determined that a plurality of cloud regions have been generated by edge detection, the center of gravity and shape of the cloud region can be specified. If the CPU 11 determines that the process has been completed for all cloud regions (YES in step S228), the process proceeds to step S229. The CPU 11 maintains the stored content of the prediction DB 153 without changing it (step S229). That is, since it is a newly generated cloud region, the CPU 11 ends without performing the prediction process.

  23 and 24 are flowcharts showing the procedure of the second prediction process. The CPU 11 reads the clear information and the cloudy information at each current coordinate (step S231). The CPU 11 determines whether or not cloudy information is stored in all current coordinates (step S232). If the CPU 11 determines that cloudiness information is stored in all current coordinates (YES in step S232), the process proceeds to step S233. The CPU 11 reads the cloudy defined value from the defined value DB 156 (step S233). The CPU 11 stores the specified value read in association with each device ID and each unit time in the prediction DB 153 (step S234).

  If the CPU 11 determines that cloud information is not stored in all current coordinates (NO in step S232), the process proceeds to step S235. The CPU 11 determines whether clear information is stored in all current coordinates (step S235). If the CPU 11 determines that clear information is stored in all the current coordinates (YES in step S235), the CPU 11 determines that the entire surface has suddenly become clear, and shifts the processing to step S236. The CPU 11 reads the date and time from the clock unit 18 (step S236). CPU11 reads the regulation value corresponding to the date and time (step S237). The CPU 11 stores the specified value read in association with each device ID and each unit time in the prediction DB 153 (step S238). For example, the predicted power generation amount of 3.05 KVA is stored for all device IDs and all unit times.

  If the CPU 11 determines that clear information is not stored in all the current coordinates (NO in step S235), the CPU 11 proceeds to step S239. The CPU 11 performs edge detection based on the read sunny information and cloudy information of each coordinate, and specifies a cloud region (step S239). Specifically, a coordinate group that changes from sunny coordinates to cloudy coordinates is extracted. The CPU 11 specifies the coordinates of the outer periphery of the cloud region based on the detected edge (step S241). The CPU 11 calculates the barycentric coordinates based on the outer peripheral coordinates (step S242).

  The CPU 11 newly generates a cloud ID (step S243). The CPU 11 stores the barycentric coordinates in the cloud DB 151 in association with the cloud ID (step S244). The CPU 11 stores the coordinates of the outer periphery and the coordinates of the outer periphery in the storage unit 15 in association with the cloud ID (step S245). Further, the CPU 11 calculates an area based on the coordinate group of the region, and stores the calculated area in the storage unit 15 in association with the cloud ID. The CPU 11 calculates the distance from the barycentric coordinate to the edge detected in step S239 for each unit angle (step S246). The CPU 11 stores the calculated distance in the cloud DB 151 in association with the cloud ID and the angle (step S247).

  The CPU 11 determines whether or not the processing has been completed for all cloud regions (step S248). If the CPU 11 determines that the process has not been completed for all cloud regions (NO in step S248), the process proceeds to step S241. If the CPU 11 determines that the processing has been completed for all cloud regions (YES in step S248), the CPU 11 proceeds to step S249. The CPU 11 maintains the stored content of the prediction DB 153 without changing it (step S249). That is, the CPU 11 ends without performing the prediction process because it is a newly generated cloud region.

  25 to 27 are flowcharts showing the procedure of the association process. CPU11 performs the following processes, when it is NO at step S201. The CPU 11 reads the clear information and the cloudy information at each current coordinate (step S251). The CPU 11 determines whether or not cloudy information is stored in all current coordinates (step S252). If the CPU 11 determines that cloudy information is stored in all current coordinates (YES in step S252), the process proceeds to step S253. The CPU 11 reads the cloudy defined value from the defined value DB 156 (step S253). The CPU 11 stores the specified value read in association with each device ID and each unit time in the prediction DB 153 (step S254).

  If the CPU 11 determines that cloud information is not stored in all current coordinates (NO in step S252), the process proceeds to step S255. The CPU 11 determines whether clear information is stored in all current coordinates (step S255). If the CPU 11 determines that clear information is stored in all current coordinates (YES in step S255), the process proceeds to step S256. The CPU 11 reads the date and time from the clock unit 18 (step S256). The CPU 11 reads a specified value corresponding to the date and time (step S257). The CPU 11 stores the specified value read in association with each device ID and each unit time of the prediction DB 153 (step S258).

  If the CPU 11 determines that clear information is not stored in all current coordinates (NO in step S255), the process proceeds to step S259. The CPU 11 performs edge detection based on the read sunny information and cloudy information of each coordinate, and specifies a cloud region (step S259). The CPU 11 specifies the coordinates of the outer periphery specified by the edge and the coordinates within the outer periphery (step S261). The CPU 11 stores the outer peripheral coordinates and the outer peripheral coordinates in the storage unit 15. Further, the CPU 11 calculates an area based on the coordinate group of the region, and stores the calculated area in the storage unit 15 in association with the cloud ID. The CPU 11 calculates the barycentric coordinates based on the outer peripheral coordinates (step S262). The CPU 11 reads the cloud ID one unit time ago from the cloud DB 151 (step S263). When a plurality of cloud IDs exist, a plurality of cloud IDs are read out.

  The CPU 11 reads the barycentric coordinates corresponding to the cloud ID one unit time ago from the cloud DB 151 (step S264). The CPU 11 calculates the distance between the current center-of-gravity coordinates and the center-of-gravity coordinates one unit time ago (step S265). The CPU 11 extracts the barycentric coordinate and the corresponding cloud ID with the shortest distance (step S266). The CPU 11 determines whether or not the calculated distance is within the threshold distance stored in the storage unit 15 (step S267).

  If the CPU 11 determines that the calculated distance is within the threshold distance (YES in step S267), the process proceeds to step S268. The CPU 11 reads the current outer peripheral coordinates and the outer peripheral coordinates one unit time ago (step S268). The CPU 11 calculates similarity by pattern matching (step S269). Note that the similarity may be calculated by a mean square error between the distance from the current center-of-gravity coordinates to the edge and the distance from the center-of-gravity coordinates one unit time ago to the edge. In this case, it is determined that the similarity is higher as the mean square error is smaller. The CPU 11 may calculate the similarity by comparing the areas stored in the storage unit 15.

  The CPU 11 reads the threshold value from the storage unit 15. The CPU 11 determines whether or not the similarity is greater than or equal to a threshold value (step 271). If the CPU 11 determines that the similarity is not equal to or greater than the threshold (NO in step S271), the process proceeds to step S272. The CPU 11 determines whether there is another cloud ID one unit time before (step S272). If the CPU 11 determines that another cloud ID exists (YES in step S272), the process proceeds to step S2721. The CPU 11 reads the barycentric coordinates and cloud ID with the next smallest distance (step S2721). The CPU 11 returns the process to step S267. Thereby, the similarity for other cloud IDs is repeatedly calculated.

  If the CPU 11 determines that there is no other cloud ID (NO in step S272) and if it is determined that the distance calculated in step S267 exceeds the threshold distance (NO in step S267), the process proceeds to step S273. Transition. The CPU 11 determines that a new cloud has been generated, and generates a new cloud ID (step S273). The CPU 11 stores the coordinates of the outer periphery and the coordinates of the outer periphery in the storage unit 15 in association with the cloud ID (step S274). The CPU 11 calculates the distance from the barycentric coordinate to the edge detected in step S259 for each unit angle (step S275). The CPU 11 stores the calculated distance in the cloud DB 151 in association with the cloud ID and the angle (step S276).

  If the CPU 11 determines that the similarity is greater than or equal to the threshold in step S271 (YES in step S271), the CPU 11 shifts the process to step S277, assuming that the cloud area is one unit time ago. The CPU 11 assigns the same cloud ID as the cloud ID extracted in step S266 or S2721 (step S277). The CPU 11 stores the coordinates of the outer periphery and the coordinates of the outer periphery in the storage unit 15 in association with the same cloud ID (step S278). The CPU 11 calculates the distance from the barycentric coordinate to the edge detected in step S259 for each unit angle (step S279). The CPU 11 stores the calculated distance in the cloud DB 151 in association with the cloud ID and the angle (step S2710).

  The CPU 11 determines whether or not the processing has been completed for all cloud regions (step S2711). If the CPU 11 determines that the process has not been completed for all cloud regions (NO in step S2711), the process proceeds to step S261. Thereby, the association between each cloud region and each cloud region one unit time ago is performed. If the CPU 11 determines that the process has been completed for all cloud regions (YES in step S2711), the process ends.

  28 to 31 are flowcharts showing the procedure of the third prediction process. The CPU 11 reads out the cloud ID to be predicted (step S281). The CPU 11 reads the current center-of-gravity coordinates of the read cloud ID and the center-of-gravity coordinates of the past plural unit times (step S282). The CPU 11 calculates the Euclidean distance between the barycentric coordinate group of another cloud ID for the past unit time and the barycentric coordinate group of the cloud ID to be predicted (step S283). In this process, after calculating the Euclidean distance, the time zones of other cloud IDs are shifted to calculate a plurality of distances. CPU11 reads a threshold value from the memory | storage part 15 (step S284). The CPU 11 extracts another cloud ID, time, and barycentric coordinate group whose Euclidean distance is equal to or smaller than the threshold (step S285).

  The CPU 11 stores the time at which the Euclidean distance is minimum and the barycentric coordinate group in association with the other cloud ID in the candidate DB 155 (step S286). The CPU 11 refers to the cloud DB 151 and stores the barycentric coordinates of the time after a predetermined unit time in the final candidate DB 157 in association with other cloud IDs (step S2860). The CPU 11 determines whether there is another cloud ID that has not been processed by the above process (step S287). If the CPU 11 determines that another unprocessed cloud ID exists (YES in step S287), the process returns to step S283. As a result, another cloud ID similar to the past trajectory of the cloud ID to be predicted is extracted, and the candidate DB 155 and the final candidate DB 157 are completed.

  If the CPU 11 determines that there is no other unprocessed cloud ID (NO in step S287), the process proceeds to step S288. The CPU 11 refers to the final candidate DB 157 and calculates the average coordinates of the barycentric coordinates for each time (step S288). Specifically, the CPU 11 reads the barycentric coordinates for each time in the time field of the final candidate DB 157 and calculates the average coordinates of the barycentric coordinates. The CPU 11 calculates the total value of the distance between the average coordinates and the barycentric coordinates of each cloud ID for each time (step S289). The CPU 11 determines whether or not the total value has been calculated for all time zones (step S291). If the CPU 11 determines that the total value has not been calculated for all time zones (NO in step S291), the process returns to step S288.

  If the CPU 11 determines that the total value has been calculated for all time zones (YES in step S291), the CPU 11 calculates the total sum of the total values for each time zone as the degree of variation (step S292). The average of the sum may be used as the degree of variation. The CPU 11 determines whether or not the degree of variation is equal to or less than the threshold value stored in the storage unit 15 (step S293). If the CPU 11 determines that the degree of variation is equal to or less than the threshold (YES in step S293), the process proceeds to step S294. The CPU 11 stores the time for a predetermined unit time from the present and the average coordinates of each time in the predicted centroid DB 158 in association with the cloud ID to be predicted as the predicted centroid coordinates (step S294). In the present embodiment, an example in which the processing from step S306 to S3113 is performed when the variation is large is described, but the present invention is not limited to this. The process of step S293 may be omitted, that is, the process may proceed to step S294 without determining the degree of variation.

  The CPU 11 reads the center of gravity coordinates of the current cloud area from the cloud DB 151 and reads the outer peripheral coordinates of the cloud area from the storage unit 15 (step S295). The CPU 11 calculates a movement vector from the read center of gravity coordinates to the predicted center of gravity coordinates corresponding to the prediction time (step S296). The CPU 11 adds the calculated movement vector to the outer peripheral coordinates of the cloud area (step S297). The CPU 11 refers to the position DB 152 and reads out the device ID belonging to the outer peripheral coordinates after the addition (step S298). The CPU 11 reads a specified value corresponding to cloudiness from the specified value DB 156 (step S299).

  The CPU 11 stores the prescribed value of cloudiness in the prediction DB 153 in association with the time and device ID (step S301). The CPU 11 refers to the specified value DB 156 and reads a specified value corresponding to the date and time output from the clock unit 18 (step S302). The CPU 11 refers to the position DB 152, and reads out the device ID located outside the predicted outer peripheral coordinates after the addition in step S298 (step S303). CPU11 memorize | stores the prescription | regulation value of fineness in prediction DB153 corresponding to time and apparatus ID (step S304).

  The CPU 11 determines whether or not the processing has been completed for all times in the predicted centroid DB 158 (step S305). If the CPU 11 determines that the process has not been completed for all times to be predicted (NO in step S305), the CPU 11 proceeds to step S296. If the CPU 11 determines that the process has been completed for all the times (YES in step S305), the process proceeds to step S3113.

  If the CPU 11 determines in step S293 that the degree of variation is not less than or equal to the threshold value (NO in step S293), the process proceeds to step S306. The CPU 11 refers to the cloud DB 151 and reads the barycentric coordinates one unit time ago (step S306). The CPU 11 calculates a movement vector from the barycentric coordinate one unit time before to the current barycentric coordinate (step S307). The CPU 11 adds the movement vector to the barycentric coordinates according to the number of unit hours (step S308). The CPU 11 stores the added barycentric coordinates in association with the cloud ID and time in the predicted barycenter DB 158 as the predicted barycentric coordinates (step S309). Specifically, the movement vector is added to the current center of gravity, and the predicted center of gravity coordinates after one unit time are calculated and stored in the predicted center of gravity DB 158. Subsequently, the movement vector is added to the predicted center-of-gravity coordinates after one unit time, and the predicted center-of-gravity coordinates after two unit times are calculated and stored in the predicted center-of-gravity DB 158. The processing described above is performed every unit time.

  The CPU 11 reads the center of gravity coordinates of the current cloud area from the cloud DB 151 and reads the outer peripheral coordinates of the cloud area from the storage unit 15 (step S311). The CPU 11 calculates a movement vector from the read current center of gravity coordinates to the predicted center of gravity coordinates corresponding to the prediction time (step S312). CPU11 adds the calculated movement vector to the outer periphery coordinate of a cloud area | region (step S313). The CPU 11 refers to the position DB 152 and reads out the device ID belonging to the outer peripheral coordinates after the addition (step S314). The CPU 11 reads a specified value corresponding to cloudiness from the specified value DB 299 (step S315).

  The CPU 11 stores the cloudy prescribed value in the prediction DB 153 in association with the time and the device ID (step S316). The CPU 11 reads the date and time output from the clock unit 18 with reference to the specified value DB 156 (step S317). The CPU 11 reads a specified value corresponding to the date and time (step S318). The CPU 11 refers to the position DB 152 and reads out the device ID located outside the predicted outer peripheral coordinates after the addition in step S314 (step S319). The CPU 11 stores a clear prescribed value in the prediction DB 153 in association with the time and the device ID (step S3111).

  The CPU 11 determines whether or not the processing has been completed for all the times in the predicted centroid DB 158 (step S3112). If the CPU 11 determines that the process has not been completed for all predicted times (NO in step S3112), the process proceeds to step S312. If the CPU 11 determines that the process has been completed for all the times (YES in step S3112), the process proceeds to step S3113.

  The CPU 11 determines whether or not the above-described processing has been completed for all current cloud IDs (step S3113). If the CPU 11 determines that the process has not been completed for all current cloud IDs (NO in step S3113), the process returns to step S281 to perform a prediction process for other cloud IDs. If the CPU 11 determines that the process has been completed for all cloud IDs (YES in step S3113), the CPU 11 stores the current power generation amount in the result DB 154 in association with the device ID (step S3114). As described above, by using the locus of the past center of gravity with high reliability, it is possible to predict the power generation amount with higher accuracy.

Embodiment 2
The second embodiment relates to a form using wind direction or wind speed. FIG. 32 is an explanatory diagram showing an outline of the information processing system according to the second embodiment. In addition to the power generator 2, the anemometer 3 is distributed. In this embodiment, an example using both the wind direction and the wind speed will be described. However, the present invention is not limited to this. Either the wind direction or the wind speed may be used. The wind direction anemometer 3 outputs the measured wind direction and wind speed to the server computer 1 via the communication network N.

  FIG. 33 is a block diagram illustrating a hardware group of the server computer 1 according to the second embodiment. A wind direction / wind speed DB 159 is stored in the storage unit 15. FIG. 34 is an explanatory diagram showing a record layout of the wind direction / wind speed DB 159. The wind direction wind speed DB 159 includes a wind direction anemometer ID field, a time field, a wind direction field, a wind speed field, and the like. In the wind direction anemometer ID field, unique identification information for identifying the wind direction anemometer 3 (hereinafter referred to as wind direction anemometer ID) is stored. The CPU 11 stores the wind direction anemometer ID, the wind direction, and the wind speed transmitted from the wind direction anemometer 3 via the communication unit 16 in the RAM 12. The CPU 11 stores the wind direction and the wind speed in association with the time in synchronization with the timing of storing the power generation amount of the power generation device 2 in the cloud DB 151 every predetermined unit time.

  The time field stores a time t1 one unit time ago and a time t2 two unit times ago. In addition, you may memorize | store the wind direction and wind speed which the wind direction anemometer 3 transmitted every predetermined unit time matched with time. In the wind direction field, the wind direction is stored in association with the wind direction anemometer ID and time. For example, the wind direction may be 0 degrees for the north wind and 180 degrees for the south wind. The wind speed is stored in the wind speed field in association with the wind direction anemometer ID and time.

  FIG. 35 is an explanatory diagram showing a record layout of the position DB 152 according to the second embodiment. Furthermore, an anemometer ID field is provided. An anemometer ID is stored in association with the device ID. In the present embodiment, for ease of explanation, it is assumed that all the power generation devices 2 arranged in the entire region shown in FIG. 5 use one anemometer 3.

  FIG. 36 is an explanatory diagram showing a record layout of the cloud DB 151 according to the second embodiment. A wind direction field and a wind speed field are newly provided. When the CPU 11 stores the wind direction and the wind speed in association with the time in the wind direction / wind speed DB 159, the CPU 11 stores the wind direction and the wind speed in the cloud DB 151 in association with the time. In addition, when the wind direction and the wind speed data based on a plurality of wind direction anemometers 3 exist in the cloud region, the average wind direction and the average wind speed in the region may be stored in the cloud DB 151 in association with the time.

  FIG. 37 is an explanatory diagram showing a record layout of the candidate DB 155 according to the second embodiment. The wind direction and the wind speed are stored in association with the final candidate cloud ID and time. If the CPU 11 determines in step S293 that the variation exceeds the threshold value, the CPU 11 performs the following processing. The CPU 11 calculates the similarity between the current wind direction and wind speed and the wind direction and wind speed stored in the candidate DB 155. The similarity may be calculated as follows. The CPU 11 refers to the candidate DB 155 to obtain the average value of the wind direction for each cloud ID. CPU11 calculates | requires the square root of the difference of the average value of the calculated | required wind direction, and the present wind direction. CPU11 makes the reciprocal of a square root the similarity of a wind direction.

  Similarly, the CPU 11 refers to the candidate DB 155 to obtain the average value of the wind speed for each cloud ID. CPU11 calculates | requires the square root of the difference of the average value of the calculated | required wind speed, and the present wind speed. CPU11 makes the reciprocal of a square root the similarity of a wind speed. The CPU 11 multiplies the wind direction similarity by a first weight greater than 1 stored in the storage unit 15, and multiplies the wind speed similarity by a second weight smaller than the first weight stored in the storage unit 15. The overall similarity is calculated by adding the values. Note that the similarity calculation method is not limited to this. For example, weights need not be used.

  In addition, the similarity of the wind direction may be calculated based on the difference between the wind direction closest to the present in time and the current wind direction. In addition, the CPU 11 may store date and time information in time and calculate the similarity based on the same wind direction or wind speed as the current time zone (for example, morning) or season (for example, winter). The CPU 11 reads the threshold value stored in the storage unit 15. The CPU 11 extracts cloud IDs whose similarity exceeds a threshold value. The CPU 11 stores the extracted cloud ID in the final candidate DB 157. The CPU 11 refers to the cloud DB 151 and reads barycentric coordinates for a predetermined unit time after the time stored in the candidate DB 155. The CPU 11 stores the read time and barycentric coordinates in the final candidate DB 157 in association with the cloud ID.

  FIG. 38 is a flowchart showing a procedure for storing wind direction and wind speed. The CPU 11 acquires the wind direction anemometer ID, the wind direction, and the wind speed output from the wind direction anemometer 3 via the communication unit 16 (step S381). The CPU 11 stores the wind direction and the wind speed in the wind direction and wind speed DB 159 in association with the wind direction anemometer ID every predetermined unit time (step S382). The CPU 11 stores the wind direction and the wind speed in the cloud DB 151 in association with the time (step S383).

  39 and 40 are flowcharts showing the procedure of the prediction process using the wind direction and the wind speed. CPU11 performs the following processes, when it is NO at Step S293. In step S286, the CPU 11 stores the wind direction and the wind speed corresponding to the time in the candidate DB 155. CPU11 reads the wind direction and the wind speed corresponding to time from candidate DB155 (step S391). CPU11 calculates an average wind direction for every cloud ID (step S392). CPU11 reads the present wind direction (step S393). The CPU 11 calculates the square root of the difference between the calculated average wind direction and the current wind direction (step S394).

  The CPU 11 calculates the reciprocal of the square root (step S395). The CPU 11 reads the first weight stored in the storage unit 15 (step S396). The CPU 11 multiplies the reciprocal by the first weight to calculate the wind direction similarity (step S397). Note that the CPU 11 may add the first weight in addition to multiplication. The CPU 11 calculates the average wind speed for each cloud ID (step S398). CPU11 reads the present wind speed (step S399). The CPU 11 calculates the square root of the difference between the calculated average wind speed and the current wind speed (step S401).

  The CPU 11 calculates the reciprocal of the square root (step S402). CPU11 reads the 2nd weight smaller than the 1st weight memorize | stored in the memory | storage part 15 (step S403). The CPU 11 calculates the wind speed similarity by multiplying the reciprocal by the second weight (step S404). Note that the CPU 11 may add the second weight. The CPU 11 adds the wind speed similarity to the calculated wind direction similarity, and calculates the similarity related to the cloud ID (step S405). The CPU 11 determines whether or not the processing described above has been executed for all cloud IDs in the candidate DB 155 (step S406).

  If the CPU 11 determines that the process has not been completed for all cloud IDs (NO in step S406), the process returns to step S391. If the CPU 11 determines that the process has been completed for all cloud IDs (YES in step S406), the process proceeds to step S407. CPU11 reads a threshold value from the memory | storage part 15 (step S407). The CPU 11 extracts only cloud IDs whose similarity is greater than or equal to a threshold value (step S408). In the present embodiment, an example in which the similarity is determined to be greater as the degree of similarity is greater is described, but the present invention is not limited to this. For example, when the square root is used as the similarity instead of the reciprocal of the square root, it is determined that the similarity is smaller as the similarity is smaller.

  The CPU 11 stores the extracted cloud ID in the final candidate DB 157 (step S409). The CPU 11 reads, from the cloud DB 151, the barycentric coordinates of the time after the predetermined time corresponding to the extracted cloud ID (step S4011). The CPU 11 associates the barycentric coordinates corresponding to the time with the cloud ID and stores them in the final candidate DB 157 (step S4012). Specifically, the CPU 11 reads the last time stored in the candidate DB 155. The CPU 11 reads the barycentric coordinates for a predetermined number of hours from the last time from the cloud DB 151. The CPU 11 stores the barycentric coordinates read in association with the cloud ID in association with a predetermined number of hours from the final time.

  The CPU 11 refers to the final candidate DB 157 and calculates average coordinates for each time (step S4013). CPU11 calculates the total value of the distance of an average coordinate and the gravity center coordinate of each cloud ID for every time (step S4014). The CPU 11 determines whether or not the total value has been calculated for all time zones (step S4015). If the CPU 11 determines that the total value has not been calculated for all time zones (NO in step S4015), the process returns to step S4013.

  If the CPU 11 determines that the total value has been calculated for all the time zones (YES in step S4015), the CPU 11 calculates the sum of the total values for each time zone as the degree of variation (step S4016). The average of the sum may be used as the degree of variation. The CPU 11 determines whether or not the degree of variation is equal to or less than the threshold value stored in the storage unit 15 (step S4017). If the CPU 11 determines that the degree of variation is equal to or less than the threshold (YES in step S4017), the CPU 11 proceeds to step S294. On the other hand, if the CPU 11 determines that the degree of variation is not equal to or less than the threshold (NO in step S4017), the process proceeds to step S306. The subsequent processing is the same as that described in the first embodiment, and is therefore omitted. Thus, the CPU 11 is limited to histories having similar wind directions or wind speeds, and can further improve the prediction accuracy.

  The second embodiment is as described above, and the other parts are the same as those of the first embodiment. Therefore, the corresponding parts are denoted by the same reference numerals, and detailed description thereof is omitted.

Embodiment 3
FIG. 41 is a functional block diagram showing the operation of the server computer 1 of the above-described form. When the CPU 11 executes the control program 15P, the server computer 1 operates as follows. Based on the power generation amount output from the output unit 26, the calculation unit 101 calculates a region including the power generation device 2 having a power generation amount equal to or less than a threshold and the barycentric coordinates in the region. The storage processing unit 102 extracts one device ID based on the calculated center-of-gravity coordinates and the center-of-gravity coordinates stored in the storage unit 15 in association with the device ID, and the region where the device ID of the extracted position is calculated by the calculation unit 101 And stored in correspondence with the barycentric coordinates. The reading unit 103 reads a barycentric coordinate group corresponding to another device ID stored in the storage unit 15. The extraction unit 104 extracts a plurality of other device IDs having a high degree of similarity based on a past barycentric coordinate group of a plurality of other device IDs stored in the storage unit 15 and a past barycentric coordinate group of one device ID. To do. Based on the barycentric coordinate group of a plurality of other device IDs extracted by the extracting unit 104, the reference point calculation unit 105 calculates a predicted barycentric coordinate related to one device ID.

  FIG. 42 is a block diagram illustrating a hardware group of the server computer 1 according to the third embodiment. A program for operating the server computer 1 reads a portable recording medium 1A such as a CD-ROM, a DVD (Digital Versatile Disc) disk, a memory card, or a USB (Universal Serial Bus) memory into a reading unit 10A such as a disk drive. It may be stored in the storage unit 15. Further, a semiconductor memory 1B such as a flash memory storing the program may be mounted in the personal computer 1. Further, the program can be downloaded from another server computer (not shown) connected via a communication network N such as the Internet. The contents will be described below.

  The server computer 1 shown in FIG. 42 reads a program for executing the above-described various software processes from the portable recording medium 1A or the semiconductor memory 1B, or from another server computer (not shown) via the communication network N. to download. The program is installed as the control program 15P, loaded into the RAM 12, and executed. Thereby, it functions as the server computer 1 described above.

  The third embodiment is as described above, and the others are the same as in the first and second embodiments. Therefore, the corresponding parts are denoted by the same reference numerals, and detailed description thereof is omitted.

  The following additional notes are further disclosed with respect to the embodiments including the first to third embodiments.

(Appendix 1)
Based on the power generation amount acquired from a plurality of distributed solar power generation devices, the control unit calculates a region including a solar power generation device whose power generation amount is equal to or less than a threshold, and a reference point in the region,
One identification information is extracted by the control unit based on the reference point stored in the storage unit in association with the calculated reference point and the region identification information, and the extracted one identification information corresponds to the calculated region and reference point And store it in the storage unit,
A reference point group corresponding to other identification information stored in the storage unit is read by the control unit,
Based on a past reference point group of a plurality of other identification information stored in the storage unit and a past reference point group of the one identification information, a plurality of other identification information having a high degree of similarity is extracted by the control unit. And
A program for causing a computer having the control unit to execute a process of calculating a prediction reference point related to the one identification information by the control unit based on a reference point group of a plurality of other pieces of identification information extracted.

(Appendix 2)
The program according to attachment 1, wherein a predicted power generation amount of the photovoltaic power generation apparatus is determined based on the calculated prediction reference point and region.

(Appendix 3)
One identification information having a reference point whose distance from the calculated reference point is within a threshold distance is extracted, and the extracted one identification information is stored in association with the calculated region and the reference point. Program.

(Appendix 4)
The area and the reference point where the distance between the calculated reference point is within the threshold distance and one piece of identification information having an area similar to the area related to the calculated reference point is extracted, and the one piece of extracted identification information is calculated The program according to appendix 1 or 2 is stored in association with.

(Appendix 5)
A Euclidean distance is calculated based on a past reference point group related to a plurality of other identification information stored in the storage unit and a past reference point group related to the one identification information, and the calculated Euclidean distance is equal to or less than a threshold value. The program according to any one of supplementary notes 1 to 4, wherein a plurality of other identification information relating to the reference point group is extracted.

(Appendix 6)
The prediction reference point related to the one identification information is calculated based on the average of the reference points related to the plurality of other extracted identification information, or the average of the reference point having the maximum value and the reference point having the minimum value. The program as described in any one of.

(Appendix 7)
Calculate the degree of variation of the reference point related to the plurality of other identification information with high similarity extracted,
7. The prediction reference point of the one region is calculated based on a reference point group related to a plurality of other pieces of extracted identification information when the calculated degree of variation is equal to or less than a threshold value. program.

(Appendix 8)
The storage unit stores identification information, a reference point, a wind direction or a wind speed in association with each other,
When the degree of variation of the calculated plurality of pieces of identification information exceeds a threshold, the wind direction or wind speed corresponding to the one piece of identification information and other identification information related to the wind direction or wind speed having a high degree of similarity are extracted,
The program according to claim 7, wherein a prediction reference point related to the one identification information is calculated based on a reference point group related to the extracted other identification information.

(Appendix 9)
When a plurality of other identification information related to the wind direction or wind speed with a high degree of similarity is extracted, the degree of variation of the reference point related to the plurality of other extracted identification information is calculated,
The program according to claim 8, wherein when the calculated degree of variation is equal to or less than a threshold value, a prediction reference point related to the one identification information is calculated based on a reference point group related to the other extracted identification information.

(Appendix 10)
When the calculated degree of variation exceeds a threshold, based on the reference point of the one identification information and the past reference point of the one identification information stored in the storage unit, the prediction reference point related to the one identification information is determined. The program according to appendix 9.

(Appendix 11)
The reference point is the center of gravity of the region. The program according to any one of supplementary notes 1 to 10.

(Appendix 12)
Based on the power generation amount acquired from a plurality of distributed solar power generation devices, a calculation unit that calculates a region including solar power generation devices whose power generation amount is equal to or less than a threshold, and a reference point in the region;
One identification information is extracted based on the reference point calculated by the calculation unit and the reference point stored in the storage unit in association with the identification information of the region, and the extracted one identification information is calculated by the calculation unit. A storage processing unit for storing in the storage unit in association with a reference point;
A reading unit for reading a reference point group corresponding to other identification information stored in the storage unit;
An extraction unit that extracts a plurality of other identification information having a high degree of similarity based on a past reference point group of a plurality of other identification information stored in the storage unit and a past reference point group of the one identification information; ,
An information processing apparatus comprising: a reference point calculation unit that calculates a prediction reference point related to the one identification information based on a reference point group of a plurality of other identification information extracted by the extraction unit.

(Appendix 13)
Based on the power generation amount acquired from a plurality of distributed solar power generation devices, the control unit calculates a region including a solar power generation device whose power generation amount is equal to or less than a threshold, and a reference point in the region,
One identification information is extracted by the control unit based on the reference point stored in the storage unit in association with the calculated reference point and the region identification information, and the extracted one identification information corresponds to the calculated region and reference point And store it in the storage unit,
A reference point group corresponding to other identification information stored in the storage unit is read by the control unit,
Based on a past reference point group of a plurality of other identification information stored in the storage unit and a past reference point group of the one identification information, the control unit extracts a plurality of other identification information having a high degree of similarity. And
An information processing method using an information processing apparatus having the control unit, wherein the control unit calculates a prediction reference point related to the one identification information based on a plurality of extracted reference point groups of other identification information.

(Appendix 14)
In an information processing system in which a plurality of distributed photovoltaic power generation devices and information processing devices are connected via a communication network,
The solar power generator is
An output unit that outputs the amount of power generation to the information processing apparatus;
The information processing apparatus includes:
Based on the power generation amount output from the output unit, a calculation unit that calculates a region including a photovoltaic power generation device with a power generation amount equal to or less than a threshold, and a reference point in the region;
One identification information is extracted based on the reference point stored in the storage unit in association with the calculated reference point and the region identification information, and the extracted one identification information is used as the region and the reference point calculated by the calculation unit. A storage processing unit that associates and stores in the storage unit;
A reading unit for reading a reference point group corresponding to other identification information stored in the storage unit;
An extraction unit that extracts a plurality of other identification information having a high degree of similarity based on a past reference point group of a plurality of other identification information stored in the storage unit and a past reference point group of the one identification information; ,
An information processing system comprising: a reference point calculation unit that calculates a prediction reference point related to the one identification information based on a reference point group of a plurality of other identification information extracted by the extraction unit.

DESCRIPTION OF SYMBOLS 1 Server computer 2 Power generation device 1A Portable recording medium 1B Semiconductor memory 3 Anemometer 10A Reading part 11 CPU
12 RAM
13 Input unit 14 Display unit 15 Storage unit 15P Control program 16 Communication unit 18 Clock unit 21 CPU
22 RAM
23 Solar Module 24 Energy Meter 26 Output Unit 28 Clock Unit 101 Calculation Unit 102 Storage Processing Unit 103 Reading Unit 104 Extraction Unit 105 Reference Point Calculation Unit 151 Cloud DB
152 Position DB
153 Prediction DB
154 Results DB
155 Candidate DB
156 Default value DB
157 Final candidate DB
158 Prediction center of gravity DB
159 Wind direction DB
N communication network

Claims (5)

Based on the power generation amount acquired from a plurality of distributed solar power generation devices, the control unit calculates a region including a solar power generation device whose power generation amount is equal to or less than a threshold, and a reference point in the region,
One identification information having a reference point that is stored in the storage unit in association with the calculated reference point and the area identification information and whose distance from the calculated reference point is within a threshold distance, or related to the calculated reference point One identification information having a region similar to the region is extracted, and the extracted one identification information is stored in the storage unit in association with the calculated region and the reference point,
A reference point group corresponding to other identification information stored in the storage unit is read by the control unit,
Based on a past reference point group of a plurality of other identification information stored in the storage unit and a past reference point group of the one identification information, a plurality of other identification information having a high degree of similarity is extracted by the control unit. And
A program for causing a computer having the control unit to execute a process of calculating a prediction reference point related to the one identification information by the control unit based on a reference point group of a plurality of other pieces of identification information extracted. The program according to claim 1, wherein a predicted power generation amount of the photovoltaic power generation apparatus is determined based on the calculated prediction reference point and region. Based on the power generation amount acquired from a plurality of distributed solar power generation devices, a calculation unit that calculates a region including solar power generation devices whose power generation amount is equal to or less than a threshold, and a reference point in the region;
One identification information having a reference point that is stored in the storage unit in association with the reference information and the region identification information calculated by the calculation unit and that has a reference point whose distance from the calculated reference point is within a threshold distance, or calculated A storage processing unit that extracts one piece of identification information having a region similar to the region related to the reference point, and stores the extracted one piece of identification information in the storage unit in association with the region calculated by the calculation unit and the reference point; ,
A reading unit for reading a reference point group corresponding to other identification information stored in the storage unit;
An extraction unit that extracts a plurality of other identification information having a high degree of similarity based on a past reference point group of a plurality of other identification information stored in the storage unit and a past reference point group of the one identification information; ,
An information processing apparatus comprising: a reference point calculation unit that calculates a prediction reference point related to the one identification information based on a reference point group of a plurality of other identification information extracted by the extraction unit. Based on the power generation amount acquired from a plurality of distributed solar power generation devices, the control unit calculates a region including a solar power generation device whose power generation amount is equal to or less than a threshold, and a reference point in the region,
One identification information having a reference point that is stored in the storage unit in association with the calculated reference point and the area identification information and whose distance from the calculated reference point is within a threshold distance, or related to the calculated reference point One identification information having a region similar to the region is extracted, and the extracted one identification information is stored in the storage unit in association with the calculated region and the reference point,
A reference point group corresponding to other identification information stored in the storage unit is read by the control unit,
Based on a past reference point group of a plurality of other identification information stored in the storage unit and a past reference point group of the one identification information, the control unit extracts a plurality of other identification information having a high degree of similarity. And
An information processing method using an information processing apparatus having the control unit, wherein the control unit calculates a prediction reference point related to the one identification information based on a plurality of extracted reference point groups of other identification information. In an information processing system in which a plurality of distributed photovoltaic power generation devices and information processing devices are connected via a communication network,
The solar power generator is
An output unit that outputs the amount of power generation to the information processing apparatus;
The information processing apparatus includes:
Based on the power generation amount output from the output unit, a calculation unit that calculates a region including a photovoltaic power generation device with a power generation amount equal to or less than a threshold, and a reference point in the region;
Calculated reference point and one identification information distance between the reference point is calculated and stored in the serial憶部in association with the identification information having a reference point is within threshold distance area, or the calculated reference point A storage processing unit that stores, in the storage unit, one piece of identification information having a region similar to the region associated with the region calculated by the calculation unit and a reference point;
A reading unit for reading a reference point group corresponding to other identification information stored in the storage unit;
An extraction unit that extracts a plurality of other identification information having a high degree of similarity based on a past reference point group of a plurality of other identification information stored in the storage unit and a past reference point group of the one identification information; ,
An information processing system comprising: a reference point calculation unit that calculates a prediction reference point related to the one identification information based on a reference point group of a plurality of other identification information extracted by the extraction unit.

Images