A satellite image-based method for assessing rooftop photovoltaic potential and its anti-interference

Abstract

The present invention discloses an anti-interference evaluation method for rooftop photovoltaic potential based on satellite images. In order to overcome the problem that the prior art is difficult to distinguish whether there is interference and error in the roof recognition result; the present invention includes the following steps: obtaining a satellite image from the top of the building; dividing the satellite image at equal distances and identifying the boundary lines; restoring and combining the matrix image according to the original position and connecting it into a number of closed areas; matching the shape of the closed area with the database to determine whether it is a building roof; evaluating the accuracy of the building roof recognition result obtained according to the closed area through the roof recognition result output by the trained recognition model, and optimizing the building roof recognition result; calculating the building roof area, and evaluating the rooftop photovoltaic potential in combination with the energy consumption and environmental information of the station area where the building is located. The recognition results are verified and evaluated by different roof recognition methods, the recognition results and the recognition process are optimized, and the accuracy of the recognition results is improved.

Description

Technical Field

[0001] The invention relates to the field of photovoltaic potential assessment, and in particular to a satellite image-based rooftop photovoltaic potential anti-interference assessment method. Background Art

[0002] With the growing global energy demand, photovoltaic power generation has attracted more and more attention due to its advantages of cleanliness, convenience, safety, and suitability for distributed networking, and has become one of the most promising renewable energy power generation methods. Quantitative evaluation of solar energy resource potential is the basis for energy planning, providing basic data and engineering construction guidance for the development and utilization of regional solar energy resources. The identification results of roof area have an important impact on the evaluation of photovoltaic potential. However, the current roof recognition process only uses one recognition method, which makes it difficult to remove the error interference of machine recognition, affecting the accuracy of recognition.

[0003] For example, a "distributed photovoltaic resource integration method, system, device and storage medium" disclosed in Chinese patent documents, with announcement number CN113076855A, includes: obtaining original pictures with roof information through satellites; inputting the obtained original pictures into a pre-trained artificial intelligence model, and filtering out the first pictures that can install photovoltaics from the original pictures; identifying the coordinate information of the first picture, matching the coordinate information with different customers through a big data algorithm, and obtaining a second picture associated with the customer; counting the coordinate information of the second picture into a database, and analyzing the potential data of the second picture in combination with meteorological data and industry data; integrating photovoltaic resources according to the obtained potential data. The roof recognition method of this scheme is single, and it is difficult to distinguish interference and errors. Summary of the invention

[0004] The present invention mainly solves the problem that it is difficult to distinguish whether there are interference and errors in the roof recognition results in the prior art; it provides an anti-interference evaluation method for roof photovoltaic potential based on satellite images, verifies the recognition results through different roof recognition methods, optimizes the recognition results and the recognition process, and improves the accuracy of the recognition results.

[0005] The above technical problems of the present invention are mainly solved by the following technical solutions:

[0006] A method for evaluating the anti-interference of rooftop photovoltaic potential based on satellite images comprises the following steps:

[0007] S1: Get satellite images from the top of the building;

[0008] S2: dividing the pre-processed satellite image into a number of matrix image units at equal distances, and identifying the boundary lines of the objects in the matrix image units respectively;

[0009] S3: Restore and combine the recognized matrix images according to the original positions, judge the continuity of the boundary lines, and connect them into several closed areas after correction;

[0010] S4: judging whether the closed area is a building roof according to the shape of the closed area and the database matching;

[0011] S5: Input the preprocessed satellite image into the trained recognition model and output the roof recognition result; evaluate the accuracy of the building roof recognition result obtained based on the closed area and optimize the building roof recognition result;

[0012] S6: Calculate the roof area of ​​the building and evaluate the rooftop photovoltaic potential based on the energy consumption and environmental information of the area where the building is located.

[0013] This solution cuts the image to identify each matrix unit separately, improving recognition efficiency and accuracy. After segmentation and recognition, the images are combined and spliced ​​to determine the continuity of the boundary lines, reducing the impact of distortion caused by satellite image shooting; the recognition results of the recognition model are then used to verify the results, evaluate the accuracy of the building roof recognition results obtained based on the closed area, intuitively obtain the accuracy of the recognition results, and improve the reliability of the recognition results.

[0014] Preferably, the satellite image is preprocessed by UNB pan-sharpening including both the MS band and the Pan band. This method preserves the color of the object image.

[0015] Preferably, the identification process of the boundary line is:

[0016] Extracting RGB data of each data point in each matrix image unit;

[0017] Traverse all data points, compare them with the RGB data of each point around the data point, and extract the boundary curve; if the RGB data difference of a pair of data points is within the set distinction threshold range, then the pair of data points is in a color value area; the intersection of each color value area is the boundary line. Cut the image to identify each matrix unit separately to improve recognition efficiency and recognition accuracy.

[0018] Preferably, the step S3 comprises the following steps:

[0019] S301: Restoring and splicing each matrix image unit according to the segmentation position;

[0020] S302: Determine whether the spliced ​​boundary curves are all connected into a continuous closed curve. If so, the process ends and proceeds to step S4; otherwise, the process proceeds to the next step of determination.

[0021] S303: determining the distance between the endpoint of the boundary line at the edge of the matrix image unit and the nearest endpoint of the boundary line at the edge of the adjacent matrix image unit; if the distance is less than the error threshold, taking the middle point of the boundary line endpoints at the connection of the two matrix image units as the correction point, moving the boundary line endpoints to the correction point, and fitting a curve; if the distance is greater than the error threshold, returning to step S2 to identify the boundary line again;

[0022] S304: Connecting the boundary lines to form a number of closed areas.

[0023] After segmentation and identification, the continuity of boundary lines is determined by combining and splicing to reduce the influence of distortion caused by satellite image shooting; preferably, the database stores closed area shapes including building roof shapes, road shapes and greening shapes; the area of ​​the closed area is calculated to determine whether the area of ​​the closed area is greater than the photovoltaic construction threshold; if so, it is determined that the closed area has the ability to build photovoltaics and the closed area is retained; otherwise, the closed area is deleted;

[0024] Determine whether the area of ​​the enclosed area is within the roof area threshold range; if so, proceed to the next step of determination; otherwise, determine that the enclosed area does not belong to the building roof and delete the enclosed area;

[0025] Match the shape of the closed area with the shape of the building roof in the database; if the match is successful, the closed area is the building roof; otherwise, match it with the data in the remaining database to determine the type of the closed area.

[0026] The image matching is used to determine whether it is a building roof. The image matching includes parameters such as shape and ratio of each side. Interference such as roads, elevated roads and green plants are removed.

[0027] Preferably, the step S5 comprises the following steps:

[0028] S501: input the preprocessed satellite image into the trained U-net network recognition, and output the roof recognition result of the recognition model;

[0029] S502: Coincidentally matching the corresponding recognition model roof recognition result and the building roof determined by the closed area according to the corresponding relationship of the coordinates;

[0030] S503: Determine the ratio of the overlapping area to the building roof determined by the closed area; if the ratio is greater than the overlapping threshold, the accuracy of the building roof recognition result is judged to be high, and enter step S6; otherwise, return to step S2.

[0031] The recognition results of the recognition model are used to verify the results, evaluate the accuracy of the building roof recognition results obtained based on the closed area, intuitively obtain the accuracy of the recognition results, and improve the reliability of the recognition results.

[0032] Preferably, for the same corresponding roof, the overlapping portion of the roof areas of the two recognition results is determined as the building roof;

[0033] The area of ​​the non-overlapping part of the roof area of ​​the two recognition results corresponding to the same roof is calculated; if the area of ​​the non-overlapping part is less than or equal to the error area threshold, the part is determined to be a non-building roof and deleted; otherwise, the next step is performed;

[0034] Determine whether the non-overlapping area is a custom regular shape; if so, determine that the area is a building roof, otherwise delete the area.

[0035] Remove roof areas that do not meet construction standards in advance to improve calculation accuracy.

[0036] Preferably, the calculation process of the rooftop photovoltaic potential in step S6 is as follows:

[0037] S601: Obtain the average daily energy consumption of the area where the building is located in each quarter according to the satellite positioning information;

[0038] S602: According to the statistical information of the meteorological bureau, the average daily light intensity of the region in each season is obtained;

[0039] S603: Evaluate the daily average energy consumption ratio by quarter;

[0040]

[0041] Among them, P is the average daily energy consumption proportion in a quarter;

[0042] ρ is the photovoltaic conversion efficiency per unit area per unit time;

[0043] L is the average daily light intensity of the current quarter;

[0044] S na is the roof area of ​​the nth building roof;

[0045] S nht is the blocked roof area of ​​the nth building roof at time t;

[0046] T is the average daily sunshine time in the current quarter;

[0047] t is the shading time of the building roof;

[0048] t1 is the occlusion start time;

[0049] t2 is the occlusion start time;

[0050] C d is the average daily energy consumption of the area in the current quarter;

[0051] S604: Establish a rooftop photovoltaic potential assessment table, with different daily average energy consumption ratios corresponding to different potential levels;

[0052] S605: Take the weighted average of the daily energy consumption proportions of the four quarters and compare it with the roof potential assessment table to determine the roof potential level.

[0053] The proportion of electricity provided by rooftop photovoltaics to power the substation is used to assess its photovoltaic potential and determine whether it is worth building.

[0054] The beneficial effects of the present invention are:

[0055] 1. Identify each matrix unit separately by cutting the image to improve recognition efficiency and accuracy.

[0056] 2. After segmentation and recognition, the images are combined and spliced ​​to determine the continuity of boundary lines and reduce the impact of distortion caused by satellite imagery.

[0057] 3. Verify the results through the recognition results of the recognition model, evaluate the accuracy of the building roof recognition results obtained based on the closed area, intuitively obtain the accuracy of the recognition results, and improve the reliability of the recognition results. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 It is a flow chart of the anti-interference evaluation method of rooftop photovoltaic potential of the present invention. DETAILED DESCRIPTION

[0059] The technical solution of the present invention is further specifically described below through embodiments and in conjunction with the accompanying drawings.

[0060] Example:

[0061] A satellite image-based rooftop photovoltaic potential anti-interference assessment method of this embodiment is as follows: Figure 1 As shown, the following steps are included:

[0062] S1: Get satellite imagery from the top of a building.

[0063] S2: Segment the preprocessed satellite image into a number of matrix image units at equal distances, and identify boundary lines of objects in the matrix image units respectively.

[0064] More than 70% of optical earth observation satellites and many modern aerial digital cameras can simultaneously capture low-resolution multispectral (MS) images and high-resolution panchromatic (Pan) images. The number of MS bands and the spectral bandwidth of Pan images often vary from sensor to sensor. In addition, seasonal and regional differences can also lead to differences in image information.

[0065] The stereo remote sensing image data is preprocessed by UNB pan-sharpening method that includes both MS band or Pan band. The color of the object image is retained, and the least square method is used to find the best fitting information between MS and Pan bands. Then this information is used to adjust the image processing parameters to achieve the best effect of Pan sharpening. During the fusion process, each group of Pan and MS effects are processed separately according to their image information. Regardless of the differences in sensors, seasons and regions, it can continuously and stably obtain good fusion effects.

[0066] The recognition process of boundary lines is as follows:

[0067] The RGB data of each data point in each matrix image unit is extracted.

[0068] Traverse all data points, compare them with the RGB data of each point around the data point, and extract the boundary curve.

[0069] If the RGB data difference of a pair of data points is within the set distinction threshold range, the pair of data points is located in a color value area; the intersection of each color value area is a boundary line.

[0070] The cut image is used to identify each matrix unit separately, thus improving the recognition efficiency and accuracy.

[0071] S3: The recognized matrix images are restored and combined according to the original positions, the continuity of the boundary lines is judged, and the corrected and connected into several closed areas.

[0072] S301: Restoring and splicing each matrix image unit according to the segmentation position.

[0073] S302: Determine whether the spliced ​​boundary curves are all connected into continuous closed curves. If so, the process ends and proceeds to step S4; otherwise, the process proceeds to the next step of determination.

[0074] S303: Determine the distance between the endpoint of the boundary line at the edge of the matrix image unit and the nearest endpoint of the boundary line at the edge of the adjacent matrix image unit. If the distance is less than the error threshold, take the middle point of the boundary line endpoint at the connection between the two matrix image units as the correction point, move the boundary line endpoint to the correction point, and fit it into a curve; if the distance is greater than the error threshold, return to step S2 to identify the boundary line again.

[0075] S304: Connecting the boundary lines to form a number of closed areas.

[0076] After segmentation and recognition, the images are combined and spliced ​​to determine the continuity of boundary lines and reduce the impact of distortion caused by satellite imagery.

[0077] S4: judging whether the closed area is a building roof based on the shape of the closed area and matching it with the database.

[0078] The database stores closed area shapes including building roof shapes, road shapes and green space shapes.

[0079] Calculate the area of ​​the closed area and determine whether the area of ​​the closed area is greater than the photovoltaic construction threshold; if so, determine that the closed area has the ability to build photovoltaics and retain the closed area; otherwise, delete the closed area;

[0080] Determine whether the area of ​​the enclosed area is within the roof area threshold range; if so, proceed to the next step of determination; otherwise, determine that the enclosed area does not belong to the building roof and delete the enclosed area;

[0081] Match the shape of the closed area with the shape of the building roof in the database; if the match is successful, the closed area is the building roof; otherwise, match it with the data in the remaining database to determine the type of the closed area.

[0082] The image matching is used to determine whether it is a building roof. The image matching includes parameters such as shape and ratio of each side. Interference such as roads, elevated roads and green plants are removed.

[0083] S5: Input the preprocessed satellite image into the trained recognition model and output the roof recognition result; evaluate the accuracy of the building roof recognition result obtained based on the closed area and optimize the building roof recognition result.

[0084] S501: Input the preprocessed satellite image into the trained U-net network recognition, and output the roof recognition result of the recognition model.

[0085] The trained recognition model is identified by the U-net network. Through machine learning, the model is continuously trained to identify various forms of roofs and remove the influence of surrounding roads and vegetation. The collected stereo remote sensing data is divided into a training set and a test set. The U-net network is continuously trained through the training set containing various roofs, and finally a trained recognition model is obtained through the test set test.

[0086] S502: According to the corresponding relationship of the coordinates, overlap the corresponding recognition model roof recognition result and the building roof determined by the closed area.

[0087] For the same corresponding roof, the overlapping part of the roof area of ​​the two recognition results is determined as the building roof.

[0088] The area of ​​the non-overlapping part of the roof area of ​​the two recognition results corresponding to the same roof is calculated; if the area of ​​the non-overlapping part is less than or equal to the error area threshold, the part of the area is determined to be a non-building roof and deleted; otherwise, the next step of judgment is carried out.

[0089] Determine whether the non-overlapping area is a custom regular shape; if so, determine that the area is a building roof, otherwise delete the area.

[0090] Remove roof areas that do not meet construction standards in advance to improve calculation accuracy.

[0091] S503: Determine the ratio of the overlapping area to the building roof determined by the closed area; if the ratio is greater than the overlapping threshold, the accuracy of the building roof recognition result is judged to be high, and enter step S6; otherwise, return to step S2.

[0092] The recognition results of the recognition model are used to verify the results, evaluate the accuracy of the building roof recognition results obtained based on the closed area, intuitively obtain the accuracy of the recognition results, and improve the reliability of the recognition results.

[0093] S6: Calculate the roof area of ​​the building and evaluate the rooftop photovoltaic potential based on the energy consumption and environmental information of the area where the building is located.

[0094] S601: Obtain the average daily energy consumption of the area where the building is located in each quarter according to the satellite positioning information;

[0095] S602: According to the statistical information of the meteorological bureau, the average daily light intensity of the region in each season is obtained;

[0096] S603: Evaluate the daily average energy consumption ratio by quarter;

[0097]

[0098] Among them, P is the average daily energy consumption proportion in a quarter;

[0099] ρ is the photovoltaic conversion efficiency per unit area per unit time;

[0100] L is the average daily light intensity of the current quarter;

[0101] S na is the roof area of ​​the nth building roof;

[0102] S nht is the blocked roof area of ​​the nth building roof at time t;

[0103] T is the average daily sunshine time in the current quarter;

[0104] t is the shading time of the building roof;

[0105] t1 is the occlusion start time;

[0106] t2 is the occlusion start time;

[0107] Cd is the average daily energy consumption of the area in the current quarter;

[0108] S604: Establish a rooftop photovoltaic potential assessment table, with different daily average energy consumption ratios corresponding to different potential levels;

[0109] S605: Take the weighted average of the daily energy consumption proportions of the four quarters and compare it with the roof potential assessment table to determine the roof potential level.

[0110] The proportion of electricity provided by rooftop photovoltaics to power the substation is used to assess its photovoltaic potential and determine whether it is worth building.

[0111] The solution of this embodiment improves recognition efficiency and accuracy by cutting the image to identify each matrix unit separately. After segmentation and recognition, the combination and splicing are used to determine the continuity of the boundary lines, reducing the influence of distortion caused by satellite image shooting; the recognition results of the recognition model are then used to verify the results, evaluate the accuracy of the building roof recognition results obtained based on the closed area, intuitively obtain the accuracy of the recognition results, and improve the reliability of the recognition results.

[0112] It should be understood that the embodiments are only used to illustrate the present invention and are not used to limit the scope of the present invention. In addition, it should be understood that after reading the content taught by the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms also fall within the scope limited by the appended claims of the application.

Claims

1. A method for anti-interference assessment of rooftop photovoltaic potential based on satellite images, characterized in that: The following steps are involved: S1: Get satellite images from the top of the building; S2: dividing the pre-processed satellite image into a number of matrix image units at equal distances, and identifying the boundary lines of the objects in the matrix image units respectively; S3: Restore and combine the recognized matrix images according to the original positions, judge the continuity of the boundary lines, and connect them into several closed areas after correction; S4: judging whether the closed area is a building roof according to the shape of the closed area and the database matching; S5: Input the preprocessed satellite image into the trained recognition model and output the roof recognition result; evaluate the accuracy of the building roof recognition result obtained based on the closed area and optimize the building roof recognition result; S6: Calculate the roof area of ​​the building and evaluate the rooftop photovoltaic potential based on the energy consumption and environmental information of the area where the building is located.

2. The method for anti-interference assessment of rooftop photovoltaic potential based on satellite images according to claim 1 is characterized in that: The satellite images are preprocessed by UNB pan-sharpening which includes both MS band and Pan band.

3. The method for anti-interference assessment of rooftop photovoltaic potential based on satellite images according to claim 1 is characterized in that: The identification process of the boundary line is as follows: Extracting RGB data of each data point in each matrix image unit; Traverse all data points, compare them with the RGB data of each point around the data point, and extract the boundary curve; if the RGB data difference of a pair of data points is within the set distinction threshold range, then the pair of data points is in a color value area; the intersection of each color value area is the boundary line.

4. A satellite image-based rooftop photovoltaic potential anti-interference assessment method according to claim 1, 2 or 3, characterized in that: The step S3 comprises the following steps: S301: restoring and splicing each matrix image unit according to the segmentation position; S302: Determine whether the spliced ​​boundary curves are all connected into a continuous closed curve. If so, the process ends and proceeds to step S4; otherwise, the process proceeds to the next step of determination. S303: determining the distance between the endpoint of the boundary line at the edge of the matrix image unit and the nearest endpoint of the boundary line at the edge of the adjacent matrix image unit; if the distance is less than the error threshold, taking the middle point of the boundary line endpoints at the connection of the two matrix image units as the correction point, moving the boundary line endpoints to the correction point, and fitting a curve; if the distance is greater than the error threshold, returning to step S2 to identify the boundary line again; S304: Connecting the boundary lines to form a number of closed areas.

5. The method for anti-interference assessment of rooftop photovoltaic potential based on satellite images according to claim 1, characterized in that: The database stores closed area shapes including building roof shapes, road shapes and greening shapes; calculates the area of ​​the closed area and determines whether the area of ​​the closed area is greater than the photovoltaic construction threshold; If so, it is determined that the enclosed area has the capacity to build photovoltaic power, and the enclosed area is retained; otherwise, the enclosed area is deleted; Determine whether the area of ​​the enclosed area is within the roof area threshold range; if so, proceed to the next step of determination; otherwise, determine that the enclosed area does not belong to the building roof and delete the enclosed area; Match the shape of the closed area with the shape of the building roof in the database; if the match is successful, the closed area is the building roof; otherwise, match it with the data in the remaining database to determine the type of the closed area.

6. A satellite image-based rooftop photovoltaic potential anti-interference assessment method according to claim 1 or 5, characterized in that: The step S5 comprises the following steps: S501: input the preprocessed satellite image into the trained U-net network recognition, and output the roof recognition result of the recognition model; S502: Coincidentally matching the corresponding recognition model roof recognition result and the building roof determined by the closed area according to the corresponding relationship of the coordinates; S503: Determine the ratio of the overlapping area to the building roof determined by the closed area; if the ratio is greater than the overlapping threshold, the accuracy of the building roof recognition result is judged to be high, and enter step S6; otherwise, return to step S2.

7. The method for anti-interference assessment of rooftop photovoltaic potential based on satellite images according to claim 6, characterized in that: For the same corresponding roof, the overlapping part of the roof area of ​​the two recognition results is determined as the building roof; The area of ​​the non-overlapping parts of the roof areas of the two recognition results corresponding to the same roof is calculated; if the area of ​​the non-overlapping parts is less than or equal to the error area threshold, the area is determined to be a non-building roof and the area is deleted; Otherwise, proceed to the next step; Determine whether the non-overlapping area is a custom regular shape; if so, determine that the area is a building roof, otherwise delete the area.

8. A satellite image-based rooftop photovoltaic potential anti-interference assessment method according to claim 1 or 7, characterized in that: The calculation process of the rooftop photovoltaic potential in step S6 is as follows: S601: Obtain the average daily energy consumption of the area where the building is located in each quarter according to the satellite positioning information; S602: According to the statistical information of the meteorological bureau, the average daily light intensity of the region in each season is obtained; S603: Evaluate the daily average energy consumption ratio by quarter; Among them, P is the average daily energy consumption proportion in a quarter; ρ is the photovoltaic conversion efficiency per unit area per unit time; L is the average daily light intensity of the current quarter; S na is the roof area of ​​the nth building roof; S nht is the blocked roof area of ​​the nth building roof at time t; T is the average daily sunshine time in the current quarter; t is the shading time of the building roof; t1 is the occlusion start time; t2 is the occlusion start time; C d is the average daily energy consumption of the area in the current quarter; S604: Establish a rooftop photovoltaic potential assessment table, with different daily average energy consumption ratios corresponding to different potential levels; S605: Take the weighted average of the daily energy consumption proportions of the four quarters and compare it with the roof potential assessment table to determine the roof potential level.

Images