Photovoltaic module installation optimization method, installation optimization device, equipment and medium

Abstract

The invention discloses an installation optimization method and an installation optimization device of a photovoltaic module, electronic equipment and a computer readable storage medium. The installation optimization method of the photovoltaic module comprises the following steps: determining a target construction area in which the photovoltaic module can be arranged according to an aerial top view of a construction site of the photovoltaic module, and converting the target construction area into a plane construction map; calculating a shadow shielding area generated by an obstacle in the plane construction map; and determining an initial arrangement mode of the photovoltaic module according to the plane construction map, and adjusting the initial arrangement mode based on the shadow shielding area to obtain an optimized arrangement mode. Therefore, the installation optimization of the photovoltaic module can be realized, and the problems of difficulty in 3D modeling and high manual design cost in the installation of the current photovoltaic module are solved.

Description

Technical Field

[0001] The present application relates to the field of photovoltaic installation technology, and in particular to a photovoltaic assembly installation optimization method, an installation optimization device, an electronic device, and a computer-readable storage medium. Background Art

[0002] Currently, the bottlenecks in developing renewable energy sources like photovoltaics lie in installation space and capacity, with installation space being a particularly prominent issue. Photovoltaic panels require ample space for installation and have strict environmental requirements, avoiding shadows. Statistics show that over 10 billion square meters of rooftops and other vertical surfaces in my country's urban and rural areas that receive sufficient sunlight could generate 2 trillion kilowatt-hours of electricity annually, representing approximately 28% of my country's total annual power generation. Furthermore, numerous ground-mounted power stations exist. Optimizing the installation of photovoltaic panels has become a pressing issue. Summary of the Invention

[0003] The embodiments of the present application provide a photovoltaic assembly installation optimization method, an installation optimization device, an electronic device, and a computer-readable storage medium to solve at least one of the above-mentioned technical problems.

[0004] The photovoltaic assembly installation optimization method according to the embodiment of the present application includes:

[0005] Determining a target construction area where the photovoltaic modules can be arranged based on an aerial view of the construction site of the photovoltaic modules, and converting the target construction area into a planar construction map;

[0006] Calculate the shadow occlusion area caused by obstacles in the planar construction diagram;

[0007] An initial arrangement of the photovoltaic components is determined according to the plan construction drawing, and the initial arrangement is adjusted based on the shadow-blocking area to obtain an optimized arrangement.

[0008] In certain embodiments, determining a target construction area where the photovoltaic modules can be arranged based on an aerial view of the construction site of the photovoltaic modules, and converting the target construction area into a planar construction map includes:

[0009] Obtaining an aerial view of the construction site of the photovoltaic module;

[0010] Based on the aerial photography scale, an image processing algorithm is used to extract the outline of the aerial top view to determine the target construction area where the photovoltaic modules can be arranged;

[0011] According to the inclination of the target construction area, the target construction area is converted into a plane construction map.

[0012] In certain embodiments, the step of extracting the contour of the aerial view using an image processing algorithm to determine a target construction area where the photovoltaic modules can be arranged includes:

[0013] performing contrast enhancement processing on the aerial top view to obtain an enhanced image;

[0014] Performing Gaussian filtering on the enhanced image to obtain a first denoised image;

[0015] performing Canning edge detection on the first denoised image to obtain a second denoised image;

[0016] Performing Hough transform shape fitting processing on the second denoised image to extract the target contour shape;

[0017] The target construction area is determined according to the target outline shape.

[0018] In some embodiments, converting the target construction area into a planar construction map according to the inclination of the target construction area includes:

[0019] When the target construction area is a planar structure, the target construction area is directly used as the planar construction map;

[0020] When the target construction area is an inclined structure, a ridge line of the target construction area is generated by a straight skeleton algorithm;

[0021] Dividing the roof line into a plurality of line segments according to the vertices of the roof line;

[0022] Determining the highest ridge line among the plurality of line segments according to the vertex positions of the plurality of line segments;

[0023] The slope areas connected by the highest roof ridge line are expanded to obtain the plan construction drawing.

[0024] In some embodiments, the target construction area includes polygon vertices and polygon contour lines, and generating the ridge line of the target construction area using a straight skeleton algorithm includes:

[0025] Determining whether the polygon outline connected by the polygon vertices is a gable edge to determine the type of the polygon vertices;

[0026] When the type of the polygon vertex is a free vertex, the free vertex is controlled to shrink inward by the straight skeleton algorithm to determine a free intersection vertex;

[0027] When the type of the polygon vertex is a partially free vertex, controlling the partially free vertex to move along the gable edge to determine a gable intersection vertex;

[0028] The ridge line is determined according to the free intersection point and the gable intersection point.

[0029] In some embodiments, the slope area is a trapezoidal slope, and the process of unfolding the slope areas connected by the highest ridge line to obtain the plan construction drawing includes:

[0030] Obtaining the top view length of the waist of the trapezoidal slope;

[0031] Determine the unfolded length of the waist of the trapezoidal slope according to the slope angle corresponding to the trapezoidal slope or the ridge line height corresponding to the highest ridge line, and the top-view length;

[0032] Keeping the upper base of the trapezoidal slope stationary, controlling the parallel movement of the lower base, and calculating the current length of the waist in real time;

[0033] When the current length is moved to be equal to the expanded length, the plane construction drawing is determined according to the upper base, lower base and waist of the trapezoidal slope.

[0034] In some embodiments, calculating the shadow occlusion area caused by obstacles in the planar construction diagram includes:

[0035] Determining the shadow distance corresponding to each point on the upper edge of the obstacle based on the geometric characteristic parameters of the obstacle and the solar radiation correlation parameters of the construction site at various times throughout the year;

[0036] Determine the outer edge of the shadow produced by the obstacle according to the shadow distance corresponding to each point of the upper edge;

[0037] The shadow-blocking area is determined according to the outer edge of the shadow.

[0038] In certain embodiments, determining the initial arrangement of the photovoltaic components according to the plan construction drawing, and adjusting the initial arrangement based on the shadow-blocking area to obtain an optimized arrangement includes:

[0039] Constructing a grid area corresponding to the photovoltaic assembly according to the length parameter, width parameter, spacing parameter, predetermined reference direction and reference direction inclination angle of the photovoltaic assembly;

[0040] Overlaying the grid area on the plan construction drawing to determine the initial arrangement of the photovoltaic modules;

[0041] The initial arrangement is adjusted according to the portion of the grid area that exceeds the planar construction drawing, the obstacle area, and the shadow-blocking area to obtain an optimized arrangement.

[0042] In certain embodiments, overlaying the grid area on the planar construction drawing to determine the initial arrangement of the photovoltaic components includes:

[0043] When the reference direction inclination angle is zero, covering the grid area on the planar construction drawing and aligning the grid area with the coordinate system of the planar construction drawing to determine the initial arrangement of the photovoltaic components;

[0044] When the reference direction inclination angle is not zero, the grid area is covered on the plane construction drawing, and then the grid area is rotated according to the reference direction inclination angle, and the grid area is expanded along the coordinate axis direction of the grid area to determine the initial arrangement of the photovoltaic components.

[0045] In certain embodiments, before overlaying the grid area on the planar construction drawing to determine the initial arrangement of the photovoltaic components, the installation optimization method further includes:

[0046] When the photovoltaic components are installed vertically, calculating the component shielding area generated by the photovoltaic components in the planar construction drawing;

[0047] The grid area is adjusted according to the component occlusion area.

[0048] In certain embodiments, the photovoltaic assembly is configured to be connected to an inverter to convert direct current into alternating current, and the installation optimization method further comprises:

[0049] An optimized wiring mode of the photovoltaic modules is determined based on the conversion efficiency of the inverter and the total length of the cables used for wiring the photovoltaic modules.

[0050] In certain embodiments, determining the optimized wiring mode of the photovoltaic assembly based on the conversion efficiency of the inverter and the total length of the cables used for wiring the photovoltaic assembly includes:

[0051] Determining an expected number of photovoltaic modules connected to each inverter according to the maximum power, maximum input DC power, capacity ratio, and minimum input DC power of the inverter;

[0052] Connecting the photovoltaic modules in a predetermined wiring manner to connect a current group of the photovoltaic modules;

[0053] When the number of connections of the photovoltaic components of the current group reaches the expected number, the connection of the photovoltaic components of the current group is stopped, and the step of connecting the photovoltaic components using the predetermined connection method is repeated to connect the next group of photovoltaic components until all the photovoltaic components are connected.

[0054] In certain embodiments, determining the optimized connection mode of the photovoltaic assembly based on the conversion efficiency of the inverter and the total length of the cables used for connecting the photovoltaic assembly further includes:

[0055] When the number of photovoltaic components connected in the last group is less than the expected number, the expected number of photovoltaic components connected to each inverter is adjusted, or the last group of photovoltaic components is connected to a single inverter.

[0056] In some embodiments, the predetermined connection mode includes any one or more of a horizontal connection mode, a vertical connection mode, a straight connection mode, a C-shaped connection mode, and a loop connection mode.

[0057] In certain embodiments, determining the optimized wiring mode of the photovoltaic assembly based on the conversion efficiency of the inverter and the total length of the cables used for wiring the photovoltaic assembly includes:

[0058] determining a graph structure according to a set of component features and an edge set of the photovoltaic component, wherein the component features include position coordinates, electrical parameters, and connection status;

[0059] A graph neural network is established based on the graph structure, and the optimized wiring mode is solved by a reinforcement learning method.

[0060] In some embodiments, the reinforcement learning method includes any one or more of a proximal policy optimization algorithm and a dominant action review algorithm.

[0061] In some embodiments, the reward function in the reinforcement learning method is:

[0062]

[0063] Among them, E is the electrical part, E=μ·η inv -σ·ReLU(R actial -R max ), η inv is the conversion efficiency of a single inverter, ReLU is the activation function, R actual is the actual capacity ratio of the inverter, R max The maximum capacity ratio set for the inverter; L ij is the length of the connection; I cross (P) is the cross detection function; α, β, γ, μ and σ are proportional coefficients.

[0064] The photovoltaic assembly installation optimization device according to the embodiment of the present application includes:

[0065] a determination module, configured to determine a target construction area where the photovoltaic modules can be arranged based on an aerial view of the construction site of the photovoltaic modules, and convert the target construction area into a planar construction map;

[0066] A calculation module, configured to calculate the shadow occlusion area caused by obstacles in the planar construction diagram;

[0067] An optimization module is used to determine an initial arrangement of the photovoltaic components according to the plan construction drawing, and adjust the initial arrangement based on the shadow-blocking area to obtain an optimized arrangement.

[0068] The electronic device of the embodiment of the present application includes one or more processors and a memory, the memory stores a computer program, and when the computer program is executed by the processor, the installation optimization method of any of the above embodiments is implemented.

[0069] The computer-readable storage medium of the embodiment of the present application stores a computer program thereon, and when the program is executed by a processor, the installation optimization method of any of the above embodiments is implemented.

[0070] The photovoltaic module installation optimization method, installation optimization device, electronic device, and computer-readable storage medium of the present application first determine the target construction area where the photovoltaic modules can be arranged based on an aerial view of the photovoltaic module construction site, and convert the target construction area into a planar construction map. The target construction area is then converted into a planar construction map. The shadow obstruction area created by obstacles in the planar construction map is then calculated. Finally, the initial arrangement of the photovoltaic modules is determined based on the planar construction map, and the initial arrangement is adjusted based on the shadow obstruction area to obtain an optimized arrangement. This method optimizes the installation of photovoltaic modules and addresses the current issues of difficult 3D modeling and high manual design costs in photovoltaic module installation.

[0071] Additional aspects and advantages of the embodiments of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through practice of the embodiments of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0073] Figure 1 is a schematic flow chart of a method for optimizing the installation of photovoltaic modules according to certain embodiments of the present application;

[0074] Figure 2 is a schematic diagram of the working process of the photovoltaic assembly installation optimization method according to certain embodiments of the present application;

[0075] Figure 3 is a schematic flow chart of a method for optimizing the installation of photovoltaic modules according to certain embodiments of the present application;

[0076] Figure 4 is a schematic flow chart of a method for optimizing the installation of photovoltaic modules according to certain embodiments of the present application;

[0077] Figure 5 is a schematic flow chart of a method for optimizing the installation of photovoltaic modules according to certain embodiments of the present application;

[0078] Figure 6 is a schematic diagram of generating a ridge line of a target construction area by using a straight skeleton algorithm in certain embodiments of the present application;

[0079] Figure 7 is a schematic flow chart of a method for optimizing the installation of photovoltaic modules according to certain embodiments of the present application;

[0080] Figure 8 is a schematic diagram of generating ridge lines for continuous gables and single gables, respectively, according to certain embodiments of the present application;

[0081] Figure 9 is a schematic flow chart of a method for optimizing the installation of photovoltaic modules according to certain embodiments of the present application;

[0082] Figure 10 This is a schematic diagram of a plan construction drawing obtained by unfolding the slope area connected by the highest ridge line in certain embodiments of the present application;

[0083] Figure 11 is a schematic diagram of a two-sided sloping roof in certain embodiments of the present application unfolded and flattened into a planar construction diagram;

[0084] Figure 12 is a schematic diagram of a four-sided sloping roof before deployment according to certain embodiments of the present application;

[0085] Figure 13 is a schematic diagram of a four-sided sloping roof unfolded and flattened into a planar construction diagram according to certain embodiments of the present application;

[0086] Figure 14 is a schematic diagram of a four-sided sloping roof unfolded and flattened into a planar construction diagram according to certain embodiments of the present application;

[0087] Figure 15 is a schematic flow chart of a method for optimizing the installation of photovoltaic modules according to certain embodiments of the present application;

[0088] Figure 16 A side view of a shadowed area created by an obstacle according to some embodiments of the present application;

[0089] Figure 17A top view of a shadowed area generated by an obstacle according to some embodiments of the present application;

[0090] Figure 18 is a schematic flow chart of a method for optimizing the installation of photovoltaic modules according to certain embodiments of the present application;

[0091] Figure 19 A schematic diagram of a grid area corresponding to a photovoltaic assembly constructed for certain embodiments of the present application;

[0092] Figure 20 A schematic diagram of a grid area overlaid on a planar construction drawing according to certain embodiments of the present application;

[0093] Figure 21 is a schematic flow chart of a method for optimizing the installation of photovoltaic modules according to certain embodiments of the present application;

[0094] Figure 22 A schematic diagram of a rotation of a coordinate system of a grid area in certain embodiments of the present application;

[0095] Figure 23 A schematic diagram of grid expansion of a grid area according to some embodiments of the present application;

[0096] Figure 24 A schematic diagram of adjusting a grid area according to a component occlusion area in certain embodiments of the present application;

[0097] Figure 25 is a schematic flow chart of a method for optimizing the installation of photovoltaic modules according to certain embodiments of the present application;

[0098] Figure 26 is a schematic flow chart of a method for optimizing the installation of photovoltaic modules according to certain embodiments of the present application;

[0099] Figure 27 is a schematic diagram of a predetermined wiring method of a photovoltaic module according to certain embodiments of the present application;

[0100] Figure 28 is a schematic diagram of a predetermined wiring method of a photovoltaic module according to certain embodiments of the present application;

[0101] Figure 29 is a schematic diagram of a predetermined wiring method of a photovoltaic module according to certain embodiments of the present application;

[0102] Figure 30 is a schematic flow chart of a method for optimizing the installation of photovoltaic modules according to certain embodiments of the present application;

[0103] Figure 31 is a schematic diagram of a module of a photovoltaic assembly installation optimization device according to certain embodiments of the present application;

[0104] Figure 32 is a schematic diagram of a module of an electronic device according to some embodiments of the present application;

[0105] Figure 33 This is a schematic diagram of the connection status between a computer-readable storage medium and a processor in certain embodiments of the present application.

[0106] Description of reference numerals:

[0107] Photovoltaic module installation optimization device 100, determination module 10, calculation module 20, optimization module 30;

[0108] Electronic device 200, processor 210, memory 220;

[0109] Computer-readable storage medium 300 , computer program 310 , processor 320 . DETAILED DESCRIPTION

[0110] The following further describes the embodiments of the present application in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements or elements having the same or similar functions. Furthermore, the embodiments of the present application described below in conjunction with the accompanying drawings are exemplary and are intended only to explain the embodiments of the present application and are not to be construed as limiting the present application.

[0111] Currently, the bottlenecks in developing renewable energy sources like photovoltaics lie in installation space and capacity, with installation space being a particularly prominent issue. Photovoltaic panels require ample space for installation and have strict environmental requirements, avoiding shadows. Statistics show that over 10 billion square meters of rooftops and other vertical surfaces in my country's urban and rural areas that receive sufficient sunlight could generate 2 trillion kilowatt-hours of electricity annually, representing approximately 28% of my country's total annual power generation. Furthermore, numerous ground-mounted power stations exist. Optimizing the installation of photovoltaic panels has become a pressing issue.

[0112] In the related art, the installation design schemes for photovoltaic modules can be mainly divided into two schemes: manual design based on Computer Aided Design (CAD) software and automatic layout based on three-dimensional (3D) fine modeling. Among them, the manual design scheme has the following disadvantages: the cost of layout design is high, which requires a lot of time and manpower costs, and the measurement is not accurate, which is easy to introduce human deviations and affect the laying accuracy of photovoltaic modules. The 3D fine modeling scheme faces implementation difficulties: it currently relies on drone patrol scanning or manual modeling. If drone patrol scanning is used, field scanning and exploration are required, and the subsequent modeling processing needs to rely on commercial software, which is expensive; if manual modeling is used, there are problems with slow speed and difficulty in ensuring accuracy. Both of these schemes are difficult to meet the requirements of large-scale photovoltaic power station construction for automated and intelligent laying.

[0113] In view of this, the embodiment of the present application provides an installation optimization method for photovoltaic modules, which is based on image processing methods and two-dimensional CAD modeling to achieve automatic arrangement and wiring optimization of photovoltaic modules, making up for the problems of difficult 3D modeling and high cost of manual design.

[0114] See also Figure 1 and Figure 2 The photovoltaic module installation optimization method of the embodiment of the present application includes:

[0115] 010: Based on the aerial view of the PV module construction site, determine the target construction area where the PV modules can be arranged, and convert the target construction area into a planar construction map;

[0116] 020: Calculate the shadow occlusion area caused by obstacles in the plane construction map;

[0117] 030: Determine the initial layout of photovoltaic panels based on the plan construction plan, and adjust the initial layout based on the shadow occlusion area to obtain the optimized layout.

[0118] The photovoltaic module installation optimization method of the present application embodiment first determines the target construction area where the photovoltaic modules can be arranged based on an aerial view of the photovoltaic module construction site, and converts the target construction area into a planar construction map. The target construction area is then calculated to determine the shadow obstruction area caused by obstacles in the planar construction map. Finally, the initial arrangement of the photovoltaic modules is determined based on the planar construction map, and the initial arrangement is adjusted based on the shadow obstruction area to obtain an optimized arrangement. This method can achieve photovoltaic module installation optimization and solve the current problems of difficult 3D modeling and high manual design costs in photovoltaic module installation.

[0119] Specifically, in 010, a target construction area for PV module layout is determined based on an aerial view of the PV module construction site, and the target construction area is converted into a plan view. Determining the target construction area based on the aerial view avoids the high cost and time consumption of manual on-site surveys. Converting the target construction area into a plan view facilitates the subsequent determination of the PV module layout based on the plan view, eliminating the need for complex 3D modeling and improving layout design efficiency.

[0120] In 020, the shadow occlusion area caused by obstacles in the plane construction map is calculated.

[0121] In 030, the initial layout of the PV panels is determined based on the plan construction plan and adjusted based on the shadowed areas to obtain an optimized layout. By calculating the shadowed areas and adjusting the PV panel layout accordingly, the PV panel layout is more scientific and reasonable, preventing PV panels from being blocked by shadows created by obstacles. This allows the PV panels to receive the most sunlight, ensuring efficient power generation, thereby improving the power generation efficiency and economic benefits of the entire PV power station.

[0122] See also Figure 3 In some embodiments, determining a target construction area where photovoltaic modules can be arranged based on an aerial view of the construction site of the photovoltaic modules and converting the target construction area into a planar construction diagram (i.e., 010) includes:

[0123] 011: Obtain an aerial view of the photovoltaic module construction site;

[0124] 012: Based on the aerial photography scale, image processing algorithms are used to extract the contours of the aerial top view to determine the target construction area where photovoltaic modules can be arranged;

[0125] 013: According to the inclination of the target construction area, the target construction area is converted into a plane construction map.

[0126] Specifically, aerial views can be obtained through a variety of channels, such as through map providers like Baidu Maps, Amap, or Google Maps. Different map providers offer aerial views at varying resolutions and update frequencies, allowing the optimal image data source to be selected based on the specific characteristics of the construction site. Of course, there are many other channels for obtaining aerial views. For example, aerial views can also be obtained through natural resources departments or professional surveying and mapping agencies.

[0127] When obtaining an aerial view through a map supply platform, first, the user can search for the construction site of the photovoltaic module on the map supply platform, such as entering the address or other keywords, to quickly locate the construction site and enter the map page of the construction site. Then, a complete aerial view of the construction site is captured on the map page. The above method is more suitable for cases where the construction site is an urban or rural building. If the construction site is a ground power station, the aerial view can also be manually marked on the map page, such as by drawing a polygon or rectangular box to circle the required aerial view, thereby obtaining a more accurate aerial view.

[0128] After obtaining the aerial top view, the aerial top view is contour extracted using an image processing algorithm based on the aerial scale to determine the target construction area where photovoltaic modules can be arranged. The aerial scale represents the proportional relationship between a unit length on the aerial top view and the actual length. In one example, the aerial top view can be first converted into an actual top view with actual size based on the aerial scale, and then the contour of the actual top view can be extracted using an image processing algorithm to determine the target construction area where photovoltaic modules can be arranged. In another example, the aerial top view can be first contour extracted using an image processing algorithm to obtain an image construction area where photovoltaic modules can be arranged, and then the image construction area can be converted into a target construction area with actual size based on the aerial scale.

[0129] After determining the target construction area, the target construction area can be converted into a planar construction diagram based on its inclination. The inclination of the target construction area can include whether the target construction area is a flat structure or a sloped structure. For example, a flat structure is a flat roof or a flat ground power station, while an sloped structure is an example of a sloped roof. Depending on the inclination of the target construction area, different methods can be used to convert the target construction area into a planar construction diagram.

[0130] In the embodiments of the present application, an image processing algorithm is used to automatically extract the outline of an aerial top-view image, which not only accurately determines the target construction area where photovoltaic modules can be arranged, but also reduces the complexity and time cost of manual operations. Furthermore, based on the inclination of the target construction area, the target construction area is converted into a planar construction map. On the one hand, this makes the embodiments of the present application applicable to target construction areas with various inclinations, such as flat roofs, flat ground power stations, and sloping roofs. On the other hand, converting the target construction area into a planar construction map can provide a more accurate two-dimensional foundation for the subsequent layout design of photovoltaic modules, eliminating design errors caused by inclination factors.

[0131] See also Figure 4 In certain embodiments, an image processing algorithm is used to extract contours from an aerial view to determine a target construction area where photovoltaic modules can be arranged, including:

[0132] 0121: Perform contrast enhancement processing on the aerial view to obtain an enhanced image;

[0133] 0122: Perform Gaussian filtering on the enhanced image to obtain a first denoised image;

[0134] 0123: Perform Canning edge detection on the first denoised image to obtain a second denoised image;

[0135] 0124: Perform Hough transform shape fitting processing on the second denoised image to extract the target contour shape;

[0136] 0125: Determine the target construction area based on the target outline shape.

[0137] Specifically, in 0121, contrast enhancement processing is performed on the aerial view to obtain an enhanced image.

[0138] It is understandable that the aerial view is affected by the weather and time of the aerial photography. The aerial view captured from the map supply platform may have low contrast and unclear details. The contrast and clarity of the image can be improved through contrast enhancement processing. In the embodiment of the present application, the contrast enhancement processing can adopt the limited contrast adaptive histogram equalization (CLAHE) algorithm. This algorithm can increase the contrast of the key areas of the image as much as possible while avoiding the amplification of noise through block processing, contrast limitation, histogram equalization and bilinear interpolation. The specific process example is as follows:

[0139] (1) Block processing: ① Divide the aerial view into multiple blocks. The size of each block can be adjusted according to the image resolution and processing requirements, for example, 8×8 or 16×16 pixels. ② Calculate the grayscale histogram distribution for each block. A grayscale histogram is a graph that shows the number of pixels at each grayscale level in an image, which can be used to show the brightness distribution of the image.

[0140] Through block processing, the aerial view is divided into multiple blocks so that each block can be processed separately. This local processing method can better adapt to the lighting conditions and contrast differences in different areas of the image.

[0141] (2) Contrast limitation: ① Calculate the histogram average of each block. ② Set the pixel number threshold T for each grayscale level in the histogram. For example, it can be 3 times the histogram average (T = 3 × average pixel value). The pixel number threshold T is used to limit the number of pixels at each grayscale level in the histogram. ③ For each grayscale level, if its pixel number exceeds the threshold T, the excess pixels are clipped. ④ The clipped pixels are evenly distributed to all grayscale levels to keep the total number of pixels in the histogram unchanged.

[0142] By limiting contrast, the number of pixels at each gray level in the histogram can be limited, avoiding the impact of high contrast in local areas on the overall image and preventing noise amplification.

[0143] (3) Histogram equalization: The cropped histogram is equalized and a grayscale mapping function is generated to map the original grayscale value to a new grayscale range.

[0144] Through histogram equalization, the grayscale distribution of the image can be made more uniform while avoiding excessive enhancement of local areas.

[0145] (4) Bilinear interpolation: ① Apply a bilinear interpolation algorithm to the pixel values ​​of each block. Bilinear interpolation is an interpolation method based on the four nearest neighbor points. It determines the pixel value of the target point by calculating the weighted average of the four neighboring points. ② After interpolation, all processed blocks are combined into a complete image to obtain the final enhanced image.

[0146] By interpolating the pixel values ​​of each block through bilinear interpolation, the grayscale value mutation between adjacent blocks can be avoided, thereby generating a smooth transition effect.

[0147] In 0122, Gaussian filtering is performed on the enhanced image to obtain a first denoised image.

[0148] Gaussian filtering is an image smoothing technique used to remove noise from an image while preserving as much detail as possible. Gaussian filtering uses a Gaussian convolution kernel to convolve an image. The Gaussian convolution kernel can be a two-dimensional Gaussian function. In one example, the Gaussian convolution kernel template with dimensions (-2ω-1, 2ω+1) is:

[0149]

[0150] In G Where u represents the row position, v represents the column position, and σ is the standard deviation of the Gaussian function. u,v∈-ω,-ω+1,…,ω-1, ω and S are normalization constants. The Gaussian convolution kernel is convolved on the enhanced image, and the value of the current pixel is updated by calculating the weighted average of the surrounding pixels, which can effectively remove the noise in the image.

[0151] In step 0123, Canny edge detection is performed on the first denoised image to obtain a second denoised image.

[0152] Canny edge detection is a classic edge detection algorithm used to extract clear edge information from an image. In the embodiments of the present application, Canny edge detection may include the following process:

[0153] ① Gradient calculation: Use the Sobel gradient operator to perform convolution along the horizontal and vertical directions of the image to solve the image gradient data. The commonly used 3×3 Sobel operator is:

[0154] Horizontal direction:

[0155] Vertical direction:

[0156] ② Non-maximum suppression: Non-maximum suppression is performed on the results of gradient solution. For the gradient values ​​in the same direction, only the largest gradient value is retained, and all other smaller values ​​are suppressed to zero to make the boundary clear.

[0157] ③Dual threshold detection: Set dual thresholds of upper and lower thresholds. The dual threshold data can be manually adjusted by the user. Only pixel data within the dual threshold range will be retained, otherwise the pixel data will be cleared to further suppress noise.

[0158] In step 0124, a Generalized Hough Transform (GHT) shape fitting process is performed on the second denoised image to extract the target contour shape.

[0159] The Hough transform is used to detect specific geometric shapes, such as lines and circles, from an image and is suitable for extracting complex contours. It is understood that the geometry of urban and rural building roofs or ground-mounted power stations can generally be considered a combination of one or more conventional geometric shapes. In the embodiments of this application, the Hough transform is used to fit geometric shapes such as line segments and circles to extract the target contour shape in the image where photovoltaic modules can be arranged.

[0160] In 0125, the target construction area is determined based on the target contour shape.

[0161] In one example, the target contour shape is a roof contour shape, and the target construction area is a roof area.

[0162] In this application, by sequentially performing contrast enhancement, Gaussian filtering, Canning edge detection, and Hough transform shape fitting, the target contour shape can be accurately extracted, thereby determining the target construction area for photovoltaic module layout. This entire process is automated through image processing algorithms, reducing manual labor, improving design efficiency, and providing greater accuracy and reliability.

[0163] See also Figure 5 In some embodiments, according to the inclination of the target construction area, converting the target construction area into a planar construction map (i.e., 013) includes:

[0164] 0131: When the target construction area is a plane structure, the target construction area is directly used as the plane construction drawing;

[0165] 0132: When the target construction area is a tilted structure, the ridge line of the target construction area is generated using the straight skeleton algorithm;

[0166] 0133: Divide the ridge line into multiple line segments according to its vertices;

[0167] 0134: Determine the highest ridge line among multiple line segments based on the vertex positions of the multiple line segments;

[0168] 0135: Expand each slope area connected by the highest ridge line to obtain a plan construction drawing.

[0169] Specifically, the planar structure is, for example, a flat roof or a flat ground power station, and the inclined structure is, for example, a sloped roof.

[0170] When the target construction area is a flat roof or flat ground power station, the target construction area is already a plane, and the target construction area can be directly used as a plane construction map for subsequent determination of the layout of photovoltaic modules.

[0171] When the target construction area is a sloping roof, the target construction area is a slope, and the target construction area needs to be unfolded and flattened into a plane construction drawing before it can be used to subsequently determine the arrangement of photovoltaic modules.

[0172] According to research, it is found that the extraction of the ridge line of the target construction area by visual recognition is susceptible to noise interference. Therefore, the embodiment of the present application automatically generates the ridge line of the target construction area through the straight skeleton algorithm, that is, the ridge line inside the roof area. The straight skeleton algorithm is an algorithm for calculating the internal structure of a two-dimensional simple polygon. It shrinks the edges of the polygon inward uniformly according to a predetermined wavefront until it shrinks to a point. The trajectory formed by each vertex of the polygon during the shrinking process is the desired ridge line result. The shrinking process is as follows: Figure 6 shown.

[0173] After generating the ridge line, since the target construction area is a top view of a sloping roof, but the actual area for laying photovoltaic modules is the area of ​​the roof slope, the shape of the actual roof slope where photovoltaic modules can be laid should be solved based on the top view information so that photovoltaic modules can be laid on the actual shape.

[0174] First, the ridge line can be divided into multiple line segments based on its vertices. These line segments are then classified into two categories based on the location of their end vertices: if both end vertices of a line segment are within the roof area's outline, or both end vertices are on the roof area's outline, the line segment is considered the highest ridge line. If one end vertex of a line segment is within the roof area's outline, while the other end vertex is on the roof area's outline, the line segment is considered a non-highest ridge line. This allows all line segments to be categorized. Next, each sloped area connected by the highest ridge line is expanded to produce a plan view.

[0175] See also Figure 7 In some embodiments, the target construction area includes polygon vertices and polygon contour lines. Generating the ridge line of the target construction area using a straight skeleton algorithm includes:

[0176] 01321: Determine whether the polygon outline connected by the polygon vertices is a gable edge to determine the type of the polygon vertex;

[0177] 01322: When the polygon vertex type is free vertex, the straight skeleton algorithm is used to control the free vertex to shrink inward to determine the free intersection vertex;

[0178] 01323: When the polygon vertex type is partially free vertex, control the movement of some free vertices along the gable edge to determine the gable intersection vertex;

[0179] 01324: Determine the ridge line based on the free intersection vertex and the gable intersection vertex.

[0180] Specifically, the target construction area includes polygon vertices and polygon contour lines. The above embodiment explains the basic principle of generating the ridge line of the target construction area by the straight skeleton algorithm. The embodiment of the present application provides a solution for determining the ridge line for the case where there is a gable side (called "wife side" in Japanese) where the surface corresponding to the contour line is perpendicular to the roof plane. The gable surface is always perpendicular to the roof plane, and the corresponding edges and vertices do not shrink inward in the straight skeleton algorithm. In order to facilitate computer programming processing, the ridge line can be generated by moving the relevant vertices.

[0181] Depending on the number of contour lines on the gable side, a variety of scenarios will occur. Two typical scenarios are as follows: Figure 8As shown. The polygon vertices in the figure can be divided into three categories according to the type of polygon contour lines they are connected to: If the two polygon contour lines connected by the polygon vertex are not gable edges, then the polygon vertex is a free vertex and can normally participate in the inward contraction of the straight skeleton algorithm; If one of the two polygon contour lines connected by the polygon vertex is a gable edge, then the polygon vertex is a partially free vertex. Partially free vertices can only move along the gable edge when the straight skeleton algorithm is used; In addition, if the two polygon contour lines connected by the polygon vertex are both gable edges, then the polygon vertex is a non-free vertex. Non-free vertices do not participate in the contraction process during the straight skeleton algorithm and cannot move along the gable edge. They are fixed points. Furthermore, the vertex generated by the collision of free vertices during the contraction process is a free intersection vertex; the intersection point generated by the movement of two partially free vertices on the gable edge during the contraction process of partially free vertices is a gable intersection vertex.

[0182] For the case of continuous gables, Figure 8 As shown in (a), first, the free vertices move normally to generate free intersection vertices, but some free vertices do not move; after the free vertices move, some free vertices move along the gable edge to generate gable intersection vertices; finally, the gable intersection vertex is connected to the nearest free intersection vertex. For the case of a single gable, Figure 8 As shown in (b), the gable intersection vertex is generated first. It then moves along the perpendicular line of the wife's side toward the interior of the roof area's contour line, colliding with the free vertex to generate a normal intersection line. It should be noted that a continuous gable is multiple single gables connected together, while a single gable is a gable that is not connected to other gables. Other gable types can be considered combinations of the aforementioned continuous gable and single gable types and are handled accordingly.

[0183] See also Figure 9 In some embodiments, the slope area is a trapezoidal slope. The slope areas connected by the highest ridge line are expanded to obtain a plan construction drawing (i.e., 0135), including:

[0184] 01351: Get the top view length of the waist of the trapezoid slope;

[0185] 01352: Determine the unfolded length of the waist of the trapezoidal slope based on the slope angle corresponding to the trapezoidal slope or the ridge line height corresponding to the highest ridge line, as well as the top-down length;

[0186] 01353: Keep the upper base of the trapezoidal slope stationary, control the parallel movement of the lower base, and calculate the current length of the waist in real time;

[0187] 01354: When the current length is equal to the expanded length, the plane construction drawing is determined according to the upper base, lower base and waist of the trapezoidal slope.

[0188] Specifically, the embodiment of the present application provides a solution for determining the plane construction drawing for the typical four-sided trapezoidal slope. The original top view of the trapezoidal slope and the actual shape after the plane is unfolded are as follows: Figure 10 (a) shows ABCD and ABC'D' respectively. Now, we take the solution of A'C' after the AC side is unfolded as an example to illustrate the principle of slope unfolding. The length of the highest ridge line AB and the corresponding contour line CD remain unchanged before and after unfolding. Let the length of line segment AC be the top view length a, which can be read from the original top view after generating the ridge line. The side view of the ridge line is as follows Figure 10 As shown in (b), let the length of A'C' to be determined be the expanded length c, and the user manually inputs the inclination angle θ or the height h of the highest ridge line from the plane, then or After determining the length of the expanded A'C' side, Figure 10 As shown in (a), the highest ridgeline AB remains stationary, while CD translates downward parallel to the original ridgeline. The distance between points A and C' is continuously calculated until the current length of AC' matches the desired unfolded length c. At this point, the new ABC'D' represents the desired roof slope. This scheme works for any simple polygon and allows you to unfold each sloping area connected by the highest ridgeline to obtain a plan view.

[0189] In certain embodiments, when the target construction area is an inclined structure, the target construction area may be expanded and flattened into a planar construction diagram through geometric substitution based on the geometric characteristic parameters of the target construction area.

[0190] At this time, first, determine the geometric characteristic parameters of the target construction area, such as length parameters, width parameters, height parameters, endpoint coordinates, etc.; then, based on the above geometric characteristic parameters, use geometric substitution to flatten the target construction area into a plane construction drawing, so as to more accurately design the layout of photovoltaic modules and avoid design errors caused by tilt factors.

[0191] The following uses two-sided / single-sided sloping roofs and four-sided sloping roofs as examples to describe the process of flattening the target construction area into a planar construction diagram through geometric substitution based on the geometric feature parameters of the target construction area.

[0192] (1) Double-sided / single-sided sloping roof

[0193] See also Figure 11For a two-sided sloping roof, the size of the roof in the top view can be determined by the target construction area, that is, the length a and width b of the target construction area. The user specifies the height h of the ridge line relative to the eaves, and sets the relevant endpoint coordinates based on the length a and width b. The six endpoint coordinates of the ridge line and the eaves before unfolding and flattening are as follows: Figure 11 As shown in the left picture, after unfolding and flattening, Figure 11 As shown in the right figure. From the height h, we can know that the vertical coordinate z3=z4=z1+h=z2+h=z5+h=z6+h. From the symmetry of the roof, we can know that By geometric substitution, the coordinates (x'1, y'1) in the expanded diagram can be deduced as:

[0194]

[0195] The coordinates of the diagonal (x'6, y'6) can be solved by substitution in a similar way. Then, by taking advantage of the fact that the short side length remains unchanged after expansion and the long sides are parallel, the coordinates of all endpoints in the expanded diagram can be calculated.

[0196] (2) Four-sided sloping roof

[0197] See also Figures 12 to 14 For a sloping roof, let the straight line PQ on which the ridge line lies be l1; A, E, and F are points on the eaves. After unfolding, E changes to E and E'. Draw a straight line l2 through point E that intersects l1 at point P. Draw a straight line through point E' that is parallel to the eaves line EF and intersects l2 at point C. On the other side, points Q and D can be obtained by the same logic. At this point, the unfolding of the sloping roof can be equated to the unfolding of the two sloping roofs mentioned above, that is, the rectangle CDPQ is unfolded into the rectangle EFPQ. Then, find the coordinates of point P (x p ,y p ) After that, the coordinates of point E can be obtained using the formula derived from the two sloping roofs mentioned above, and then the other coordinates can be solved. The analytical expression of line l1 is:

[0198]

[0199] The analytical expression of the straight line l2 is:

[0200]

[0201] Solve for the intersection point P of the two straight lines, then we have:

[0202]

[0203] The coordinates of point P are:

[0204]

[0205] It can be understood that the above describes two typical ways of unfolding sloping roofs into plan construction drawings. Other types of sloping roofs can be split and decomposed into combinations of these two typical roofs. More complex roof situations can also be solved by manual splitting by the user, etc., which are not limited here.

[0206] See also Figure 15 In some embodiments, calculating the shadow occlusion area (i.e., 020) caused by obstacles in the planar construction diagram includes:

[0207] 021: Determine the shadow distance corresponding to each point on the upper edge of the obstacle based on the geometric characteristic parameters of the obstacle and the solar radiation correlation parameters of the construction site at all times of the year;

[0208] 022: Determine the outer edge of the shadow created by the obstacle based on the shadow distance corresponding to each point on the upper edge;

[0209] 023: Determine the shadow occlusion area based on the outer edge of the shadow.

[0210] It is understandable that over time, the backlight side of an obstacle will cast a shadow, which can block the photovoltaic panels, causing the batteries to not generate electricity properly and even risk burning them. Therefore, the implementation method of this application requires calculating the shadow obstruction area created by the obstacle in the plan construction diagram and adjusting the layout of the photovoltaic panels based on the shadow obstruction area.

[0211] Specifically, the outline of the obstacle can be obtained through the aforementioned process of extracting the outline of the aerial view using an image processing algorithm. That is to say, while extracting the target contour shape, the outline of the obstacle can also be extracted to determine the geometric characteristic parameters of the obstacle. In an example, the geometric characteristic parameters of the obstacle may include the height parameter, width parameter, etc. of the obstacle. The solar radiation correlation parameters of the construction site at various times throughout the year may include time information, latitude information, solar altitude angle, solar azimuth angle, solar declination angle, solar hour angle, etc. Based on the solar radiation correlation coefficient of the construction site at various times throughout the year in a certain year, the shadow distance generated by the obstacle in a year can be calculated for subsequent determination of the shadow-blocking area.

[0212] See also Figure 16 and Figure 17 , take the obstacle as an example, which is an occluder with a height of dy and a width of dx. At a certain moment, the shadow length of the occluder is d, the distance between the far end of the shadow and the vertex of the occluder is r, and the sun altitude angle is α s , side view as Figure 16 As shown, from the geometric relationship we can know:

[0213]

[0214] Top view Figure 17 As shown, from the geometric relationship, Among them, γ s is the solar azimuth.

[0215] Therefore, we can get have:

[0216]

[0217] The local latitude is The solar declination angle is δ, and the solar hour angle is ω. Substituting in the spherical coordinate system, we can get:

[0218]

[0219] The sine of the sun's altitude angle is:

[0220]

[0221] Substituting into the above formula we can get:

[0222]

[0223] Therefore, the shadow distance d can be expressed as:

[0224]

[0225] set up is the proportional coefficient s, where latitude It can be directly obtained from the local geographic coordinates. The declination angle δ has the following simplified calculation formula:

[0226]

[0227] Where n is the number of days in the year, for example, January 14th corresponds to n = 14. The hour angle ω = (true solar time - 12) · 15°. Usually, the time displayed on a clock is the mean solar time, which is directly related to the sun's azimuth. When the true solar time is 12:00, the sun's azimuth is 0, while the mean solar time is not 12:00. There is an hour difference between the two, which is recorded as T s , the time difference is calculated as:

[0228] T s =0.0028-1.9587sin(b)+9.9059sin(2b)-7.0924cos(b)-0.6882cos(2b)

[0229] Here, b is the sun angle, which can be calculated using the following formula:

[0230]

[0231] Where n is defined in the same way as in the above declination angle formula, and n0 is a year-dependent parameter, which is calculated as follows:

[0232] n0=78.801+[0.2422×(year-1969)-round(0.25×(year-1969)]

[0233] Here, "round" represents rounding up, and "year" represents the current year. This value is relatively stable, with an average fluctuation of 1% over 20 years. Therefore, n0 (the first year of installation) can be used for calculation.

[0234] From the above, we can see that the proportional coefficient s only needs the accurate local time t (accurate to seconds) and the geographical latitude of the power station For actual obstructions, the shadow is mainly considered to be the upper edge of the cross section with the largest cross section area in the sun direction. Let the distance from the upper edge of the cross section to the ground be y. Establish a coordinate system, then y can be expressed as a function of the horizontal coordinate x, denoted as y = f(x). Then the shadow distance D corresponding to each point on the upper edge of the obstacle can be expressed as:

[0235]

[0236] By calculating the above formula, we can obtain the outer edge of the shadow cast by the obstacle, and based on this outer edge, we can determine the shadow-blocked area. It should be noted that if the obstacle has a complex shape (such as a through hole or window on the side), we can perform a piecewise integral calculation on the shadow distance D to determine the outer edge of the shadow cast by the obstacle.

[0237] See also Figure 18 In certain embodiments, determining an initial arrangement of photovoltaic modules based on a planar construction drawing and adjusting the initial arrangement based on shadow-blocked areas to obtain an optimized arrangement (i.e., 030) includes:

[0238] 031: Construct a grid area corresponding to the photovoltaic module according to the length parameter, width parameter, spacing parameter, predetermined reference direction and reference direction inclination angle of the photovoltaic module;

[0239] 032: Overlay the grid area on the plan to determine the initial layout of the photovoltaic panels;

[0240] 033: Adjust the initial layout according to the part of the grid area that exceeds the plane construction map, the obstacle area and the shadow area to obtain the optimized layout.

[0241] Specifically, see Figure 19 , users can pre-specify the length a and width b of the photovoltaic modules, and the upper and lower spacing d between photovoltaic modules yThe left and right distance d between the photovoltaic modules x , reference direction and reference direction inclination θ, to establish Figure 19 The grid area shown. The size of a single grid unit is (a+d x， b+d y ), the default reference direction is due south, which can be obtained from the map in the 010 process above. Of course, the reference direction can also be different from due south. Users can specify the reference direction and reference angle, or select a reference line or roof edge as a reference. Vertical grid lines in the grid area will automatically align with the reference direction.

[0242] See also Figure 20 , the grid area is covered to the greatest extent possible on the unfolded roof plan, ensuring that every part of the roof is covered by the grid. The resulting PV panel layout in the grid area is the initial layout. If the PV panel area in the grid area exceeds the roof range, or intersects with obstacles (including actual obstacles and objects that can be considered obstacles, such as maintenance passages), or shadowed areas, the PV panel area is removed. The remaining grid area is the actual area on the roof where PV panels can be arranged, thus obtaining the optimized layout.

[0243] See also Figure 21 In some embodiments, overlaying the grid area on the plan construction drawing to determine the initial arrangement of photovoltaic modules (i.e., 032) includes:

[0244] 0321: When the reference direction inclination angle is zero, overlay the grid area on the plane construction drawing and align the grid area with the plane construction drawing coordinate system to determine the initial arrangement of the PV panels;

[0245] 0322: When the reference direction inclination angle is not zero, the grid area is covered on the plane construction drawing, and then the grid area is rotated according to the reference direction inclination angle, and the grid area is expanded along the coordinate axis direction of the grid area to determine the initial arrangement of the photovoltaic modules.

[0246] Specifically, the original coordinate system of the grid area is set to XY. When the reference direction angle is zero, the rotation angle is 0. x , b+d y )'s rectangular grid cells are aligned with the origin at their upper left corners and arranged without spacing.

[0247] See also Figure 22In practical applications, due to the artificially specified reference direction or reference direction inclination, the actual coordinate system will rotate relative to the original coordinate system, and the reference direction inclination is not zero. Let the rotation angle to be rotated be θ. At this time, the coordinate system used for arrangement is rotated and transformed. The rotation transformation matrix is Then for any point P(x,y) in the original coordinate system, the rotated coordinate P'(x',y'):

[0248]

[0249] At the same time, each photovoltaic module rotates around its own geometric center to ensure that the photovoltaic module always faces the reference direction. Figure 22 As shown; after the rotation, the originally densely arranged roofs have gaps in the arrangement. At this time, you can select the reference coordinates in the new coordinate system X'Y'. For example, here we select the coordinates of the photovoltaic modules in the leftmost column as the reference, and expand the grid appropriately in the y' and x' directions to ensure that the roof is still densely arranged after the rotation. The effect is as follows Figure 23 shown.

[0250] In certain embodiments, before overlaying the grid area on the plan construction drawing to determine the initial arrangement of the photovoltaic components (i.e., 032), the installation optimization method further includes:

[0251] When PV modules are installed vertically, calculate the shading area caused by the PV modules in the plan construction diagram;

[0252] Adjust the grid area based on the component occlusion area.

[0253] Specifically, if the photovoltaic modules are installed in a slope-following manner (i.e., the photovoltaic module brackets are attached to the roof surface and the photovoltaic modules are installed along the roof slope), there is no need to calculate the shadows produced by the photovoltaic modules.

[0254] However, in actual applications, photovoltaic panels are often installed vertically, such as with single-sided vertical brackets, double-sided vertical brackets, or building-integrated photovoltaics (BIPV). In this case, the photovoltaic panels are at a certain height from the roof, and shadows will be cast over time. Therefore, when arranging in this situation, the grid area should be adjusted to correct the shadow effects of the photovoltaic panels. The calculation method for the component shading area caused by the photovoltaic panels in the plan construction diagram is the same as the calculation method for the shadow shading area caused by obstacles in the plan construction diagram. That is, the photovoltaic panels are treated as obstacles and the component shading area is calculated using the methods in 021, 022, and 023 above. We will not repeat them here.

[0255] After calculating the module shielding area, the area of ​​the module shielding area and the top view projection of the photovoltaic module itself can be combined as a new grid unit to update the grid area. Figure 24 As shown in the figure, in the updated grid area, the size of a single grid unit becomes (a'+d x ', b'+d y ').

[0256] See also Figure 25 In some embodiments, the photovoltaic module is connected to an inverter to convert direct current into alternating current, and the installation optimization method further includes:

[0257] 040: Determine the optimal wiring method for PV panels based on the inverter's conversion efficiency and the total length of cables used for PV panel wiring.

[0258] It is understandable that the direct output of photovoltaic modules is direct current (DC), while whether it is for industrial and commercial self-use or for power grid connection, it requires alternating current (AC). Therefore, in actual power station construction, photovoltaic modules need to be connected to an inverter for direct current to alternating current (DC-AC) conversion to convert it into actual usable electrical energy. The performance parameters of the inverter include the maximum input DC power P dc , output AC power P ac , minimum input DC power P dc-th and conversion efficiency η, while Among them, P L is the power loss. For economic reasons, it is generally desirable to maximize the inverter's conversion efficiency η while minimizing the total cable length L used to connect the photovoltaic modules in the final power station. The present embodiment determines an automatic wiring solution for photovoltaic modules based on considerations of the inverter's conversion efficiency and the total cable length used to connect the photovoltaic modules.

[0259] The following describes the automatic wiring schemes for photovoltaic modules in simple wiring situations and fine wiring situations respectively.

[0260] See also Figure 26 In certain embodiments, determining an optimized wiring method (i.e., 040) for photovoltaic modules based on the conversion efficiency of the inverter and the total length of the cables used for wiring the photovoltaic modules comprises:

[0261] 041: Determine the expected number of PV panels connected to each inverter based on the inverter's maximum power, maximum DC input power, capacity ratio, and minimum DC input power.

[0262] 042: Connect the photovoltaic panels using a predetermined connection method to connect the current group of photovoltaic panels;

[0263] 043: When the number of connections of the current group of photovoltaic modules reaches the expected number, the connection of the current group of photovoltaic modules is stopped, and the steps of connecting the photovoltaic modules using the predetermined connection method are repeated to connect the next group of photovoltaic modules until all photovoltaic modules are connected.

[0264] Specifically, in the case of simple wiring, only a single type of inverter and photovoltaic module is considered for the shortest wiring optimization, and the inverter only considers the maximum input DC power P ac , minimum input DC power P dc-th , photovoltaic modules only consider the lowest temperature T and power temperature coefficient γ of the installation site P and the maximum power of the photovoltaic module P m The module power P under standard test conditions (STC) can be obtained from the data sheet of the photovoltaic module. m-STC And the component temperature T under STC conditions STC However, in actual applications, the working conditions of photovoltaic modules are often different from the STC conditions, so it is necessary to m In one example, the maximum power P m Can be corrected to P m =P m-STC ×(1+γ P ×(TT STC )). Of course, the maximum power P m The correction method is not limited to this. The maximum power after correction P m Will be used for subsequent calculations.

[0265] Assume that the expected number of photovoltaic modules connected to each inverter is n, and define the inverter capacity ratio as R. The value of R can be specified by yourself, usually it can be set to 1.1~1.4, then the value of n should meet the following requirements: ① ②n·P m ≥P dc-th That is, the expected number of PV panels connected to each inverter is n, which is determined by the maximum power P of the inverter. m , input maximum DC power P dc , capacity ratio R, minimum input DC power P dc-th Determine. Among them, the maximum power of the inverter P m By the component power P m-STC , power temperature coefficient γ P , minimum temperature T, component temperature T STC Determine the capacity ratio R by the expected number n and the maximum power P m , input maximum DC power P ac Sure.

[0266] After determining the expected number n of PV panels to be connected to each inverter, the grid can be connected in series, either horizontally or vertically, starting from the upper left corner of the grid. Diagonal connections are prohibited. Once the number of connected PV panels reaches n, the current connection is disconnected, and the series connection is continued, starting with the next adjacent PV panel, until all PV panels are fully connected. In the above examples, the predetermined connection method uses a horizontal or vertical connection method. It will be understood that the predetermined connection method is not limited to this.

[0267] In some embodiments, the predetermined connection mode includes any one or more of a horizontal connection mode, a vertical connection mode, a straight connection mode, a C-shaped connection mode, and a loop connection mode.

[0268] Specifically, if Figure 27 As shown, in the horizontal connection method, photovoltaic modules are arranged in rows, and the photovoltaic modules in each row are connected in series. In the vertical connection method, photovoltaic modules are arranged in columns, and the photovoltaic modules in each column are connected in series. In the straight-line connection method, photovoltaic modules are connected in series into a long string to form a straight line. It can be understood that the straight-line connection method and the aforementioned horizontal connection method and vertical connection method may have a cross-inclusion relationship. The above-mentioned horizontal connection method, vertical connection method, and straight-line connection method can simplify wiring, reduce the use of cables, and help reduce costs. In the C-shaped connection method, photovoltaic modules are arranged in a C shape. As shown Figure 28 As shown in the figure, in the loop connection method, the photovoltaic panels are connected in a loop or leapfrog manner. The above C-shaped connection method and loop connection method are used to maximize space utilization or adapt to specific roof shapes.

[0269] In practical applications, a combination of one or more of the above connection methods can be selected (such as Figure 29 to connect PV panels to meet specific design requirements and optimize system performance.

[0270] See also Figure 26 In certain embodiments, determining an optimized connection method (i.e., 040) for photovoltaic modules based on the conversion efficiency of the inverter and the total length of the cables used for connecting the photovoltaic modules further includes:

[0271] 044: When the number of connected PV panels in the last group is less than the expected number, adjust the expected number of PV panels in a group connected to each inverter, or connect the last group of PV panels to a separate inverter.

[0272] Specifically, for the connection methods in the above 041-043, there may be a string of strings with the number of photovoltaic modules being m at the end, but m < n. At this time, different strategies can be adopted for optimization according to requirements. One of the optimization strategies can be: Suppose that when connecting to the last m photovoltaic modules, i strings have been connected. If the inverter is allowed to exceed the capacity ratio R appropriately, when m ≤ i, then at this time, the wiring is readjusted so that each of the previously connected m strings is connected to one more photovoltaic module, that is, the expected number of a set of photovoltaic modules connected to each inverter is adjusted to n + 1; when m > i, then optimization is not possible, and an inverter needs to be set up separately for these m photovoltaic modules. And if the inverter is not allowed to exceed the capacity ratio R, then regardless of the number m of the remaining photovoltaic modules, an inverter needs to be set up separately.

[0273] Please refer to Figure 30 , in some embodiments, based on the conversion efficiency of the inverter and the total length of the cables used for the wiring of the photovoltaic modules, an optimized wiring method for the photovoltaic modules (i.e., 040) is determined, including:

[0274] 045: Determine the graph structure according to the set of component features and the edge set of the photovoltaic modules, and the component features include position coordinates, electrical parameters, and connection status;

[0275] 046: Establish a graph neural network according to the graph structure, and solve the optimized wiring method through a reinforcement learning method.

[0276] Specifically, the above simple wiring situation only considers single-type inverters and photovoltaic modules, but in actual applications, there may be multiple different types of inverters and photovoltaic modules. When performing automatic wiring, it is necessary to take into account that the conversion efficiency η of the inverter is the highest and the total length L of the cables used for the wiring of the photovoltaic modules is the shortest. At the same time, there may be multiple topological structures for the grid, and it is difficult to directly solve the wiring result. Therefore, the embodiments of the present application adopt a solution based on a graph neural network and reinforcement learning.

[0277] Let the set of component features of the photovoltaic modules be where p i is a single component feature, including the position coordinates (x i , y i ) of the photovoltaic module, the electrical parameter P mi and the connection status s i ∈ {0, 1}. Then the grid area can be represented as the graph structure G = {P, ε}, where P is the set of component features of the photovoltaic modules, and ε is the edge set, defined as According to this graph structure, a graph neural network (GNN) is established, and the optimal wiring method can be automatically solved using a reinforcement learning method.

[0278] In some embodiments, the reinforcement learning method includes any one or more of a proximal policy optimization algorithm and a dominant action-critic algorithm.

[0279] Specifically, Proximal Policy Optimization (PPO) is a policy gradient method that improves the performance of an agent by optimizing its policy function. The core idea of ​​PPO is to limit the magnitude of each update when updating the policy, thereby ensuring the stability and reliability of the policy update.

[0280] The Advantage Actor-Critic (A2C) algorithm is a reinforcement learning algorithm that combines policy gradient and value estimation. It maintains the policy function and value function and uses the advantage function to update the policy, enabling the agent to learn the optimal policy more effectively.

[0281] PPO is suitable for scenarios requiring high stability in policy updates, while A2C is suitable for scenarios requiring fast convergence and efficient sample utilization. In practical applications, you can choose the appropriate algorithm based on specific needs, or combine the two to find the optimal wiring method.

[0282] In some embodiments, the reward function in the reinforcement learning method is:

[0283]

[0284] Among them, E is the electrical part, E=μ·η inv -σ·ReLU(R actual -R max ), η inv is the conversion efficiency of a single inverter, ReLU is the activation function, R actual is the actual capacity ratio of the inverter, R max The maximum capacity ratio set for the inverter; L ij is the length of the connection; I cross (P) is the cross detection function; α, β, γ, μ and σ are proportional coefficients.

[0285] Specifically, the length of the connection can be calculated as I cross (P) can be calculated by taking the corresponding direction vector for two connecting line segments Perform cross product, the angle between the two vectors is θ, then the two line segments For a system with N line segments, the overall intersection detection value is In the above formula, the proportional coefficients α, β, γ, μ, and σ can be set manually according to user needs to adjust the proportion of relevant items in the reward function.

[0286] In the implementation mode of the present application, by defining a reward function and comprehensively considering various aspects such as the inverter's conversion efficiency, capacity ratio, connection length and cross detection, it is possible to clarify the optimization goals and balance the relationship between the goals in the process of solving the optimal wiring method, so that the algorithm can select the wiring method that can maximize the inverter's conversion efficiency and the shortest total cable length, thereby reducing costs.

[0287] See also Figure 2 The following describes a specific application example of the photovoltaic module installation optimization method according to the embodiment of the present application. The process of automatic two-dimensional arrangement and automatic wiring of a photovoltaic power station includes the following steps:

[0288] In steps S101-S103, the user searches for a PV panel construction site on an online map provider, accesses the site's map page, and then captures an aerial view from that page. An image processing algorithm is then used to extract the outlines of the aerial view, generating an image of the construction area where the PV panels can be deployed. This image of the construction area is then converted to the actual size of the target construction area based on the aerial scale. If the target construction area is a sloped roof, the target construction area is flattened into a planar construction map using geometric substitution based on the target construction area's geometric characteristic parameters.

[0289] S104: The user inputs the latitude and longitude of the construction site, reference direction and other related information for subsequent automatic layout and automatic wiring.

[0290] S105: Calculate the shadow area caused by the obstacle.

[0291] S106-S108, the user selects the installation method of the photovoltaic modules. If it is a slope-following installation method, the shadows between the photovoltaic modules can be ignored; if it is a vertical bracket installation method, the module shading area generated by each photovoltaic module needs to be calculated, and the grid area of ​​the automatic module arrangement is updated according to the module shading area.

[0292] S109-S111, make a judgment based on the intersection of the grid area with the obstacle area and the shadow area. If the photovoltaic module area in the grid area intersects with the obstacle area or the shadow area, then remove the photovoltaic module area and mark it as a place where photovoltaic modules cannot be arranged; if there is no obstacle area or shadow area, then photovoltaic modules can be arranged normally.

[0293] S112, after arranging the photovoltaic modules, the user selects which wiring method to use (such as horizontal connection method, vertical connection method, straight connection method, C-shaped connection method, loop connection method, etc.), and finally automatically generates the best wiring plan based on the electrical parameters of the inverter and photovoltaic modules.

[0294] See also Figure 31 The present application also provides an installation optimization device 100 for photovoltaic modules. The installation optimization device 100 includes a determination module 10, a calculation module 20, and an optimization module 30. The determination module 10 is configured to determine a target construction area where photovoltaic modules can be arranged based on an aerial view of the photovoltaic module construction site, and to convert the target construction area into a planar construction diagram. The calculation module 20 is configured to calculate the shadow obstruction area caused by obstacles in the planar construction diagram. The optimization module 30 is configured to determine an initial arrangement of photovoltaic modules based on the planar construction diagram, and to adjust the initial arrangement based on the shadow obstruction area to obtain an optimized arrangement.

[0295] In some embodiments, the determination module 10 is specifically used to: obtain an aerial top view of the construction site of the photovoltaic module; based on the aerial scale, use an image processing algorithm to extract the contours of the aerial top view to determine the target construction area where the photovoltaic modules can be arranged; and convert the target construction area into a plane construction map according to the inclination of the target construction area.

[0296] In some embodiments, the determination module 10 is specifically used to: perform contrast enhancement processing on the aerial top view to obtain an enhanced image; perform Gaussian filtering on the enhanced image to obtain a first denoised image; perform Canning edge detection on the first denoised image to obtain a second denoised image; perform Hough transform shape fitting on the second denoised image to extract the target contour shape; and determine the target construction area based on the target contour shape.

[0297] In some embodiments, the determination module 10 is specifically configured to: when the target construction area is a planar structure, directly use the target construction area as a planar construction drawing; when the target construction area is an inclined structure, expand and flatten the target construction area into a planar construction drawing through geometric substitution based on the geometric characteristic parameters of the target construction area.

[0298] In some embodiments, the calculation module 20 is specifically used to: determine the shadow distance corresponding to each point on the upper edge of the obstacle based on the geometric characteristic parameters of the obstacle and the solar radiation-related parameters of the construction site at various times throughout the year; determine the outer edge of the shadow produced by the obstacle based on the shadow distance corresponding to each point on the upper edge; and determine the shadow-blocking area based on the outer edge of the shadow.

[0299] In certain embodiments, the optimization module 30 is specifically used to: construct a grid area corresponding to the photovoltaic component based on the length parameters, width parameters, spacing parameters, predetermined reference direction, and reference direction inclination of the photovoltaic component; overlay the grid area on the plane construction drawing to determine the initial arrangement of the photovoltaic component; adjust the initial arrangement according to the portion of the grid area that exceeds the plane construction drawing, the obstacle area, and the shadow area to obtain an optimized arrangement.

[0300] In certain embodiments, the optimization module 30 is specifically configured to: when the reference direction inclination angle is zero, overlay the grid area on the plane construction drawing, and align the grid area with the coordinate system of the plane construction drawing to determine the initial arrangement of the photovoltaic components; when the reference direction inclination angle is not zero, overlay the grid area on the plane construction drawing, then rotate the grid area according to the reference direction inclination angle, and expand the grid area along the coordinate axis direction of the grid area to determine the initial arrangement of the photovoltaic components.

[0301] In some embodiments, the optimization module 30 is further configured to: calculate the component shielding area generated by the photovoltaic components in the planar construction diagram when the photovoltaic components are installed in a vertical manner; and adjust the grid area according to the component shielding area.

[0302] In some embodiments, the photovoltaic module is connected to an inverter to convert direct current into alternating current. The optimization module 30 is further configured to determine an optimal connection method for the photovoltaic module based on the conversion efficiency of the inverter and the total length of the cables used for connecting the photovoltaic module.

[0303] In certain embodiments, the optimization module 30 is specifically used to: determine the expected number of photovoltaic modules connected to each inverter based on the maximum power, maximum input DC power, capacity ratio and minimum input DC power of the inverter; connect the photovoltaic modules in a predetermined connection method to connect the current group of photovoltaic modules; when the number of connections of the current group of photovoltaic modules reaches the expected number, stop connecting the current group of photovoltaic modules, and re-perform the step of connecting the photovoltaic modules in a predetermined connection method to connect the next group of photovoltaic modules, until all photovoltaic modules are connected.

[0304] In some embodiments, the optimization module 30 is specifically configured to adjust the expected number of photovoltaic modules connected to each inverter when the number of photovoltaic modules connected to the last group is less than the expected number, or connect the last group of photovoltaic modules to a separate inverter.

[0305] In some embodiments, the predetermined connection mode includes any one or more of a horizontal connection mode, a vertical connection mode, a straight connection mode, a C-shaped connection mode, and a loop connection mode.

[0306] In some embodiments, the optimization module 30 is specifically used to: determine a graph structure based on a set of component features and an edge set of the photovoltaic module, where the component features include position coordinates, electrical parameters, and connection status; establish a graph neural network based on the graph structure, and solve the optimized wiring method through reinforcement learning methods.

[0307] In some embodiments, the reinforcement learning method includes any one or more of a proximal policy optimization algorithm and a dominant action-critic algorithm.

[0308] In some embodiments, the reward function in the reinforcement learning method is:

[0309]

[0310] Among them, E is the electrical part, E=μ·η inv -σ·ReLU(R actual -R max ), η inv is the conversion efficiency of a single inverter, ReLU is the activation function, R actual is the actual capacity ratio of the inverter, R max The maximum capacity ratio set for the inverter; L ij is the length of the connection; I cross (P) is the cross detection function; α, β, γ, μ and σ are proportional coefficients.

[0311] It should be noted that the explanation of the photovoltaic module installation optimization method in the aforementioned embodiment is also applicable to the photovoltaic module installation optimization device 100 in the embodiment of the present application, and will not be elaborated here.

[0312] See also Figure 32 The present application also provides an electronic device 200. The electronic device 200 includes one or more processors 210 and a memory 220. The memory 220 stores a computer program, and when the computer program is executed by the processor 210, the installation optimization method of any of the above embodiments is implemented.

[0313] For example, when the computer program is executed by the processor 210, the following installation optimization method is implemented:

[0314] 010: Based on the aerial view of the PV module construction site, determine the target construction area where the PV modules can be arranged, and convert the target construction area into a planar construction map;

[0315] 020: Calculate the shadow occlusion area caused by obstacles in the plane construction map;

[0316] 030: Determine the initial layout of photovoltaic panels based on the plan construction plan, and adjust the initial layout based on the shadow occlusion area to obtain the optimized layout.

[0317] For another example, when the computer program is executed by the processor 210, the following installation optimization method is implemented:

[0318] 011: Obtain an aerial view of the photovoltaic module construction site;

[0319] 012: Based on the aerial photography scale, image processing algorithms are used to extract the contours of the aerial top view to determine the target construction area where photovoltaic modules can be arranged;

[0320] 013: According to the inclination of the target construction area, the target construction area is converted into a plane construction map.

[0321] It should be noted that the explanation of the photovoltaic assembly installation optimization method in the aforementioned embodiment is also applicable to the electronic device 200 of the embodiment of the present application, and will not be elaborated here.

[0322] See also Figure 33 The present application also provides a computer-readable storage medium 300 on which a computer program 310 is stored. When the program is executed by the processor 320, the installation optimization method of any of the above embodiments is implemented.

[0323] For example, when the program is executed by the processor 320, the following installation optimization method is implemented:

[0324] 010: Based on the aerial view of the PV module construction site, determine the target construction area where the PV modules can be arranged, and convert the target construction area into a planar construction map;

[0325] 020: Calculate the shadow occlusion area caused by obstacles in the plane construction map;

[0326] 030: Determine the initial layout of photovoltaic panels based on the plan construction plan, and adjust the initial layout based on the shadow occlusion area to obtain the optimized layout.

[0327] For another example, when the program is executed by the processor 320, the following installation optimization method is implemented:

[0328] 011: Obtain an aerial view of the photovoltaic module construction site;

[0329] 012: Based on the aerial photography scale, image processing algorithms are used to extract the contours of the aerial top view to determine the target construction area where photovoltaic modules can be arranged;

[0330] 013: According to the inclination of the target construction area, the target construction area is converted into a plane construction map.

[0331] It should be noted that the explanation of the photovoltaic assembly installation optimization method in the aforementioned embodiment is also applicable to the computer-readable storage medium 300 in the embodiment of the present application, and will not be elaborated here.

[0332] In summary, the photovoltaic module installation optimization method, photovoltaic module installation optimization device 100, electronic device 200, and computer-readable storage medium 300 of the embodiments of the present application first determine the target construction area where the photovoltaic modules can be arranged based on an aerial view of the photovoltaic module construction site, and convert the target construction area into a planar construction map. The target construction area is then converted into a planar construction map. The shadow obstruction area created by obstacles in the planar construction map is then calculated. Finally, the initial arrangement of the photovoltaic modules is determined based on the planar construction map, and the initial arrangement is adjusted based on the shadow obstruction area to obtain an optimized arrangement. This allows for optimized photovoltaic module installation and addresses the current issues of difficult 3D modeling and high manual design costs associated with photovoltaic module installation.

[0333] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and features of different embodiments or examples without contradiction.

[0334] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present application belong.

[0335] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable storage medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a computer-readable storage medium can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable storage media include the following: an electrical connection having one or more wires (electronic devices), a portable computer disk cartridge (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable storage medium may even be paper or other suitable medium on which the program is printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing it in another suitable manner as necessary, and then stored in a computer memory.

[0336] It should be understood that various parts of the present application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0337] Those skilled in the art will appreciate that all or part of the steps carried out in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a program, and the program can be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment. In addition, the various functional units in the various embodiments of the present application can be integrated into a processing module, or each unit can exist physically alone, or two or more units can be integrated into one module. The above-mentioned integrated module can be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. The storage medium mentioned above can be a read-only memory, a disk or an optical disk, etc.

[0338] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are illustrative and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application. The scope of the present application is defined by the claims and their equivalents.

Claims

1. A photovoltaic module installation optimization method, characterized in that: include: Determining a target construction area where the photovoltaic modules can be arranged based on an aerial view of the construction site of the photovoltaic modules, and converting the target construction area into a planar construction map; Calculate the shadow occlusion area caused by obstacles in the planar construction diagram; An initial arrangement of the photovoltaic components is determined according to the plan construction drawing, and the initial arrangement is adjusted based on the shadow-blocking area to obtain an optimized arrangement.

2. The installation optimization method according to claim 1, characterized in that: The step of determining a target construction area where the photovoltaic modules can be arranged based on the aerial view of the construction site of the photovoltaic modules, and converting the target construction area into a planar construction map includes: Obtaining an aerial view of the construction site of the photovoltaic module; Based on the aerial photography scale, an image processing algorithm is used to extract the outline of the aerial top view to determine the target construction area where the photovoltaic modules can be arranged; According to the inclination of the target construction area, the target construction area is converted into a plane construction map.

3. The installation optimization method according to claim 2, characterized in that: The method of extracting the contour of the aerial view using an image processing algorithm to determine a target construction area where the photovoltaic modules can be arranged includes: performing contrast enhancement processing on the aerial top view to obtain an enhanced image; Performing Gaussian filtering on the enhanced image to obtain a first denoised image; performing Canning edge detection on the first denoised image to obtain a second denoised image; Performing Hough transform shape fitting processing on the second denoised image to extract the target contour shape; The target construction area is determined according to the target outline shape.

4. The installation optimization method according to claim 2, characterized in that: The step of converting the target construction area into a planar construction map according to the inclination of the target construction area includes: When the target construction area is a planar structure, the target construction area is directly used as the planar construction map; When the target construction area is an inclined structure, a ridge line of the target construction area is generated by a straight skeleton algorithm; Dividing the roof line into a plurality of line segments according to the vertices of the roof line; Determining the highest ridge line among the plurality of line segments according to the vertex positions of the plurality of line segments; The slope areas connected by the highest roof ridge line are expanded to obtain the plan construction drawing.

5. The installation optimization method according to claim 4, characterized in that: The target construction area includes polygon vertices and polygon contour lines, and the ridge line of the target construction area is generated by the straight skeleton algorithm, including: Determining whether the polygon outline connected by the polygon vertices is a gable edge to determine the type of the polygon vertices; When the type of the polygon vertex is a free vertex, the free vertex is controlled to shrink inward by the straight skeleton algorithm to determine a free intersection vertex; When the type of the polygon vertex is a partially free vertex, controlling the partially free vertex to move along the gable edge to determine a gable intersection vertex; The ridge line is determined according to the free intersection point and the gable intersection point.

6. The installation optimization method according to claim 4, characterized in that: The slope area is a trapezoidal slope, and the plan construction drawing is obtained by unfolding the slope areas connected by the highest ridge line, including: Obtaining the top view length of the waist of the trapezoidal slope; Determine the unfolded length of the waist of the trapezoidal slope according to the slope angle corresponding to the trapezoidal slope or the ridge line height corresponding to the highest ridge line, and the top-view length; Keeping the upper base of the trapezoidal slope stationary, controlling the parallel movement of the lower base, and calculating the current length of the waist in real time; When the current length is moved to be equal to the expanded length, the plane construction drawing is determined according to the upper base, lower base and waist of the trapezoidal slope.

7. The installation optimization method according to claim 1, characterized in that: The calculating of the shadow occlusion area caused by the obstacles in the planar construction diagram includes: Determining the shadow distance corresponding to each point on the upper edge of the obstacle based on the geometric characteristic parameters of the obstacle and the solar radiation correlation parameters of the construction site at various times throughout the year; Determine the outer edge of the shadow produced by the obstacle according to the shadow distance corresponding to each point of the upper edge; The shadow-blocking area is determined according to the outer edge of the shadow.

8. The installation optimization method according to claim 1, characterized in that: Determining the initial arrangement of the photovoltaic components according to the plan construction drawing, and adjusting the initial arrangement based on the shadow-blocking area to obtain an optimized arrangement includes: Constructing a grid area corresponding to the photovoltaic assembly according to the length parameter, width parameter, spacing parameter, predetermined reference direction and reference direction inclination angle of the photovoltaic assembly; Overlaying the grid area on the plan construction drawing to determine the initial arrangement of the photovoltaic modules; The initial arrangement is adjusted according to the portion of the grid area that exceeds the planar construction drawing, the obstacle area, and the shadow-blocking area to obtain an optimized arrangement.

9. The installation optimization method according to claim 8, characterized in that: Overlaying the grid area on the planar construction drawing to determine the initial arrangement of the photovoltaic components includes: When the reference direction inclination angle is zero, covering the grid area on the planar construction drawing and aligning the grid area with the coordinate system of the planar construction drawing to determine the initial arrangement of the photovoltaic components; When the reference direction inclination angle is not zero, the grid area is covered on the plane construction drawing, and then the grid area is rotated according to the reference direction inclination angle, and the grid area is expanded along the coordinate axis direction of the grid area to determine the initial arrangement of the photovoltaic components.

10. The installation optimization method according to claim 8, characterized in that: Before overlaying the grid area on the planar construction drawing to determine the initial arrangement of the photovoltaic components, the installation optimization method further includes: When the photovoltaic components are installed vertically, calculating the component shielding area generated by the photovoltaic components in the planar construction drawing; The grid area is adjusted according to the component occlusion area.

11. The installation optimization method according to claim 1, characterized in that: The photovoltaic assembly is used to be connected to an inverter to convert direct current into alternating current. The installation optimization method further includes: An optimized wiring mode of the photovoltaic modules is determined based on the conversion efficiency of the inverter and the total length of the cables used for wiring the photovoltaic modules.

12. The installation optimization method according to claim 11, characterized in that: The step of determining the optimized wiring mode of the photovoltaic modules based on the conversion efficiency of the inverter and the total length of the cables used for wiring the photovoltaic modules comprises: Determining an expected number of photovoltaic modules connected to each inverter according to the maximum power, maximum input DC power, capacity ratio, and minimum input DC power of the inverter; Connecting the photovoltaic modules in a predetermined wiring manner to connect a current group of the photovoltaic modules; When the number of connections of the photovoltaic components of the current group reaches the expected number, the connection of the photovoltaic components of the current group is stopped, and the step of connecting the photovoltaic components using the predetermined connection method is repeated to connect the next group of photovoltaic components until all the photovoltaic components are connected.

13. The installation optimization method according to claim 12, characterized in that: The step of determining the optimal connection mode of the photovoltaic modules based on the conversion efficiency of the inverter and the total length of the cables used for connecting the photovoltaic modules further includes: When the number of photovoltaic components connected in the last group is less than the expected number, the expected number of photovoltaic components connected to each inverter is adjusted, or the last group of photovoltaic components is connected to a single inverter.

14. The installation optimization method according to claim 12, characterized in that: The predetermined connection mode includes any one or more of a horizontal connection mode, a vertical connection mode, a straight connection mode, a C-shaped connection mode, and a loop connection mode.

15. The installation optimization method according to claim 11, characterized in that: The step of determining the optimized wiring mode of the photovoltaic modules based on the conversion efficiency of the inverter and the total length of the cables used for wiring the photovoltaic modules comprises: determining a graph structure according to a set of component features and an edge set of the photovoltaic component, wherein the component features include position coordinates, electrical parameters, and connection status; A graph neural network is established based on the graph structure, and the optimized wiring mode is solved by a reinforcement learning method.

16. The installation optimization method according to claim 15, characterized in that: The reinforcement learning method includes any one or more of a proximal strategy optimization algorithm and a dominant action review algorithm.

17. The installation optimization method according to claim 15, characterized in that: The reward function in the reinforcement learning method is: Among them, E is the electrical part, E=μ·η inv -σ·ReLU(R actualb -R max ), η inv is the conversion efficiency of a single inverter, ReLU is the activation function, R actual is the actual capacity ratio of the inverter, R max The maximum capacity ratio set for the inverter; L ij is the length of the connection; I cross (P) is the cross detection function; α, β, γ, μ and σ are proportional coefficients.

18. A photovoltaic module installation optimization device, characterized in that: include: a determination module, configured to determine a target construction area where the photovoltaic modules can be arranged based on an aerial view of the construction site of the photovoltaic modules, and convert the target construction area into a planar construction map; A calculation module, configured to calculate the shadow occlusion area caused by obstacles in the planar construction diagram; An optimization module is used to determine an initial arrangement of the photovoltaic components according to the plan construction drawing, and adjust the initial arrangement based on the shadow-blocking area to obtain an optimized arrangement.

19. An electronic device, characterized in that: The electronic device includes one or more processors and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, the installation optimization method according to any one of claims 1 to 17 is implemented.

20. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the installation optimization method described in any one of claims 1 to 17 is implemented.

Images