Building facade photovoltaic panel layout method and system based on developable area identification

Abstract

The present invention discloses a method and system for layout of photovoltaic panels on building facades based on identification of developable area, the method comprising: obtaining the developable area and corresponding shadow area of ​​the exterior facade of the building to be developed; establishing a power generation accounting function of the photovoltaic component based on the physical potential value, the area coefficient of the photovoltaic component and the energy conversion coefficient; establishing the initial investment cost function, maintenance cost function and decommissioning cost function of the photovoltaic component, and combining to establish the total economic cost calculation function of the photovoltaic component; optimizing and solving the power generation accounting function and the total economic cost calculation function under the constraints and the goals of maximizing power generation and minimizing economic cost, and laying out the facade according to the solution results. The above technical solution can improve the power generation of the photovoltaic system on the building facade while reducing the installation and maintenance costs of the photovoltaic system.

Description

Technical Field

[0001] The present invention relates to the field of renewable energy power generation technology, and in particular to a building facade photovoltaic panel layout method and system based on developable area identification. Background Art

[0002] In the field of building energy conservation, solar photovoltaic building integration (BIPV) technology has emerged and become a research hotspot. At present, the evaluation methods for the photovoltaic power generation potential of urban building roofs are relatively mature, mainly including sampling method, comprehensive evaluation method and machine learning method. The sampling method is suitable for scenarios with scarce spatial data or large-scale evaluation. By classifying representative buildings and sampling them in layers, the availability of their roofs is evaluated, thereby estimating the photovoltaic power generation potential of large-scale building roofs.

[0003] Although significant progress has been made in the study of rooftop photovoltaics, in-depth exploration of photovoltaics on building facades is still insufficient. The energy density of solar power generation is relatively low, and a large area must be laid to generate considerable electricity. In modern cities, due to the high population density and compact layout of high-rise buildings, the available roof area is limited, and it is difficult to meet local energy needs by relying solely on rooftop photovoltaic power generation. In contrast, although the facade of a building is less exposed to solar radiation and is easily blocked by the shadows of neighboring buildings, its total area is huge. If the available area of ​​the exterior wall is fully utilized for laying photovoltaics, the annual power generation is expected to reach three times the power generation of rooftop photovoltaics.

[0004] Therefore, the photovoltaic layout of building facades has shown great development potential and research value in the design of new buildings, renovation of old buildings and low-carbon urban renewal plans. However, in practical applications, the implementation of photovoltaic systems on building facades requires a lot of financial support, and as the height of buildings increases, the installation and maintenance costs also increase. In addition, the power generation efficiency of different types of photovoltaic modules varies, which directly affects the overall power generation and economic benefits of the entire system. Summary of the invention

[0005] Purpose of the invention: The present invention provides a method and system for layout of photovoltaic panels on building facades based on identification of developable area, aiming to solve the technical problems in the prior art of lack of reasonable and effective layout scheme of photovoltaic components on building facades, resulting in high installation and maintenance costs of photovoltaic systems on building facades and low power generation.

[0006] Technical solution: The present invention provides a photovoltaic panel layout method for building facades based on the identification of developable area, comprising: obtaining the appearance photos of the buildings to be developed and inputting them into the developable area identification model to obtain the developable area and the corresponding shadow area of ​​the facade of the buildings to be developed; citing the developable area of ​​the developable area, combining the orientation of the corresponding facade, calculating the average solar radiation received by the building facade as the physical potential value of the facade; using the shadow area of ​​the shadow area and the installation ratio coefficients of various types of photovoltaic modules to calculate the photovoltaic module area coefficient of the facade; using the energy conversion efficiency of various types of photovoltaic modules as the energy conversion coefficient of the facade. number; based on the physical potential value, the area coefficient of photovoltaic modules and the energy conversion coefficient, the power generation accounting function of photovoltaic modules is established; the initial investment cost function, maintenance cost function and decommissioning cost function of photovoltaic modules are established, and the total economic cost calculation function of photovoltaic modules is combined; there is a correlation between the maintenance cost function and the installation height of various types of photovoltaic modules; under the constraints and the goals of maximizing power generation and minimizing economic costs, the power generation accounting function and the total economic cost calculation function are optimized and solved, and the solution results include the type of photovoltaic modules, the corresponding installation height and the area ratio of various types of photovoltaic modules, and the facade is laid out according to the solution results.

[0007] Specifically, the establishment and training process of the developable area recognition model includes: using the YOLOv8 algorithm to construct the developable area recognition model, and integrating a dynamic attention module therein; using photos with developed areas of building facades marked as training sets to train the developable area recognition model; in the marked photos, the shadow area is also marked, the shadow area is set at the bottom of the building facade, the wide side of the shadow area is consistent with the wide side of the building facade, and the long side of the shadow area is calculated based on the wide side of the building facade and the corresponding average shadow area; the average shadow area of ​​the building facade is calculated as follows: using the solar altitude angle and azimuth at multiple time nodes, combined with the orientation of the building facade to simulate the actual shadow area, obtain the shadow area at multiple time nodes, and calculate the corresponding average shadow area.

[0008] Specifically, the calculation of the average solar radiation received by the building facade as the physical potential value of the facade includes: the physical potential value is calculated using the following formula:

[0009] ,

[0010] Among them, YPP represents the physical potential value, A j represents the developable area of ​​the building facade in the jth direction, SR jIt represents the monthly average hourly solar radiation in the jth direction of the building facade, and m represents the total number of orientation directions.

[0011] Specifically, the calculation of the photovoltaic module area coefficient of the facade includes: the photovoltaic module area coefficient of the facade is calculated using the following formula:

[0012] ,

[0013] Among them, k i represents the area factor of the ith type of photovoltaic module, z i represents the feasible installation ratio coefficient of the ith type of PV modules, S j Represents the shaded area of ​​the shaded region in the j-th direction.

[0014] Specifically, the energy conversion efficiency of various types of photovoltaic modules is used as the energy conversion coefficient of the facade, including: using the following formula to calculate the energy conversion coefficient:

[0015] ,

[0016] Among them, γ i represents the energy conversion coefficient of the ith type of photovoltaic module, η i,module represents the efficiency of the ith photovoltaic module, η i,inverter Represents the inverter efficiency corresponding to the i-th PV module.

[0017] Specifically, the initial investment cost function of the photovoltaic module is as follows:

[0018] ,

[0019] Among them, C initial represents the initial investment cost of PV modules, C Pi represents the unit cost of the ith PV module, r i represents the proportion of the i-th photovoltaic module to the total installation area of ​​the building facade, AT represents the total installation area of ​​the building facade, which is obtained by deducting the corresponding shadow area from the developable area; ρ i represents the laying density of the i-th photovoltaic module, k i represents the area factor of the ith type of photovoltaic module, C Ii represents the unit cost of the inverter corresponding to the i-th PV module, C Ci represents the unit cost of the cable corresponding to the i-th PV module, L Ci represents the length of the cable corresponding to the i-th PV module, and n represents the total number of PV module types.

[0020] Specifically, the maintenance cost function of the photovoltaic module is as follows:

[0021] ,

[0022] Among them, C maintenance represents the maintenance cost of PV panels, O base,i represents the maintenance cost per unit area of ​​the ith PV module at the base installation height, α i H represents the incremental coefficient of the maintenance cost of the ith PV module installation height, avg,j represents the average installation height of the jth height interval of the building facade, T j represents the installation area of ​​the jth height interval, and L represents the total number of height intervals divided by the building facade.

[0023] Specifically, the decommissioning cost function of the photovoltaic module is as follows:

[0024] ,

[0025] Among them, C decommission represents the decommissioning cost of PV modules, C decom,i represents the single decommissioning cost of the i-th PV module.

[0026] Specifically, the constraints include:

[0027] ;

[0028] ,

[0029] Among them, N i represents the number of the ith photovoltaic module, N Ii represents the number of inverters corresponding to the i-th PV module, P inRate,i represents the inverter rated power corresponding to the i-th PV module, H start,j and H end,j They represent the starting height and ending height of the jth height interval of the building facade, respectively. total Indicates the total height of the facade;

[0030] The objective function is as follows:

[0031] ,

[0032] Among them, maxY p and minC total They represent the maximum power generation and the minimum economic cost respectively; the Grey Wolf Algorithm is used to perform multiple iterations of the objective function to solve the multi-objective problem.

[0033] The present invention also provides a building facade photovoltaic panel layout system based on developable area identification, comprising: an identification unit, a power generation accounting unit, a cost accounting unit and a layout unit, wherein: the identification unit is used to obtain an appearance photo of a building to be developed and input it into a developable area identification model to obtain a developable area and a corresponding shadow area of ​​the facade of the building to be developed; the power generation accounting unit is used to quote the developable area of ​​the developable area, and in combination with the orientation of the corresponding facade, calculate the average solar radiation received by the building facade as the physical potential value of the facade; use the shadow area of ​​the shadow area and the installation ratio coefficient of various types of photovoltaic modules to calculate the photovoltaic module area coefficient of the facade; use the energy conversion efficiency of various types of photovoltaic modules as the energy conversion coefficient of the facade number; based on the physical potential value, the area coefficient of the photovoltaic module and the energy conversion coefficient, the power generation accounting function of the photovoltaic module is established; the cost accounting unit is used to establish the initial investment cost function, the maintenance cost function and the decommissioning cost function of the photovoltaic module, and the total economic cost calculation function of the photovoltaic module is established in combination; there is a correlation between the maintenance cost function and the installation height of various types of photovoltaic modules, and the installation area of ​​the photovoltaic module does not include the shadow area corresponding to the building facade; the layout unit is used to optimize and solve the power generation accounting function and the total economic cost calculation function under the constraints and the goals of maximizing power generation and minimizing economic costs. The solution results include the type of photovoltaic module, the corresponding installation height and the area ratio of various types of photovoltaic modules, and the facade is laid out according to the solution results.

[0034] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages: while increasing the power generation of the photovoltaic system on the building facade, it reduces the installation and maintenance costs of the photovoltaic system. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 A schematic flow chart of a method for layout of photovoltaic panels on building facades based on identification of developable areas provided by the present invention. DETAILED DESCRIPTION

[0036] The technical solution of the present invention is further described below in conjunction with the accompanying drawings.

[0037] See also Figure 1 , which is a flow chart of the building facade photovoltaic panel layout method based on developable area identification provided by the present invention.

[0038] In the embodiment of the present invention, an exterior photo of a building to be developed is obtained and input into a developable area recognition model to obtain a developable area and a corresponding shadow area of ​​the exterior facade of the building to be developed.

[0039] In an embodiment of the present invention, the YOLOv8 algorithm is used to construct the developable area recognition model, and a dynamic attention module is integrated therein; photos of the developable area of ​​the building facade are annotated as a training set to train the developable area recognition model; in the annotated photos, a shadow area is also annotated, the shadow area is set at the bottom of the building facade, the wide side of the shadow area is consistent with the wide side of the building facade, and the long side of the shadow area is calculated according to the wide side of the building facade and the corresponding average shadow area; the average shadow area of ​​the building facade is calculated as follows: the solar altitude angle and azimuth at multiple time nodes are used, combined with the orientation of the building facade to simulate the actual shadow area, the shadow area at multiple time nodes is obtained, and the corresponding average shadow area is calculated.

[0040] In the specific implementation, in order to accurately identify the developable area and the corresponding area of ​​the building facade, the present invention can use the YOLOv8 algorithm to build a recognition model. In order to enhance the model's detection performance for small targets in complex backgrounds, a dynamic attention mechanism (DAttention module) is integrated into the backbone network. The module can dynamically adjust the model's receptive field to improve feature extraction capabilities.

[0041] The backbone network of the recognition model includes 5 convolutional layers, 4 C2f modules, 1 SPPF module and 1 DAttention module. The specific structure is as follows: Convolutional layer 1: 64 3×3 convolution kernels with a step size of 2; Convolutional layer 2: 128 3×3 convolution kernels with a step size of 2; C2f module (repeated 3 times): the number of channels is 128; Convolutional layer 3: 256 3×3 convolution kernels with a step size of 2; C2f module (repeated 6 times): the number of channels is 256; Convolutional layer 4: 512 3×3 convolution kernels with a step size of 2; C2f module (repeated 6 times): the number of channels is 512; Convolutional layer 5: 1024 3×3 convolution kernels with a step size of 2; C2f module (repeated 3 times): the number of channels is 1024; DAttention module: integrated in the 10th layer for dynamic feature extraction; SPPF module: the pyramid level is 5 and the number of channels is 1024.

[0042] The detection head of the recognition model includes 2 upsampling layers, 4 concatenation layers, 9 C2f modules, 2 convolution layers and 1 detection layer. The specific structure is as follows: Upsampling layer: the feature map size is enlarged by 2 times; Concatenation layer: concatenate with the feature map P4 in the backbone network in the channel dimension; C2f module (repeated 3 times): the number of channels is 512; Upsampling layer: the feature map size is enlarged by 2 times again; Concatenation layer: concatenate with the feature map P3 in the backbone network in the channel dimension; C2f module (repeated 3 times): the number of channels is 256; Convolution layer: 256 3×3 convolution kernels, stride 2; Concatenation layer: concatenate with the previous feature map in the channel dimension; C2f module (repeated 3 times): the number of channels is 512; Convolution layer: 256 3×3 convolution kernels, stride 2; Concatenation layer: concatenate with the earlier feature map in the channel dimension; C2f module (repeated 3 times): the number of channels is 1024.

[0043] Furthermore, the DAttention module is introduced into the backbone network of YOLOv8, which enables the model to dynamically adjust the receptive field for the input feature map through dynamic sampling offset and attention weight. Given a query vector (including content features and two-dimensional reference points), this mechanism only selects a small number of key points for sampling, reducing the amount of calculation and enhancing the sensitivity to key features. This module receives feature maps as input at the 10th layer of the feature extraction stage, dynamically generates attention weights based on feature importance, and realizes selective emphasis and suppression of features. The weighted feature maps are fused and interacted in subsequent convolution and upsampling layers, and finally converged at the detection layer to generate accurate target detection output.

[0044] In the specific implementation, regarding the photos of the developed areas of the building facades, the developed areas and shadow areas are marked with emphasis. Because in different situations, such as different solar altitude angles and azimuths, the shadow area usually presents an irregular shape. On the one hand, the irregular shape is not conducive to accurate and efficient standard work, and on the other hand, the irregular shape does not make use of the rapid calculation of the corresponding area. At the same time, in most cases, the main area of ​​the shadow appears at the bottom of the building facade. Therefore, in order to further simplify the calculation process, the shadow area of ​​the building facade at multiple time nodes is simulated, and the average shadow area of ​​these multiple time nodes is calculated. The average shadow area is all regarded as appearing at the bottom of the building facade. Therefore, the wide side of the shadow area is consistent with the wide side of the building facade, and the long side of the shadow area is calculated by dividing the average shadow area by the wide side of the building facade. In this way, the marked shadow area is completely at the bottom of the building facade.

[0045] In the specific implementation, in order to determine the shadow area and corresponding area on the building facades in different orientations, ArcGIS software can be used. The calculation formula of the sun altitude angle and azimuth angle is used to simulate the actual shadow effect. The specific formula is as follows:

[0046] ,

[0047] Among them, hs represents the solar altitude angle, φ represents the geographical latitude, δ represents the declination, Ω represents the hour angle, As represents the solar azimuth, H represents the building height, and x and y represent the offset of the horizontal and vertical coordinates of the shadow area relative to the reference coordinates.

[0048] Secondly, according to the slope analysis results, extract the shaded side of the building. In this example, the shaded side of 12 points is output, and the expression is:

[0049] .

[0050] The height information of the backlit surface was extracted and combined with the shaded surface data to calculate the shadow area of ​​the building. In this process, the shadow raster data was converted into vector surface data in ArcGIS software, and the raster with a value of 0 was screened out through the attribute table to determine the shadow area.

[0051] In the specific implementation, after determining the model structure, the model is trained and tested based on the labeled data set. In the model training configuration, after grid search, the following strategies can be adopted: the Adam optimizer is used, the initial learning rate is set to 0.01, and the learning rate decay factor is 0.001 to fine-tune the learning rate in the later stage of training; the momentum parameter is set to 0.937 to accelerate convergence and reduce oscillation; the weight decay coefficient (weight_decay) is set to 0.0005 to prevent the model from overfitting; the warm-up strategy includes 3 warm-up cycles, the warm-up momentum is 0.8, and the warm-up bias learning rate is 0.1 to ensure a smooth start of the training process; in terms of loss function weight, the mean absolute error (MAE) is used as the loss function, and L2 regularization is introduced to improve the robustness of the model. The weights of the loss function are set as: bounding box loss (box) is 7.5, category loss (cls) is 0.5, depth loss (dfl) is 1.5, posture loss (pose) is 12.0, and key point loss (kobj) is 1.0; the data enhancement strategy includes batch size 64, HSV color space enhancement parameters are h: 0.015, s: 0.7, v: 0.4; geometric transformation parameters are translation (translate) 0.1, scaling (scale) 0.5, horizontal flip (fliplr) probability is 0.5, and hybrid data enhancement technology (mosaic 0.1, mixup 0.05) is used; in terms of parallel processing, the number of workers (workers) is set to 12 to improve data loading efficiency. After inspection, the comprehensive prediction accuracy of the model on the test set reached more than 90%, with good performance and can effectively identify the developable area of ​​the building facade.

[0052] After completing the model training and evaluation, the building facades in the optimized area are photographed on site to obtain high-quality image data. After the image is corrected, it is input into the trained model to obtain the coordinate information of the recognition frame in the format of (x min ,y min ,x max ,y max ). The developable area is calculated by the following formula: Width: width=x max -x min ; Height: height=y max -y min ; Area: area = width × height. Combined with parameters such as camera magnification, superimposed calculations can be performed to obtain the developable area and shadow area of ​​the actual building facade. This provides an accurate data basis for the subsequent calculation of the photovoltaic system's power generation potential.

[0053] In the embodiment of the present invention, the developable area of ​​the developable area is referenced and combined with the orientation of the corresponding facade to calculate the average solar radiation received by the building facade as the physical potential value of the facade; the photovoltaic module area coefficient of the facade is calculated using the shadow area of ​​the shadow area and the installation ratio coefficient of various types of photovoltaic modules; the energy conversion efficiency of various types of photovoltaic modules is used as the energy conversion coefficient of the facade; based on the physical potential value, the photovoltaic module area coefficient and the energy conversion coefficient, a power generation accounting function of the photovoltaic module is established.

[0054] In the embodiment of the present invention, the physical potential value refers to the maximum total amount of solar energy that can be captured by the building facade based on solar radiation under ideal conditions. The calculation formula is as follows:

[0055] ,

[0056] Among them, YPP represents the physical potential value, A j represents the developable area of ​​the building facade in the jth direction, SR j The specific implementation process is as follows: First, use PVsyst software to input the geographical location, date, time and other parameters of the building to be optimized, and use the Perez model to calculate the solar radiation SR in each direction. i , Secondly, by superimposing the received amounts in each direction, the total physical potential YPP is obtained.

[0057] In the embodiment of the present invention, in the actual application of the photovoltaic system, due to the influence of multiple factors such as geographical location, building orientation and shadows of surrounding structures, the installation area of ​​the photovoltaic module is often limited. The photovoltaic module can only effectively use a certain proportion of the installation area. This proportion is the feasible installation proportion coefficient of the facade photovoltaic module. The photovoltaic module area coefficient of the building facade can be further calculated to evaluate the actual effective area size of the facade for installing photovoltaic modules. The calculation is performed using the following formula:

[0058] ,

[0059] Among them, k i represents the area factor of the ith type of photovoltaic module, z i represents the feasible installation ratio coefficient of the ith type of PV module, S j Represents the shaded area of ​​the shaded area in the jth direction. The feasible installation ratio coefficients of different types of photovoltaic modules (photovoltaic curtain walls, sunshade facades and wall surfaces, etc.) are different, and the feasible installation ratio coefficients of the same photovoltaic modules on different types of buildings are also different. For example, for commercial buildings, the feasible installation ratio coefficients of photovoltaic curtain walls, sunshade facades and walls are 0.64, 0.18 and 0.22 respectively; for office buildings, the feasible installation ratio coefficients of sunshade facades and walls are 0.18 and 0.7 respectively; for commercial buildings, the feasible installation ratio coefficients of photovoltaic curtain walls, sunshade facades and walls are 0.7, 0.18 and 0.22 respectively.

[0060] In the embodiment of the present invention, different photovoltaic modules have different energy conversion efficiencies under standard test conditions. It is necessary to calculate the energy conversion efficiency of each module as the energy conversion coefficient of the facade. The formula is as follows:

[0061] ,

[0062] Among them, γ i represents the energy conversion coefficient of the ith type of photovoltaic module, η i,module represents the efficiency of the ith photovoltaic module, η i,inverter represents the inverter efficiency corresponding to the i-th photovoltaic module. Considering that the building facade is usually in a vertical or nearly vertical state and the sunlight resources are relatively limited, photovoltaic modules suitable for weak light environments are selected. Specifically, polycrystalline silicon modules are used for the exterior wall and photovoltaic shading parts, and their η module As for the photovoltaic curtain wall, due to its indoor light transmission requirements, thin film components are selected, and its η moduleThe inverter efficiency is determined to be 13%. The inverter efficiency is differentiated according to the scale of the photovoltaic system. Photovoltaic sunshades usually cover a small area and are regarded as small BIPV systems. Micro inverters or small string inverters are used, and their efficiency is between 90% and 96%. inverter The photovoltaic curtain wall and the exterior wall are regarded as large BIPV systems, and usually use string inverters or modular inverters, whose efficiency is between 96% and 98%. In the embodiment of the present invention, η inverter It is set at 97%.

[0063] In the embodiment of the present invention, the power generation calculation function of the photovoltaic module is the sum of the physical potential value, the photovoltaic module area coefficient and the energy conversion coefficient, as shown in the following formula:

[0064] ,

[0065] Among them, Y p Indicates the power generation of the photovoltaic module, k i and γ i They represent the area coefficient and energy conversion coefficient of the i-th photovoltaic module respectively, and n represents the total number of photovoltaic module types.

[0066] In the embodiment of the present invention, an initial investment cost function, a maintenance cost function and a decommissioning cost function of a photovoltaic module are established, and a total economic cost calculation function of the photovoltaic module is established in combination;

[0067] In the embodiment of the present invention, there is a correlation between the maintenance cost function and the installation heights of various types of photovoltaic modules.

[0068] In the embodiment of the present invention, the scope of the initial investment is limited to the cost of key equipment, and the specific calculation formula is as follows:

[0069] ,

[0070] Among them, C initial represents the initial investment cost of PV modules, C Pi represents the unit cost of the ith PV module, r i represents the proportion of the i-th photovoltaic module to the total installation area of ​​the building facade, AT represents the total installation area of ​​the building facade, which is obtained by deducting the corresponding shadow area from the developable area; ρ i represents the laying density of the ith photovoltaic module (pieces / square meter), C Ii represents the unit cost of the inverter corresponding to the i-th PV module (yuan / unit), C Ci represents the unit cost of the cable corresponding to the i-th photovoltaic module (yuan / meter), L Cirepresents the length of the cable corresponding to the i-th PV module (meters), and n represents the total number of PV module types. The inverter cost is calculated by multiplying the inverter unit price by the power generation of the corresponding PV module.

[0071] In the embodiment of the present invention, the maintenance cost involves activities such as regular inspection, cleaning, repair and replacement. As the increase in building height will increase the difficulty and cost of maintenance, it is necessary to consider the impact of the installation height on the maintenance cost. The maintenance costs of different photovoltaic modules at the same height are different. Therefore, the height range related to the photovoltaic module type is introduced as a variable for solving the target, and it is calculated at which heights which photovoltaic module is more suitable. The formula is as follows:

[0072] ,

[0073] Among them, C maintenance represents the maintenance cost of PV panels, O base,i represents the maintenance cost per unit area of ​​the ith PV module at the base installation height (yuan / m2), α i H represents the incremental coefficient of the maintenance cost of the installation height of the ith PV module, avg,j represents the average installation height of the jth height interval of the building facade, T j represents the installation area of ​​the jth height interval, and L represents the total number of height intervals divided by the building facade.

[0074] In a specific implementation, the building facade can be equally divided into multiple height intervals, and the length of each height interval can be determined according to the actual application scenario.

[0075] In the embodiment of the present invention, the decommissioning cost includes equipment dismantling and waste disposal costs, and the formula is as follows:

[0076] ,

[0077] Among them, C decommission represents the decommissioning cost of PV modules, C decom,i represents the single-block decommissioning cost of the i-th PV module, C decom,i The specific calculation formula is as follows:

[0078] ,

[0079] Among them, C material,i represents the material cost of the i-th photovoltaic module, C recycling,i represents the recycling cost of the ith photovoltaic module, V residual,i represents the residual value of the i-th PV module when it is recycled, L lifetime,i Represents the expected life of the i-th PV module.

[0080] In specific implementation, the total economic cost calculation function is the sum of the initial investment cost function, the maintenance cost function and the decommissioning cost function, and the formula is as follows:

[0081] ,

[0082] Among them, C total Represents the total economic cost.

[0083] In an embodiment of the present invention, under the constraints and the goals of maximizing power generation and minimizing economic costs, the power generation accounting function and the total economic cost calculation function are optimized and solved. The solution results include the types of photovoltaic modules, the corresponding installation heights and the area ratios of various types of photovoltaic modules. The facade is laid out according to the solution results.

[0084] In the embodiment of the present invention, the constraint conditions include:

[0085] ;

[0086] ,

[0087] Among them, N i represents the number of the ith photovoltaic module, N Ii represents the number of inverters corresponding to the i-th PV module, P inRate,i represents the inverter rated power corresponding to the i-th PV module, H start,j and H end,j They represent the starting height and ending height of the jth height interval of the building facade, respectively. total Indicates the total height of the facade.

[0088] In the specific implementation, the constraints include the number and area restrictions of photovoltaic modules, as well as the height range restrictions. Further, when solving the objective function, the effective information on the number, installation area and installation height of different types of photovoltaic modules can be obtained, which can be used as the basis for the layout of the building facade.

[0089] In the embodiment of the present invention, the objective function is as follows:

[0090] ,

[0091] Among them, maxY p and minC total Represent the maximum power generation and minimum economic cost respectively.

[0092] In the specific implementation, the Grey Wolf Algorithm can be introduced to perform multiple iterations of the objective function to solve the multi-objective problem.

[0093] The steps of the Grey Wolf Algorithm include:

[0094] (1) Initialize the gray wolf population, randomly generate the initial gray wolf population position, and set the population size to N. The position of each gray wolf is represented by a vector X i =[x i1 ,x i2 ,…,x iD ], where D is the dimension of the decision variables, including the type, area ratio, installation height (range), etc. of PV modules.

[0095] (2) Calculate the fitness function Fitness, for each gray wolf X i , calculate the corresponding power generation Y according to its position P (X i ) and economic cost C total (X i ), construct a fitness function to balance the two optimization objectives:

[0096] ,

[0097] Among them, w 1 and w 2 They respectively represent the weights of the corresponding optimization solution objectives and are set according to actual needs.

[0098] (3) Update the position of the gray wolf. According to the mechanism of the gray wolf algorithm, the position of the gray wolf is updated according to the social level (Alpha, Beta, Delta, Omega) and hunting behavior of the gray wolf.

[0099] (4) Boundary processing: After updating the position of the gray wolf, it is necessary to ensure that it is within the search space. If the position of a gray wolf exceeds the boundary, it needs to be corrected or re-randomly generated to ensure the feasibility of the solution.

[0100] (5) Iterative solution, repeat the above process until the maximum number of iterations is reached. Through continuous iteration, the overall fitness of the population will gradually improve and eventually converge to the optimal solution.

[0101] In the specific implementation, when the maximum number of iterations is reached or the convergence condition is met, the iteration is terminated. The position of the gray wolf with the highest fitness (i.e. the position of the Alpha wolf) is output as the optimal solution. The parameters corresponding to this position are the optimal layout plan of photovoltaic panels on the building facade, including the type of photovoltaic modules, the installation area, area ratio and installation height corresponding to different types.

[0102] The present invention also provides a building facade photovoltaic panel layout system based on developable area identification, comprising: an identification unit, a power generation accounting unit, a cost accounting unit and a layout unit, wherein: the identification unit is used to obtain an appearance photo of a building to be developed and input it into a developable area identification model to obtain a developable area and a corresponding shadow area of ​​the facade of the building to be developed; the power generation accounting unit is used to quote the developable area of ​​the developable area, and in combination with the orientation of the corresponding facade, calculate the average solar radiation received by the building facade as the physical potential value of the facade; use the shadow area of ​​the shadow area and the installation ratio coefficient of various types of photovoltaic modules to calculate the photovoltaic module area coefficient of the facade; use the energy conversion efficiency of various types of photovoltaic modules as the energy conversion coefficient of the facade number; based on the physical potential value, the area coefficient of the photovoltaic module and the energy conversion coefficient, the power generation accounting function of the photovoltaic module is established; the cost accounting unit is used to establish the initial investment cost function, the maintenance cost function and the decommissioning cost function of the photovoltaic module, and the total economic cost calculation function of the photovoltaic module is established in combination; there is a correlation between the maintenance cost function and the installation height of various types of photovoltaic modules, and the installation area of ​​the photovoltaic module does not include the shadow area corresponding to the building facade; the layout unit is used to optimize and solve the power generation accounting function and the total economic cost calculation function under the constraints and the goals of maximizing power generation and minimizing economic costs. The solution results include the type of photovoltaic module, the corresponding installation height and the area ratio of various types of photovoltaic modules, and the facade is laid out according to the solution results.

[0103] In a specific implementation, the building facade photovoltaic panel layout system based on developable area identification provided by the present invention, wherein the unit used to execute methods, steps or functions, and the methods, steps or functions executed by them, can refer to the building facade photovoltaic panel layout method based on developable area identification provided by the present invention.

Claims

1. A building facade photovoltaic panel layout method based on developable area identification, characterized in that: include: Obtain exterior photos of the buildings to be developed, input them into the developable area recognition model, and obtain the developable area and corresponding shadow area of ​​the facades of the buildings to be developed; The developable area of ​​the developable area is referenced, combined with the corresponding orientation of the facade, to calculate the average solar radiation received by the building facade as the physical potential value of the facade; the shadow area of ​​the shadow area and the installation ratio coefficient of various types of photovoltaic modules are used to calculate the photovoltaic module area coefficient of the facade; the energy conversion efficiency of various types of photovoltaic modules is used as the energy conversion coefficient of the facade; based on the physical potential value, the photovoltaic module area coefficient and the energy conversion coefficient, the photovoltaic module power generation accounting function is established; the physical potential value is calculated using the following formula: ， Among them, YPP represents the physical potential value, A j represents the developable area of ​​the building facade in the jth direction, SR j represents the monthly average hourly solar radiation in the jth direction of the building facade, and m represents the total number of orientation directions; The photovoltaic module area coefficient of the facade is calculated using the following formula: ， Among them, k i represents the area factor of the ith type of photovoltaic module, z i represents the feasible installation ratio coefficient of the ith type of PV modules, S j represents the shaded area of ​​the shaded region in the jth direction; The energy conversion coefficient is calculated using the following formula: ， Among them, γ i represents the energy conversion coefficient of the ith type of photovoltaic module, η i,module represents the efficiency of the ith photovoltaic module, η i,inverter represents the inverter efficiency corresponding to the i-th PV module; The initial investment cost function, maintenance cost function and decommissioning cost function of photovoltaic modules are established, and the total economic cost calculation function of photovoltaic modules is established in combination; the maintenance cost function is associated with the installation height of various types of photovoltaic modules; the initial investment cost function of photovoltaic modules is as follows: ， Among them, C initial represents the initial investment cost of PV modules, C Pi represents the unit cost of the ith PV module, r i represents the proportion of the i-th photovoltaic module to the total installation area of ​​the building facade, AT represents the total installation area of ​​the building facade, which is obtained by deducting the corresponding shadow area from the developable area; ρ i represents the laying density of the ith photovoltaic module, C Ii represents the unit cost of the inverter corresponding to the i-th PV module, C Ci represents the unit cost of the cable corresponding to the i-th PV module, L Ci represents the length of the cable corresponding to the i-th PV module, and n represents the total number of PV module types; The maintenance cost function of the photovoltaic module is as follows: ， Among them, C maintenance represents the maintenance cost of PV panels, O base,i represents the maintenance cost per unit area of ​​the ith PV module at the base installation height, α i H represents the incremental coefficient of the maintenance cost of the installation height of the ith PV module, avg,j represents the average installation height of the jth height interval of the building facade, T j represents the installation area of ​​the jth height interval, and L represents the total number of height intervals divided by the building facade; The decommissioning cost function of the photovoltaic module is as follows: ， Among them, C decommission represents the decommissioning cost of PV modules, C decom,i represents the single-block decommissioning cost of the i-th PV module; Under the constraints, and the goals of maximizing power generation and minimizing economic costs, the power generation accounting function and the total economic cost calculation function are optimized and solved. The solution results include the types of photovoltaic modules, the corresponding installation heights, and the area ratios of various types of photovoltaic modules. The facade layout is carried out according to the solution results. The constraints include: ； ， Among them, N i represents the number of the ith photovoltaic module, N Ii represents the number of inverters corresponding to the i-th PV module, P inRate,i represents the inverter rated power corresponding to the i-th PV module, H start,j and H end,j They represent the starting height and ending height of the jth height interval of the building facade, respectively. total Indicates the total height of the facade.

2. The building facade photovoltaic panel layout method based on developable area identification according to claim 1 is characterized in that: The process of establishing and training the developable area recognition model includes: The developable area recognition model is constructed using the YOLOv8 algorithm, and a dynamic attention module is integrated into the model; The photos with the developed areas of the building facades marked are used as training sets to train the developable area recognition model; in the marked photos, the shadow area is also marked, and the shadow area is set at the bottom of the building facade. The wide side of the shadow area is consistent with the wide side of the building facade, and the long side of the shadow area is calculated according to the wide side of the building facade and the corresponding average shadow area; The average shadow area of ​​the building facade is calculated as follows: the solar altitude angle and azimuth at multiple time nodes are used to simulate the actual shadow area in combination with the orientation of the building facade, the shadow area at multiple time nodes is obtained, and the corresponding average shadow area is calculated.

3. The building facade photovoltaic panel layout method based on developable area identification according to claim 1 is characterized in that: The optimization and solution of the power generation accounting function and the total economic cost calculation function under the constraints and the objectives of maximizing power generation and minimizing economic cost include: The objective function is as follows: ， Among them, maxY p and minC total They represent the maximum power generation and minimum economic cost respectively; The Grey Wolf Algorithm is used to perform multiple iterations of the objective function to solve the multi-objective problem.

4. A building facade photovoltaic panel layout system based on developable area identification, characterized in that: include: Identification unit, generation accounting unit, cost accounting unit and layout unit, where: The identification unit is used to obtain the appearance photo of the building to be developed, input it into the developable area identification model, and obtain the developable area and corresponding shadow area of ​​the facade of the building to be developed; The power generation accounting unit is used to quote the developable area of ​​the developable area, combine the orientation of the corresponding facade, calculate the average solar radiation received by the building facade as the physical potential value of the facade; use the shadow area of ​​the shadow area and the installation ratio coefficient of various types of photovoltaic modules to calculate the photovoltaic module area coefficient of the facade; use the energy conversion efficiency of various types of photovoltaic modules as the energy conversion coefficient of the facade; based on the physical potential value, the photovoltaic module area coefficient and the energy conversion coefficient, establish the power generation accounting function of the photovoltaic module; the physical potential value is calculated using the following formula: ， Among them, YPP represents the physical potential value, A j represents the developable area of ​​the building facade in the jth direction, SR j represents the monthly average hourly solar radiation in the jth direction of the building facade, and m represents the total number of orientation directions; The photovoltaic module area coefficient of the facade is calculated using the following formula: ， Among them, k i represents the area factor of the ith type of photovoltaic module, z i represents the feasible installation ratio coefficient of the ith type of PV modules, S j represents the shaded area of ​​the shaded region in the jth direction; The energy conversion coefficient is calculated using the following formula: ， Among them, γ i represents the energy conversion coefficient of the ith type of photovoltaic module, η i,module represents the efficiency of the ith photovoltaic module, η i,inverter represents the inverter efficiency corresponding to the i-th PV module; The cost accounting unit is used to establish the initial investment cost function, maintenance cost function and retirement cost function of the photovoltaic module, and to combine them to establish the total economic cost calculation function of the photovoltaic module; there is a correlation between the maintenance cost function and the installation height of various types of photovoltaic modules, and the installation area of ​​the photovoltaic module does not include the shadow area corresponding to the building facade; the initial investment cost function of the photovoltaic module is as follows: ， Among them, C initial represents the initial investment cost of PV modules, C Pi represents the unit cost of the ith PV module, r i represents the proportion of the i-th photovoltaic module to the total installation area of ​​the building facade, AT represents the total installation area of ​​the building facade, which is obtained by deducting the corresponding shadow area from the developable area; ρ i represents the laying density of the ith photovoltaic module, C Ii represents the unit cost of the inverter corresponding to the i-th PV module, C Ci represents the unit cost of the cable corresponding to the i-th PV module, L Ci represents the length of the cable corresponding to the i-th PV module, and n represents the total number of PV module types; The maintenance cost function of the photovoltaic module is as follows: ， Among them, C maintenance represents the maintenance cost of PV panels, O base,i represents the maintenance cost per unit area of ​​the ith PV module at the base installation height, α i H represents the incremental coefficient of the maintenance cost of the installation height of the ith PV module, avg,j represents the average installation height of the jth height interval of the building facade, T j represents the installation area of ​​the jth height interval, and L represents the total number of height intervals divided by the building facade; The decommissioning cost function of the photovoltaic module is as follows: ， Among them, C decommission represents the decommissioning cost of PV modules, C decom,i represents the single-block decommissioning cost of the i-th PV module; The layout unit is used to optimize and solve the power generation accounting function and the total economic cost calculation function under the constraints and the objectives of maximizing power generation and minimizing economic costs. The solution results include the types of photovoltaic modules, the corresponding installation heights, and the area ratios of various types of photovoltaic modules. The facade is laid out according to the solution results. The constraints include: ； ， Among them, N i represents the number of the ith photovoltaic module, N Ii represents the number of inverters corresponding to the i-th PV module, P inRate,i represents the inverter rated power corresponding to the i-th PV module, H start,j and H end,j They represent the starting height and ending height of the jth height interval of the building facade, respectively. total Indicates the total height of the facade.

Images