Optimal layout method of photovoltaic green plant facade based on light radiation response and visual continuity constraint

Abstract

This invention discloses a method for optimizing the layout of photovoltaic green facades based on light radiation response and visual continuity constraints. The method includes: S1, facade discretization and calculation of total annual light radiation; S2, constructing a parameterized facade module model, defining a type gene and a length parameter gene for each cell; S3, establishing a multi-objective optimization function model: a life-cycle cost function, a net carbon emission function, and a visual discontinuity penalty function; S4, employing a multi-objective evolutionary algorithm, minimizing each objective function in the multi-objective optimization function model, iteratively optimizing the population composed of the type genes and length parameter genes of all cells to obtain a Pareto optimal solution set; S5, selecting a solution from the Pareto optimal solution set for photovoltaic and green facade layout. This invention enables refined calculations based on microscopic light environment conditions, balances the dual objectives of cost and carbon reduction, and ensures that the facade visual effect presents natural, continuous, and gradual characteristics.

Description

Technical Field

[0001] This invention relates to the field of green building and low-carbon energy-saving design, and in particular to a method for optimizing the layout of photovoltaic green facades based on light radiation response and visual continuity constraints. Background Technology

[0002] Green building and low-carbon energy-saving design have become an inevitable trend in the construction industry. Building facades, as the medium for interaction between buildings and the external environment, are a key area for achieving energy conservation and emission reduction in buildings. Currently, the two most widely used active green technologies on building facades are building-integrated photovoltaics (BIPV) and vertical greening systems.

[0003] Existing building facade designs typically consider either photovoltaic (PV) or green facades independently. While PV facades generate clean energy, large-scale installation often leads to increased facade temperatures, raising indoor air conditioning loads. Furthermore, dark-colored PV panels can create a monotonous, industrial look in cities, lacking ecological aesthetics. Green facades, on the other hand, use plant photosynthesis to fix carbon and release oxygen, mitigating the urban heat island effect and beautifying the environment. However, green modules themselves require irrigation and maintenance costs and do not generate direct economic benefits (such as electricity), resulting in long cost recovery periods when applied on a large scale. Therefore, relying solely on one technology cannot simultaneously meet the dual demands of high economic returns (power generation) and high ecological benefits (carbon sequestration and cooling).

[0004] To address the aforementioned issue of single-faceted photovoltaic (PV) systems, some existing technologies attempt to combine PV with greenery. For example, in some demonstration projects, designers, based on experience, place PV panels on the south-facing side of buildings and greenery on the north-facing side; or they create simple strip-like divisions by floor (such as PV panels on windowsills and greenery on balconies). However, this approach is typically based on a one-size-fits-all zoning of macro-orientation, lacking a refined response to local microclimates (especially changes in shading caused by surrounding buildings). This can lead to some PV panels experiencing low power generation efficiency due to localized shading, or some light-loving plants malfunctioning due to insufficient sunlight.

[0005] In recent years, with the development of parametric design, techniques have emerged that utilize tools such as genetic algorithms to arrange facade modules, including methods for determining photovoltaic panel locations based on maximizing radiation. However, these algorithms still have the following shortcomings in practical applications and require improvement:

[0006] 1. Lack of comprehensive cost-benefit consideration: Most existing methods only take "maximizing power generation" as the single objective, ignoring the complex ratio between the cost (installation and maintenance) and carbon reduction benefits (power generation emission reduction and plant carbon sequestration) of photovoltaics and green plants throughout their entire life cycle.

[0007] 2. Ignoring visual continuity (fragmentation problem): This is the most significant flaw in existing technologies. Current optimization algorithms typically discretize the facade into independent grid cells for calculation, and the selection of each cell (photovoltaic or greenery) depends solely on the numerical value of that point. This "pixelated" processing method easily leads to a fragmented distribution of results, that is, a few greenery cells appearing abruptly in a large area of ​​photovoltaics, and vice versa.

[0008] 3. Lack of smooth transition (hard boundary problem): When transitioning from photovoltaic areas to green areas, existing technologies often create uniform or abruptly separated "hard boundaries." This rigid splicing is aesthetically unappealing. Existing algorithms lack a mechanism to control the gradual change in module length and the probabilistic change in module type, failing to achieve a smooth transition effect like that found in nature. Summary of the Invention

[0009] The technical problem to be solved by the present invention is to address the shortcomings of the prior art by providing a photovoltaic greening facade optimization layout method based on light radiation response and visual continuity constraints. This photovoltaic greening facade optimization layout method based on light radiation response and visual continuity constraints can perform fine calculations based on micro-sight lighting environment, take into account the dual goals of cost and carbon reduction, and ensure that the facade visual effect presents natural, continuous and gradual characteristics.

[0010] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0011] A method for optimizing the layout of photovoltaic green facades based on constraints of light radiation response and visual continuity includes the following steps.

[0012] S1. Discretize the target building facade into several cells arranged in an array, and calculate the total annual solar radiation of each cell.

[0013] S2. Construct a parametric facade module model, defining a type gene and a length parameter gene for each cell; the type gene is used to determine whether the corresponding cell is a photovoltaic module or a green plant module; the length parameter gene is used to determine the actual length of the photovoltaic or green plant module in the corresponding cell.

[0014] S3. Establish a multi-objective optimization function model. The objective function should include at least: a full life cycle cost function based on module type and length, a net carbon emission function based on total solar radiation, and a visual discontinuity penalty function to quantify the type and length differences between adjacent cells.

[0015] S4. Using a multi-objective evolutionary algorithm, with the goal of minimizing each objective function in the multi-objective optimization function model established in S3, the population composed of type genes and length parameter genes of all cells is iteratively optimized to obtain the Pareto optimal solution set.

[0016] S5. Based on specific usage requirements, select a solution from the Pareto optimal solution set and implement the photovoltaic and greening facade layout.

[0017] In S1, when discretizing the target building facade, the cell width is set to the standard width value, and the cell height is set to the building floor height.

[0018] In S1, the method for calculating the total annual solar radiation of each cell is as follows: using a light environment simulation tool, import typical annual meteorological data of the target building's location, and calculate the total annual solar radiation of each cell considering the surrounding environment's shading.

[0019] In S2, the type gene takes a value of 0 or 1; when When =1, it means that the cell in the i-th row and j-th column is a photovoltaic module; when When =0, it means that the cell in the i-th row and j-th column is the green plant module;

[0020] The length parameter gene has a value range of 0 to 1 and is used to determine the actual length of the module. The linear interpolation calculation is performed using the following method:

[0021]

[0022] In the formula, and These are the minimum and maximum values ​​of the physical length of the photovoltaic module, respectively.

[0023] and These are the minimum and maximum physical lengths of the green plant module, respectively.

[0024] In S3, the lifecycle cost function is based on module type and length. The expression is:

[0025]

[0026] In the formula, and These represent the comprehensive unit area cost of photovoltaic power and green plants, respectively; W is the cell width.

[0027] In S3, the net carbon emission function based on total solar radiation is the negative of the sum of the annual carbon reduction from power generation of the photovoltaic module and the annual carbon sequestration of the green plant module; the annual carbon reduction from power generation of the photovoltaic module is a nonlinear function of total solar radiation, and the annual carbon sequestration of the green plant module is a nonlinear function of total solar radiation.

[0028] In S3, the net carbon emission function is based on the total solar radiation. The expression is:

[0029]

[0030] In the formula, Based on the total annual solar radiation The function of annual carbon reduction from photovoltaic power generation;

[0031] Based on the total annual solar radiation The annual carbon sequestration function of plants.

[0032] In S3, the visual discontinuity penalty function The expression is:

[0033] In the formula, and These are the type difference weighting coefficient and the length difference weighting coefficient, respectively, and are set values;

[0034] The sum of type differences between the cell in row i and column j and its von Neumann neighborhood; the smaller the difference, the more clustered similar modules are, and the more likely they are to form a "strip" effect;

[0035] It is the sum of squares of the length differences between the cell in row i and column j and the neighboring modules; the smaller the difference, the smoother the length change, and the more it can create a "gradual" effect.

[0036] In S4, the multi-objective evolutionary algorithm is the non-dominated sorting genetic algorithm NSGA-II with an elite strategy.

[0037] In S4, the non-dominated sorting genetic algorithm NSGA-II uses the spatial autocorrelation mutation operator to perform mutation operations during the evolution process. That is, when the gene of a certain cell is mutated, the genes of one or more of its neighboring cells are mutated in the same direction at a preset probability.

[0038] In S5, after the photovoltaic and greenery facade layout is completed, at the boundary between the photovoltaic module area and the greenery module area, the actual length of the module is... It presents a gradual transition.

[0039] The present invention has the following beneficial effects:

[0040] 1. Economic and ecological benefits

[0041] Existing technologies often rely on empirical zoning (such as "photovoltaics on the south side and greenery on the north side"), ignoring the local microclimate differences caused by surrounding building shading, the impact on module efficiency, and the inability to comprehensively consider the cost-effectiveness of photovoltaic power generation and plant carbon sequestration under different radiation intensities.

[0042] This invention calculates radiation at the microscopic level within a facade grid, establishing a multi-objective optimization-based decision-making mechanism to accurately identify the resource potential of each cell, maximizing the building's carbon reduction potential while ensuring low costs. Specifically: economically, this invention avoids installing high-cost photovoltaic modules in high-frequency shaded areas, thus reducing ineffective investment. Ecologically, by using a multi-objective optimization algorithm to find the Pareto optimal solution for "cost-carbon reduction," it can select the configuration with the lowest net carbon emissions within limited budget constraints, effectively contributing to the achievement of carbon neutrality goals throughout the building's lifecycle.

[0043] 2. Aesthetics and Social Benefits

[0044] Existing discretization designs can easily lead to facade fragmentation and visual abrupt changes; existing parametric algorithms can easily result in a messy layout of photovoltaic panels and greenery, creating a chaotic distribution on the facade and harsh boundaries between different functional areas.

[0045] This invention creatively introduces a visual continuity penalty mechanism and a length gradient control strategy into the objective function. By introducing neighborhood continuity constraints and length gradient control, it eliminates local visual abrupt changes, achieving a "clustered" aggregation of photovoltaic modules and green plant modules in terms of type distribution. At the boundary between the two functional areas, it utilizes the continuous variation in module length to form a soft transition wave. This completely solves the "hard boundary" problem of traditional spliced ​​facades, giving the technical facade a rhythmic and continuous feel similar to the skin of a natural organism, greatly enhancing the urban landscape quality of green buildings.

[0046] 3. This invention can automatically adjust the arrangement logic of module lengths according to environmental field data, so that the building facade presents a gradient texture effect that is highly unified in functionality and aesthetics, thereby solving the problem of adaptive fusion under complex boundaries.

[0047] 4. This invention can improve the survival rate and ecological stability of vertical greening.

[0048] Traditional designs often overlook excessive localized radiation or shading, leading to poor plant growth or death on facades and increasing the cost of later maintenance and replacement. The plant carbon sequestration model established in this invention not only considers carbon sequestration efficiency but also implicitly filters for light adaptability (i.e., in radiation ranges unsuitable for plant growth, the algorithm automatically favors photovoltaics or adjusts parameters due to low carbon sequestration efficiency). This invention enables the micro-application of "suitable plants for suitable locations," ensuring that greening modules are distributed only in areas with the most suitable light conditions for their growth, thereby significantly reducing plant mortality and maintenance frequency.

[0049] 5. Improved Design Efficiency: The invention provides an automated solution process based on NSGA-II, which can generate a series of non-dominated optimal solution sets (Pareto Fronts) in a short time (usually within a few hours). Designers no longer need to repeatedly perform manual trial and error; they only need to select from the solution set that prioritizes "economy," "low carbon," or "aesthetics" according to project requirements. This shortens the design decision-making cycle by more than 80% while ensuring the mathematical rigor of the decision results. Attached Figure Description

[0050] Figure 1 This is a flowchart of the photovoltaic greening facade optimization layout method based on light radiation response and visual continuity constraints of the present invention.

[0051] Figure 2 This is the optimized layout diagram of the south facade of the Hongshan Scientific Research and Office Building (Phase II) of Jiangsu Academy of Building Sciences in this embodiment.

[0052] Figure 3 for Figure 2 A partially enlarged schematic diagram of the south-facing facade. Detailed Implementation

[0053] The present invention will now be described in further detail with reference to the accompanying drawings and specific preferred embodiments.

[0054] This embodiment selects the south facade of the Jiangsu Academy of Building Sciences Hongshan Scientific Research and Office Building (Phase II) project as the design and optimization object. This building is a typical office building in a hot-summer, cold-winter region, and its exterior envelope is climate-adaptable. The south facade has good solar radiation conditions; therefore, based on a frame and unit design, various technologies such as photovoltaics, greenery, and shading are integrated to form a unified interface. The total width of the facade is 77.5 meters, and the height is 22.8 meters (5 floors in total, with the first floor having a height of 4.5 meters and the second to fifth floors having a height of 4.0 meters).

[0055] like Figure 1 As shown, a method for optimizing the layout of photovoltaic green facades based on light radiation response and visual continuity constraints includes the following steps.

[0056] S1. Discretize the target building facade into several cells arranged in an array, and calculate the total annual solar radiation of each cell.

[0057] The preferred calculation method is as follows: using a light environment simulation tool, import typical meteorological data of the target building's location, and calculate the total annual solar radiation of each cell considering the surrounding environmental shading.

[0058] A. Hardware and software environment: The preferred configuration in this embodiment is as follows:

[0059] Workstation configuration: Intel i9-12900K CPU, 64GB RAM, NVIDIA RTX 3080 graphics card.

[0060] Software platforms: Rhino 7 (modeling), Grasshopper (parametric logic), Ladybug Tools (environment simulation), Wallacei X (multi-objective evolutionary algorithm engine).

[0061] B. Mesh Generation: Constructing the basic geometric model of the building facade, dividing the facade into... A two-dimensional grid matrix.

[0062] In this embodiment, according to cell width 1.4 meters (corresponding to the standard column span module of an office building), height

[0063] In this embodiment, according to cell width 1.4 meters (corresponding to the standard column span module of an office building), height

[0064] Meters (corresponding to the standard floor height) divide the south facade of the Jiangsu Academy of Building Sciences Hongshan Scientific Research Comprehensive Office Building (Phase II) into 54 (columns) × 4 (rows), totaling 216 cells.

[0065] C. Radiation Simulation

[0066] (1) Using a light environment simulation tool (such as Ladybug Tools), import the typical annual meteorological data (CHN_Jiangsu.Nanjing.582380_CSWD.epw) of the building's location (in this example, Nanjing).

[0067] (2) Establish a model of the surrounding building blocks, especially the shadows of the two high-rise residential buildings and trees on the south side of the building, which will cast a significant shadow in the afternoon.

[0068] (3) Run the LB Incident Radiation component to calculate the total annual radiation of the 216 cell center points. .

[0069] (4) Data results: Calculations show that the top floor (4th floor) of the facade has the highest radiation, which is approximately The bottom layer (layer 1) receives the lowest radiation due to vegetation and surrounding shading, approximately [amount missing]. The radiation on the east side of the facade is significantly lower than that on the west side due to the obstruction of the adjacent building.

[0070] This step provides a precise micro-environmental data foundation for subsequent module selection and efficiency calculations, avoiding the errors of making decisions based solely on macroscopic orientation in traditional designs.

[0071] S2. Construct a parametric facade module model and define a type gene and a length parameter gene for each cell. The type gene is used to determine whether the corresponding cell is a photovoltaic module or a green plant module. The length parameter gene is used to determine the actual length of the photovoltaic or green plant module in the corresponding cell (i.e., the vertical length of the facade).

[0072] The above types of genes take values ​​of 0 or 1; when When, it indicates that the cell in the i-th row and j-th column is a photovoltaic module; when When, it indicates that the cell in the i-th row and j-th column is the green plant module;

[0073] The above length parameter takes a continuous value between 0 and 1, and is used for the actual length of the module. The linear interpolation calculation is performed using the following method:

[0074]

[0075] In the formula, and These represent the minimum and maximum physical lengths of the photovoltaic module, respectively. In this embodiment, considering the windowsill depth and shading requirements, the preferred values ​​are: , .

[0076] and These are the minimum and maximum physical lengths of the green plant module, respectively; in this embodiment, the preferred values ​​are: , .

[0077] This step transforms the facade design into mathematical variables that can be processed by a computer. In particular, the introduction of the length parameter gene breaks the limitation of fixed module size in the traditional way, providing a geometric basis for achieving "gradient".

[0078] S3. Establish a multi-objective optimization function model. The objective function should include at least: a full life cycle cost function based on module type and length, a net carbon emission function based on total solar radiation, and a visual discontinuity penalty function to quantify the type and length differences between adjacent cells.

[0079] A. Lifecycle cost function based on module type and length The preferred expression for (minimizing) is:

[0080]

[0081] In the formula, and These represent the comprehensive unit area cost of photovoltaic power and green plants, respectively; in this embodiment, the preferred value is: This is the comprehensive unit price of photovoltaic (including inverters and mounting systems). The price is the comprehensive unit price for green plants (including drip irrigation systems).

[0082] B. Net carbon emissions function based on total solar radiation The sum of the annual carbon reduction from power generation of the photovoltaic module and the annual carbon sequestration from the green plant module is negative; the annual carbon reduction from power generation of the photovoltaic module is a nonlinear function of the total solar radiation, and the annual carbon sequestration from the green plant module is a nonlinear function of the total solar radiation.

[0083] The above net carbon emission function The preferred expression for minimizing is:

[0084]

[0085] in:

[0086]

[0087]

[0088] In the formula, Based on the total annual solar radiation The function of annual carbon reduction from photovoltaic power generation.

[0089] Let be the area of ​​the cell in the i-th row and j-th column.

[0090] , and These are the photoelectric conversion efficiency, system efficiency coefficient, and average carbon emission factor of the regional power grid, respectively; in this embodiment, a monocrystalline silicon photovoltaic panel is used, therefore... =19.5%, K=0.8, average carbon emission factor of East China regional power grid .

[0091] Based on the total annual solar radiation The annual carbon sequestration function of plants.

[0092] This refers to the annual carbon sequestration per unit area; in this embodiment, the green plant module is preferably a mixed planting module of Sedum lineare and Ivy, therefore... .

[0093] The above life cycle cost function and net carbon emissions function The design aims to find a balance between cost and carbon reduction, ensuring both the economic viability and ecological sustainability of the solution.

[0094] C. Visual discontinuity penalty function The expression for (minimizing) is:

[0095] In the formula, and These are the type difference weighting coefficient and the length difference weighting coefficient, respectively, and are set values; in this embodiment, the preferred values ​​are: , .

[0096] It is the sum of the type differences between the cell in row i and column j and its von Neumann neighborhood (the four neighbors above, below, left, and right); the smaller the difference, the more clustered the similar modules are, and the more they can form a "strip" effect.

[0097] It is the sum of squares of the length differences between the cell in row i and column j and the neighboring modules; the smaller the difference, the smoother the length change, and the more it can create a "gradual" effect.

[0098] The aforementioned visual discontinuity penalty function, through a mathematical penalty mechanism, forces the algorithm to eliminate schemes that are randomly interspersed or have "hard boundaries," and selects solutions with smooth transition characteristics. This is the core improvement of the present invention and is used to quantify the degree of "gradual change" of the facade.

[0099] S4. Using a multi-objective evolutionary algorithm, with the goal of minimizing each objective function in the multi-objective optimization function model established in S3, the population composed of type genes and length parameter genes of all cells is iteratively optimized to obtain the Pareto optimal solution set.

[0100] Furthermore, the aforementioned multi-objective evolutionary algorithm is preferably the non-dominated sorting genetic algorithm NSGA-II with an elitist strategy.

[0101] Furthermore, the non-dominated sorting genetic algorithm NSGA-II employs a spatial autocorrelation mutation operator during the evolution process. That is, when a gene in a cell is mutated, the genes in one or more neighboring cells are mutated in the same direction with a preset probability. Preferably, the following steps are included.

[0102] S4-1. Initialization: Randomly generated Each individual is used as the initial population and contains [number] individuals. Group type genes and length parameter genes.

[0103] S4-2, Assessment: Calculate the results for each individual. Three target values.

[0104] S4-3. Non-dominated ranking: Stratify the population according to Pareto dominance and prioritize retaining individuals that perform well on at least one objective and are not completely dominated by other solutions.

[0105] S4-4, Genetic Operations: The next generation is generated through selection, crossover, and mutation. In the mutation operation, a spatial autocorrelation mutation operator is introduced, meaning that when a cell mutates, there is a certain probability that it will cause its surrounding cells to mutate in the same direction, in order to accelerate the convergence to a continuous form.

[0106] S4-5. Termination: Repeat the above process until the preset algebra is reached, and output the Pareto optimal solution set (ParetoFront).

[0107] In this embodiment, the Wallacei X plugin is preferably used for solving the problem, and the algorithm parameters are set as follows:

[0108] Population Size: 50

[0109] Algebra (Generation): 100

[0110] Crossover rate: 0.8

[0111] Mutation Rate: 0.1 (using spatial autocorrelation mutation strategy)

[0112] Furthermore, the dynamic iteration process is described as follows:

[0113] Generations 1-10: Initial solutions are randomly distributed, resulting in a messy "mosaic" appearance on the facade. This approach is costly and produces inconsistent carbon reduction effects.

[0114] Generations 30-50: The algorithm begins to converge, and high-radiation areas (upper layer, west side) are gradually occupied by photovoltaics, while low-radiation areas (lower layer, shaded areas) are gradually occupied by vegetation. However, abrupt changes still exist at the boundaries.

[0115] Generations 80-100: Due to The punitive effect causes isolated modules to be absorbed, forming contiguous clusters. Simultaneously, the module length parameters at the boundaries are automatically adjusted, creating a continuous distribution from long photovoltaic panels. Short photovoltaic panels short green plants The wavy transition of the evergreen plants.

[0116] S5. From the Pareto optimal solution set, select a compromise solution based on the emphasis (e.g., prioritizing cost, carbon reduction, or aesthetics). Decode the genotype of this compromise solution and generate a model like this in 3D modeling software. Figure 2 and Figure 3 The photovoltaic panel and greening trough model shown automatically generates the actual length of the module at the boundary between the photovoltaic module area and the greening module area. Depend on Reduced to Then switch to Growth to The wave-like gradient shape achieves a smooth interweaving of the two.

[0117] To verify the effectiveness of this invention, this embodiment selects three schemes for data comparison:

[0118] Comparative Example 1 (Traditional Solution): No optimization algorithm is used at all. Based on experience, the solution consists of "two layers of photovoltaic panels on top and two layers of greenery on the bottom", with a uniform length of 1.0m.

[0119] Comparative Example 2 (Conventional Optimization Scheme): Only optimizes cost and carbon emissions, excluding visual continuity constraints. The module length can be freely varied.

[0120] Example (Solution of the present invention): Compromise solution selected from the Pareto front (98th generation, individual 12).

[0121] Table 1. Comparison of performance indicators for different implementation schemes

[0122]

[0123] Detailed effect analysis:

[0124] Compared to Comparative Example 1, the annual net carbon reduction of the present invention is increased by approximately 21.7% (15.1 vs 12.4). This is because the present invention, through radiation calculation, precisely places the photovoltaic panels in locations with high radiation levels, avoiding the problem of low efficiency caused by some photovoltaic panels being located in the lower shaded area in Comparative Example 1.

[0125] Compared to Comparative Example 2 (conventional optimization): Although Comparative Example 2 achieved a slightly higher carbon reduction (15.8), its "visual discontinuity score" was as high as 1850, resulting in a fragmented facade, difficult construction, and extremely unsightly appearance. While the carbon reduction of this invention is only slightly sacrificed (approximately 4%), it reduces the discontinuity score by 98%, achieving a perfect balance between engineering feasibility and aesthetics.

[0126] In the final BIM model, the photovoltaic areas are mainly concentrated in the upper left and upper-middle parts of the facade; the green areas are concentrated in the lower right shaded area. At the boundary between the two, the module length exhibits a sinusoidal curve-like pattern. The rhythmic changes did not reveal any structural flaws such as "photovoltaics being adjacent to greenery with abrupt changes in length", verifying the effectiveness of the invention in controlling visual continuity.

[0127] In summary, this invention, without increasing the physical construction cost of the building facade, achieves the maximization of energy output, the minimization of carbon emissions, and the optimization of architectural aesthetics through algorithmic optimization, and has extremely high value for promotion and application.

[0128] The preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the specific details in the above embodiments. Within the scope of the technical concept of the present invention, various equivalent transformations can be made to the technical solutions of the present invention, and these equivalent transformations all fall within the protection scope of the present invention.

Claims

1. A method for optimizing the layout of photovoltaic green facades based on constraints of light radiation response and visual continuity, characterized in that: include: S1. Discretize the target building facade into several cells arranged in an array, and calculate the total annual solar radiation of each cell. S2. Construct a parametric facade module model, defining a type gene and a length parameter gene for each cell; the type gene determines whether the corresponding cell is a photovoltaic module or a green plant module; the length parameter gene determines the actual length of the photovoltaic or green plant module within the corresponding cell. S3. Establish a multi-objective optimization function model. The objective function should include at least: a full life cycle cost function based on module type and length, a net carbon emission function based on total solar radiation, and a visual discontinuity penalty function to quantify the type and length differences between adjacent cells. S4. Using a multi-objective evolutionary algorithm, with the goal of minimizing each objective function in the multi-objective optimization function model established in S3, iterative optimization is performed on the population composed of type genes and length parameter genes of all cells to obtain the Pareto optimal solution set. S5. Based on specific usage requirements, select a solution from the Pareto optimal solution set and implement the photovoltaic and greening facade layout.

2. The photovoltaic greening facade optimization layout method based on light radiation response and visual continuity constraints as described in claim 1, characterized in that: In S1, when discretizing the target building facade, the cell width is set to the standard width value, and the cell height is set to the building floor height.

3. The photovoltaic greening facade optimization layout method based on light radiation response and visual continuity constraints as described in claim 1, characterized in that: In S1, the method for calculating the total annual solar radiation of each cell is as follows: using a light environment simulation tool, import typical annual meteorological data of the target building's location, and calculate the total annual solar radiation of each cell considering the surrounding environment's shading.

4. The photovoltaic greening facade optimization layout method based on light radiation response and visual continuity constraints as described in claim 1, characterized in that: In S2, the type gene takes a value of 0 or 1; when When =1, it means that the cell in the i-th row and j-th column is a photovoltaic module; when When =0, it means that the cell in the i-th row and j-th column is the green plant module; The length parameter gene has a value range of 0 to 1 and is used to determine the actual length of the module. The linear interpolation calculation is performed using the following method: ； In the formula, and These are the minimum and maximum values ​​of the physical length of the photovoltaic module, respectively. and These are the minimum and maximum physical lengths of the green plant module, respectively.

5. The photovoltaic greening facade optimization layout method based on light radiation response and visual continuity constraints as described in claim 4, characterized in that: In S3, the lifecycle cost function is based on module type and length. The expression is: ； In the formula, and These represent the comprehensive unit area cost of photovoltaic power and green plants, respectively; W is the cell width.

6. The photovoltaic greening facade optimization layout method based on light radiation response and visual continuity constraints as described in claim 4, characterized in that: In S3, the net carbon emission function based on total solar radiation is the negative of the sum of the annual carbon reduction from power generation of the photovoltaic module and the annual carbon sequestration of the green plant module; the annual carbon reduction from power generation of the photovoltaic module is a nonlinear function of total solar radiation, and the annual carbon sequestration of the green plant module is a nonlinear function of total solar radiation.

7. The photovoltaic greening facade optimization layout method based on light radiation response and visual continuity constraints as described in claim 6, characterized in that: In S3, the net carbon emission function is based on the total solar radiation. The expression is: ； In the formula, Based on the total annual solar radiation The function of annual carbon reduction from photovoltaic power generation; Based on the total annual solar radiation The annual carbon sequestration function of plants.

8. The photovoltaic greening facade optimization layout method based on light radiation response and visual continuity constraints as described in claim 4, characterized in that: In S3, the visual discontinuity penalty function The expression is: ; In the formula, and These are the type difference weighting coefficient and the length difference weighting coefficient, respectively, and are set values; The sum of type differences between the cell in row i and column j and its von Neumann neighborhood; the smaller the difference, the more clustered similar modules are, and the more likely they are to form a "strip" effect; It is the sum of squares of the length differences between the cell in row i and column j and the neighboring module; the smaller the difference, the smoother the length change, and the more it can create a "gradual" effect.

9. The photovoltaic greening facade optimization layout method based on light radiation response and visual continuity constraints as described in claim 1, characterized in that: In S4, the multi-objective evolutionary algorithm is the non-dominated sorting genetic algorithm NSGA-II with an elite strategy. During the evolution process, the non-dominated sorting genetic algorithm NSGA-II uses the spatial autocorrelation mutation operator to perform mutation operations. That is, when the gene of a certain cell is mutated, the genes of one or more of its neighboring cells are mutated in the same direction at a preset probability.

10. The photovoltaic greening facade optimization layout method based on light radiation response and visual continuity constraints according to claim 1, characterized in that: In S5, after the photovoltaic and greenery facade layout is completed, at the boundary between the photovoltaic module area and the greenery module area, the actual length of the module is... It presents a gradual transition.

Images