A method and system for optimizing the rooftop photovoltaic potential of building complexes under development intensity constraints.
By dividing spatial operation units within urban building clusters, generating parametric building geometry operations, adjusting building geometry models, and performing photovoltaic potential calculations and multi-objective optimization, the constraints on photovoltaic potential assessment and computational efficiency issues in existing technologies are resolved, thereby improving the executability and reliability of the optimization results.
Patent Information
- Application Number
- CN202611532969.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-30
- Publication Date
- 2026-11-17
AI Technical Summary
Existing technologies struggle to generate specific building outlines and height adjustment schemes when assessing the potential of rooftop photovoltaics in urban building complexes, while meeting constraints such as block boundaries, building height, building spacing, and development intensity. This results in optimization results that are difficult to directly translate into spatial entities, leading to confusion between the results of increased photovoltaic potential and reduced development intensity. Furthermore, the assessment of shading impact is insufficient, computational costs are high, surrogate model predictions are inconsistent, and optimization results lack physical model back-substitution verification.
By acquiring spatial and photovoltaic (PV) baseline data of urban blocks, dividing them into spatial operation units, calculating the PV power generation potential of benchmark rooftops, diagnosing PV confinement status, generating parametric building geometry operations, adjusting building geometry models, imposing rooftop resource and development intensity constraints, constructing a mechanism decomposition proxy model, predicting PV response results, establishing a multi-objective optimization model, solving for non-dominated candidate solutions, and performing back-substitution verification of the PV potential calculation model.
It improves the spatial feasibility and reliability of building form optimization results, reduces the risk of shading loss transfer, enhances computational efficiency and the fairness of optimization results, and obtains diversified candidate solutions that take into account both photovoltaic benefits and form perturbation.
Smart Images

Figure CN123117481A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of urban building cluster spatial optimization and building photovoltaic resource assessment technology, and in particular to a method and system for optimizing the rooftop photovoltaic potential of building clusters under development intensity constraints. Background Technology
[0002] Urban building rooftops are important spatial carriers for distributed photovoltaic (PV) systems. Existing rooftop PV potential assessment technologies typically use a given building outline, building height, and roof boundary as inputs to calculate the deployable roof area, solar irradiance level, building shading loss, and expected power generation. These technologies are mainly used for assessing current potential or comparing existing planning schemes.
[0003] The aforementioned technologies can determine the rooftop photovoltaic potential available under specific building cluster morphology conditions, but they typically cannot, under the premise of meeting constraints such as block boundaries, building height, building spacing, fire safety, and development intensity, further determine how to adjust the building outline, roof boundary, and building height to form a spatially feasible rooftop photovoltaic potential enhancement scheme.
[0004] Existing methods for optimizing the form of urban building blocks still have the following problems:
[0005] (1) Optimization parameters are difficult to directly translate into spatial entities. Some existing methods for optimizing the form of building complexes use statistical indicators such as compactness, patch index, average proximity distance, and height variation coefficient as optimization variables. These indicators can describe the overall morphological characteristics of the building complex, but they usually cannot uniquely determine the specific adjustment methods for building outlines, roof boundaries, or building heights. Therefore, even if a better combination of indicators is obtained, it is difficult to directly generate architectural space schemes that can be used for planning and design, 3D modeling, or engineering verification.
[0006] (2) The results of increased photovoltaic potential are easily confused with reduced development intensity. Reducing building height, narrowing building outline, increasing building spacing, or reducing the number of buildings can usually reduce shading between buildings, but these operations may also cause a decrease in total building area, building volume, or floor area ratio. Without constraints or compensation for changes in development intensity, the incremental photovoltaic power generation obtained after optimization may partly come from the reduction in construction capacity rather than a real improvement in the spatial organization efficiency of building clusters, making it difficult to compare different form schemes under equal conditions.
[0007] (3) The volume compensation process lacks an assessment of the incremental shading impact. To maintain development intensity, some methods can transfer the reduced building area or volume to other locations. However, the compensation location is usually determined only based on the remaining construction capacity, the upper limit of building height, or the available construction space, without evaluating the incremental shading caused by the unit compensation volume in different locations and the corresponding photovoltaic energy loss. As a result, shading in a certain area may be reduced, while the compensation building forms new shading in other areas, causing the photovoltaic loss to be spatially transferred, or even offsetting the original optimization benefits.
[0008] (4) The physical computation cost of large-scale morphological parameter search is high. After the building outline, roof boundary or building height changes, the three-dimensional geometry of the building, the deployable area of the roof, the solar position relationship, the building shadow and photovoltaic power generation need to be updated accordingly. When there are many parameter combinations or a large number of spatial operation units, repeating the complete physical simulation for each candidate scheme will generate high computation time and data storage overhead, which is difficult to meet the needs of large-scale scheme search and rapid iteration.
[0009] (5) The surrogate models that directly predict the final power generation lack action path constraints. Some existing surrogate models directly establish a mapping relationship between building form parameters and the final photovoltaic power generation, without distinguishing the impact of building outline adjustment on the roof deployable area and deployability rate, and the impact of building height adjustment on solar irradiance and shading loss. When the training samples are insufficient or the candidate parameters exceed the existing sample distribution, such surrogate models may output results that are inconsistent with the actual geometric changes or photovoltaic calculation relationships, reducing the credibility of the optimization results.
[0010] (6) Lack of physical model back-substitution verification of surrogate optimization results. Some existing methods directly use candidate solutions as recommendations after the surrogate model or multi-objective optimization algorithm outputs them, without regenerating the corresponding building geometry model and performing recalculation of roof deployability, dynamic shading, and photovoltaic power generation potential. Therefore, it is difficult to confirm whether the surrogate prediction error, development intensity error, building geometry validity, and related planning constraints truly meet the requirements, and it is also difficult to eliminate false non-dominated solutions caused by surrogate model errors. Summary of the Invention
[0011] In view of this, the purpose of this invention is to provide a method and system for optimizing the potential of rooftop photovoltaics in building complexes under development intensity constraints. Under the conditions of satisfying constraints such as street boundaries, building height, building spacing and development intensity, the method transforms the photovoltaic limitation state diagnosis of building complexes into an executable building geometry adjustment scheme, taking into account volume balance, shading control, computational efficiency and physical verification of candidate schemes, thereby improving the feasibility and reliability of the optimization results.
[0012] In a first aspect, embodiments of the present invention provide a method for optimizing the rooftop photovoltaic potential of building complexes under development intensity constraints, comprising:
[0013] S100 acquires spatial and photovoltaic basic data of the target city blocks and divides the target city blocks into spatial operation units.
[0014] S200: Based on the aforementioned spatial and photovoltaic baseline data, calculate the baseline rooftop photovoltaic power generation potential of the spatial operation unit, identify the photovoltaic confinement status of the spatial operation unit in conjunction with diagnostic indicators, and determine the corresponding geometric operation access type.
[0015] S300, based on the geometric operation admission type, generate parametric building geometric operations, adjust the geometric features of the building, and generate the adjusted building geometric model.
[0016] S400, apply roof resource constraints and development intensity constraints to the building geometry model. When there is a development capacity deficit, determine a capacity compensation scheme based on the shading sensitivity of the candidate receiving locations and update the building geometry model.
[0017] S500: Input the updated building geometry model into the photovoltaic potential calculation model, recalculate roof resources, roof deployability, shading characteristics and photovoltaic power generation potential, and generate response labels.
[0018] S600, based on the response label and geometric operation parameters, constructs a mechanism decomposition surrogate model to predict the photovoltaic response results after building geometry adjustment.
[0019] S700, Based on the prediction results of the surrogate model, a multi-objective optimization model including the incremental potential of photovoltaic power generation and the degree of geometric perturbation is established, and the non-dominated candidate solutions are obtained by solving the model.
[0020] S800, reconstruct the building geometry model for the non-inferior candidate scheme, perform back-substitution verification through the photovoltaic potential calculation model, and output the rooftop photovoltaic optimization scheme of the building group that meets the constraints based on the verification results.
[0021] In conjunction with the first aspect, the present invention provides a first possible implementation of the first aspect, wherein step S200 specifically includes:
[0022] S210, based on the aforementioned spatial and photovoltaic basic data, determine the first... Baseline rooftop resource base for each space operation unit Baseline roof deployment rate Reference occlusion constraint coefficient Comprehensive coefficient Unit conversion factor The baseline roof deployment rate The value range of the reference occlusion constraint coefficient is [0,1]. The value range is [0,1].
[0023] S220, Calculate the baseline rooftop photovoltaic power generation potential of the space operation unit. The calculation formula is: .
[0024] S230, based on the aforementioned baseline roof deployability Roof continuity index Identify deployability-restricted identifiers The calculation formula is: ,in, The diagnostic threshold corresponding to the baseline roof deployability rate. The diagnostic threshold corresponding to the roof continuity index.
[0025] S240, based on the reference occlusion constraint coefficient and building height variation index Determine if obstruction is restricted. The calculation formula is: ,in, The diagnostic threshold corresponding to the reference occlusion constraint coefficient. The diagnostic threshold corresponding to the building height variation index.
[0026] S250, based on the deployability-restricted identifier and the occlusion-limited identifier Determine the geometric operation admission type of the spatial operation unit when When, it is determined that the spatial operation unit enters the first type of geometric operation parameter space, when When, it is determined that the spatial operation unit enters the second type of geometric operation parameter space, when and When the spatial operation unit enters the third type of geometric operation parameter space, it is determined that the spatial operation unit simultaneously meets multiple admission conditions, and corresponding multiple types of geometric operation scenarios are generated simultaneously.
[0027] In conjunction with the first aspect, this embodiment of the invention provides a second possible implementation of the first aspect, wherein step S300 specifically includes:
[0028] S310, when the spatial operation unit enters the first type of geometric operation parameter space, it generates a first type of parametric architectural geometric operation to enhance the continuity and boundary regularity of roof patches. The first type of parametric architectural geometric operation includes neighboring roof patch fusion, roof boundary morphological closing operation, boundary smoothing, small-scale concave-convex regularity, and inter-building connection body setting.
[0029] S320, when the spatial operation unit enters the second type of geometric operation parameter space, it generates a second type of parametric building geometry operation to reduce the shading loss caused by the building height field and the solar sensitive direction. The second type of parametric building geometry operation includes building height field smoothing, local building layer cutting, solar sensitive direction retreat, local building outline contraction, and height gradient adjustment.
[0030] S330, when the spatial operation unit enters the third type of geometric operation parameter space, it generates a third type of parameterized architectural geometric operation and performs the first type of parameterized architectural geometric operation and the second type of parameterized architectural geometric operation collaboratively in the same architectural geometric model.
[0031] S340, based on the generated geometric operation and control parameters, adjust the geometric features of the building geometric model within the spatial operation unit to obtain the adjusted building geometric model. The geometric features include the building base outline, roof outline, building height, layer cutting position, setback position, roof fusion position, building volume change, and geometric effectiveness status.
[0032] In conjunction with the first aspect, this embodiment of the invention provides a third possible implementation of the first aspect, wherein, in S400, the roof resource constraint and development intensity constraint on the building geometry model includes:
[0033] S410, for the first type of parametric building geometry operation that primarily alters the continuity and boundary regularity of roof patches, calculate the change in the roof resource base before and after the operation. The calculation formula is: At the same time satisfy ,in, This refers to the rooftop resource base of the spatial operation unit after geometric operations. Based on the baseline rooftop resources, This represents the error in the baseline data for rooftop resources.
[0034] S420, When local roof filling, connection, or boundary regularization causes an increase in the roof resource base exceeding the roof resource base error. Context-adaptive processing should be performed.
[0035] S430, defining the first The development capacity of each of the aforementioned space operation units is The development capacity includes the total above-ground building area, the equivalent building volume, and the allowable building area converted from the plot ratio and land area, and the corresponding development intensity is calculated. The calculation formula is: ,in, The area of the spatial operation unit.
[0036] S440, Calculate the development capacity deficit caused by Type II or Type III parametric building geometry operations. The calculation formula is: When satisfied When, capacity compensation is not performed, when When this occurs, low-occlusion sensitive location compensation is activated, whereby... For the first geometric operation The baseline development capacity of each of the aforementioned space operation units, After geometric operations, the first The existing development capacity of the aforementioned space operation units, This is the allowable development capacity error.
[0037] In conjunction with the first aspect, this embodiment of the invention provides a fourth possible implementation of the first aspect, wherein, in S400, determining the capacity compensation scheme based on the occlusion sensitivity of the candidate receiving location and updating the building geometry model includes:
[0038] S450, generates candidate receiving locations for development capacity compensation within the target space operation unit. The candidate receiving position The maximum capacity is .
[0039] S460, regarding the candidate receiving position Apply unit increase in building volume In the representative set of solar states The candidate receiving positions are calculated below. occlusion sensitivity The calculation formula is: ,in, for Represents the planar irradiance of the component under solar conditions. The candidate receiving position The incremental shaded area resulting from increasing the unit capacity The deployable roof area for evaluation. Penalties for border or prohibited construction areas This is a penalty for exceeding the building height limit. Penalties for building spacing or fire safety spacing , and The corresponding penalty weights and occlusion sensitivity are used to determine the appropriate penalty weights. The smaller the value, the smaller the energy-weighted occlusion effect caused by increasing the development capacity per unit at the candidate location.
[0040] S470, aiming to minimize the impact of incremental occlusion and the cost of capacity transfer, constructs a compensatory capacity allocation model, expressed by the formula: At the same time satisfy , ,in, To be assigned to the candidate receiving position The compensation capacity, The spatial distance or transfer cost from the capacity-deficient location to the candidate receiving location. This represents the weight of the capacity transfer cost.
[0041] S480, according to the stated occlusion sensitivity The compensation capacity is allocated to each of the candidate receiving positions from low to high scores. Calculate the development capacity after compensation. The calculation formula is: At the same time satisfy .
[0042] S490, according to the allocated compensation capacity Update the building height and geometry of the candidate site to obtain the updated building geometry model.
[0043] In conjunction with the first aspect, this embodiment of the invention provides a fifth possible implementation of the first aspect, wherein step S500 specifically includes:
[0044] S510 re-enters the building footprint, roof profile, and building height (after geometric operations and capacity compensation) into a photovoltaic potential calculation model with the same calculation caliber as the baseline scenario to calculate the morphological scenario. Next The potential for rooftop photovoltaic power generation in a single space operation unit The calculation formula is: When the photovoltaic technology parameters and meteorological conditions do not change with the scenario, the comprehensive coefficient The composite coefficient remains unchanged when geometric operations involve changes in roof slope, orientation, or component arrangement. Adjust to scenario-dependent parameters ,in, For the first The rooftop deployment rate after adjusting the spatial operation unit For the first The occlusion constraint coefficient after adjustment of each spatial operation unit.
[0045] S520, Calculate the morphological scenario The four types of response tags include:
[0046] Potential absolute incremental label ,
[0047] Potential relative change rate label ,
[0048] Deployability Response Coefficient Label ,
[0049] Occlusion reduces response coefficient label .
[0050] S530, for each set of parameterized morphological scenarios We constructed and saved a full-feature dataset containing morphological features, constraint errors, and photovoltaic response performance.
[0051] S540, based on geometric validity, planning and development intensity constraints, and photovoltaic efficiency gain index, the morphological scenario is... A comprehensive assessment was conducted, and morphological scenarios that simultaneously met geometric topological compliance, planning control requirements, and positive photovoltaic response benefits were marked as feasible scenarios.
[0052] In conjunction with the first aspect, this embodiment of the invention provides a sixth possible implementation of the first aspect, wherein step S600 specifically includes:
[0053] S610, construct a multi-dimensional attribute feature set including morphological geometric parameters, benchmark photovoltaic environment characteristics, three-dimensional spatial layout indicators and solar environment factors, and establish a decomposition surrogate model corresponding to the three types of parametric building geometric operations.
[0054] For the first type of parametric building geometry operation, a first deployability surrogate model is established to predict changes in roof deployability based on the parameters of the first type of geometry operation and the baseline roof morphology characteristics. And rooftop resource change proxy model When the first type of operation may cause changes in the base amount of rooftop resources, a proxy model for changes in rooftop resources is established.
[0055] For the second type of parametric building geometry operations, an occlusion response proxy model is established. It is used to predict shading changes based on height adjustment, layer cutting, setback parameters, reference building height field, and solar geometry.
[0056] For the third type of parametric building geometry operations, a second deployability proxy model is established. and composite occlusion response proxy model The third type of parametric building geometry operation surrogate model is used to predict the combined effects of composite morphology reorganization on roof deployability and shading constraints. The actual recalculation results after the composite operation are used as training labels, instead of directly adding the prediction results of the first and second types.
[0057] S620, based on the intermediate response labels predicted by the decompositional surrogate model, the morphological scenario... The optimized state variables and photovoltaic power generation potential are physically reconstructed under the given conditions, and the calculation formula is as follows: , Apply physical range constraints to make , , This can reduce the risk of surrogate models producing predictions outside the physical scope.
[0058] S630, based on a multidimensional performance index matrix, adapts corresponding decompositional surrogate models for different geometric operations and response variables.
[0059] In conjunction with the first aspect, this embodiment of the invention provides a seventh possible implementation of the first aspect, wherein, in S700, the process of establishing a multi-objective optimization model and solving for non-dominated candidate solutions includes:
[0060] S710, a morphological perturbation degree composed of geometric changes, is constructed to evaluate the trade-off between the incremental potential of photovoltaic power generation and the degree of building form adjustment. The calculation formula is as follows: ,in, To optimize the symmetrical difference in area between the front and rear building or roof outlines, The base building area or roof area, For the first The base area of the building. For the first Changes in the height of the building To accommodate the area of retreat, contraction, or boundary adjustment, The area of the spatial operation unit. To be assigned to a candidate position The compensation capacity, For capacity transfer distance, For scale normalization parameters, , , and The weights are non-negative.
[0061] S720 establishes a bi-objective optimization model with the objective functions of maximizing the incremental potential of photovoltaic power generation and minimizing the morphological perturbation. The calculation formula is as follows: and .
[0062] S730, apply multidimensional space and performance constraints to the bi-objective optimization model.
[0063] S740 employs a multi-objective optimization algorithm to obtain a set of non-dominated candidate solutions that simultaneously satisfy the multi-dimensional space and performance constraints and form a non-dominated relationship between the photovoltaic power generation potential increment and geometric perturbation degree.
[0064] In conjunction with the first aspect, this embodiment of the invention provides an eighth possible implementation of the first aspect, wherein step S800 specifically includes:
[0065] S810, based on the geometric operation type and parameters of the non-inferior candidate scheme, regenerate the corresponding building base outline, roof outline, building height and updated building geometric model, and re-input the building geometric model into the photovoltaic potential calculation model for physical back-substitution recalculation.
[0066] S820 performs multi-dimensional rigorous verification on the back-substitution recalculation results. The multi-dimensional rigorous verification includes verification of proxy prediction error, verification of resources and development intensity, verification of shading and power generation potential, verification of planning constraints, verification of geometric topology, and verification of consistency between building-level and spatial operation unit-level.
[0067] S830 triggers an adaptive correction mechanism for candidate schemes that fail the review, and outputs a rooftop photovoltaic optimization scheme for building complexes that includes spatial geometry, photovoltaic efficiency and constraint verification results for candidate schemes that pass the review, and presents it in a multi-dimensional visualization.
[0068] Secondly, embodiments of the present invention also provide a system for optimizing the potential of rooftop photovoltaic power in building complexes under strength constraints, comprising:
[0069] The data acquisition and unit division module is used to acquire spatial and photovoltaic basic data of the target city blocks and divide the target city blocks into spatial operation units.
[0070] The baseline potential calculation and access diagnosis module is used to calculate the baseline rooftop photovoltaic power generation potential of the space operation unit based on the space and photovoltaic basic data, identify the photovoltaic confinement status of the space operation unit in combination with diagnostic indicators, and determine the corresponding geometric operation access type.
[0071] The parametric geometry model adjustment module is used to generate parametric building geometry operations based on the geometry operation admission type, adjust the geometric features of the building, and generate the adjusted building geometry model.
[0072] The capacity constraint and compensation update module is used to impose roof resource constraints and development intensity constraints on the building geometry model. When there is a development capacity deficit, a capacity compensation scheme is determined based on the shading sensitivity of the candidate receiving locations, and the building geometry model is updated.
[0073] The physical recalculation and label generation module is used to input the updated building geometry model into the photovoltaic potential calculation model, recalculate roof resources, roof deployability, shading characteristics and photovoltaic power generation potential, and generate response labels.
[0074] The surrogate model construction and prediction module is used to construct a mechanism decomposition surrogate model based on the response label and geometric operation parameters, which is used to predict the photovoltaic response results after building geometry adjustment.
[0075] The multi-objective optimization solution module is used to establish a multi-objective optimization model that includes the incremental potential of photovoltaic power generation and the degree of geometric perturbation based on the prediction results of the surrogate model, and solve for non-dominated candidate solutions.
[0076] The geometric reconstruction and back-substitution verification output module is used to reconstruct the building geometric model of the non-inferior candidate scheme, perform back-substitution verification through the photovoltaic potential calculation model, and output the rooftop photovoltaic optimization scheme of the building group that meets the constraints based on the verification results.
[0077] The beneficial effects of the embodiments of the present invention are:
[0078] This invention improves the spatial feasibility of building form optimization results. Based on roof deployability, building shading, and building complex morphological characteristics, it transforms different photovoltaic-limited states into parametric geometric operations that directly affect the building outline, roof boundary, and building height, thus avoiding the problem of only optimizing abstract morphological indicators without generating specific spatial solutions.
[0079] This invention improves the fairness of comparisons between different scenarios. It sets constraints on the base number of rooftop resources and development intensity for different geometric operations, controlling changes in rooftop resources, building volume, total building area, or floor area ratio, thereby reducing inflated photovoltaic revenues obtained by increasing rooftop resources or decreasing development intensity.
[0080] This invention reduces the risk of additional shading caused by development capacity compensation. By calculating the incremental shading impact on surrounding deployable rooftops after increasing the development capacity of candidate locations by one unit, it prioritizes low-shading-sensitive locations to accept the missing capacity, thus reducing the risk of shading losses shifting from the original adjustment location to other locations.
[0081] This invention balances the efficiency of morphological scenario search with the physical consistency of surrogate predictions. It utilizes the results of building geometry scenario recalculation to train decompositional and scenario-adaptive surrogate models, predicting intermediate responses such as roof resources, deployability, and shading. Based on the photovoltaic potential decomposition relationship, it reconstructs the power generation potential, reducing the number of recalculations in multi-objective searches and lowering the possibility of physically inconsistent results from black-box predictions.
[0082] This invention aims to obtain a diverse set of candidate solutions that balance photovoltaic revenue and morphological perturbation. With the goal of maximizing the incremental potential of photovoltaic power generation and minimizing pure geometric perturbation, it obtains a Pareto non-dominant solution set under constraints such as spatial boundaries, development intensity, building height, building spacing, shading, and geometric effectiveness. This set can be further developed into representative solutions of technology priority, balanced approach, and low-perturbation type.
[0083] This invention enhances the credibility and traceability of optimization results. It regenerates building geometry from candidate schemes and performs back-substitution recalculation of photovoltaic potential, verifying proxy prediction errors, resource and development intensity errors, shading changes, power generation potential, and planning constraints. Schemes that pass verification can output building outlines, heights, adjustment locations, capacity transfer locations, and potential change results, providing spatial auxiliary basis for comparing planning schemes. Attached Figure Description
[0084] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.
[0085] Figure 1 This is a flowchart illustrating the method for optimizing the rooftop photovoltaic potential of building complexes under strength constraints, as described in this invention.
[0086] Figure 2 This is a schematic diagram of three types of geometric operations in the method for optimizing the rooftop photovoltaic potential of building complexes under strength constraints in this invention;
[0087] Figure 3 This is a schematic diagram of the hierarchical constraints and condition compensation in the method for optimizing the rooftop photovoltaic potential of building complexes under strength constraints in this invention;
[0088] Figure 4 This is a schematic diagram of the proxy model construction and mechanism decomposition in the optimization method for the rooftop photovoltaic potential of building complexes under strength constraints in this invention;
[0089] Figure 5 This is a schematic diagram illustrating the multi-objective optimization and Pareto front in the method for optimizing the potential of rooftop photovoltaic power in building complexes under strength constraints, as described in this invention.
[0090] Figure 6 This is a schematic diagram of the physical verification and closed-loop process in the method for optimizing the potential of rooftop photovoltaic power in building complexes under strength constraints in this invention.
[0091] Figure 7 This diagram illustrates the baseline photovoltaic potential decomposition, constraint type diagnosis, and scenario admission in the method for optimizing the rooftop photovoltaic potential of building complexes under strength constraints, as described in this invention. Detailed Implementation
[0092] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.
[0093] Please refer to Figures 1 to 7The first embodiment of the present invention provides a method for optimizing the rooftop photovoltaic potential of building complexes under development intensity constraints, comprising: S100, acquiring spatial and photovoltaic basic data of a target urban block, dividing the target urban block into spatial operation units, wherein the spatial and photovoltaic basic data includes building spatial data, rooftop photovoltaic basic data, planning and spatial constraint data, solar and meteorological data, and topographic and three-dimensional environmental data, and the spatial operation units include regular grids, road-enclosed blocks, planned plots, residential areas, commercial parks, office parks, campuses, industrial parks, and internal units of transportation hubs; S200, calculating the baseline rooftop photovoltaic power generation potential of the spatial operation units based on the spatial and photovoltaic basic data, identifying the photovoltaic constraint status of the spatial operation units in conjunction with diagnostic indicators, and determining the corresponding geometric operation access type; S300, generating parametric building geometric operations based on the geometric operation access type, adjusting the geometric features of the buildings, and generating adjusted... The building geometry model is constructed as follows: S400, roof resource constraints and development intensity constraints are applied to the building geometry model. When there is a development capacity deficit, a capacity compensation scheme is determined based on the shading sensitivity of the candidate receiving locations, and the building geometry model is updated; S500, the updated building geometry model is input into the photovoltaic potential calculation model to recalculate roof resources, roof deployability, shading characteristics, and photovoltaic power generation potential, and a response label is generated; S600, based on the response label and geometric operation parameters, a mechanism decomposition surrogate model is constructed to predict the photovoltaic response results after the building geometry adjustment; S700, based on the prediction results of the surrogate model, a multi-objective optimization model including the photovoltaic power generation potential increment and the degree of geometric perturbation is established to solve for non-dominated candidate schemes; S800, the building geometry model is reconstructed for the non-dominated candidate schemes, and back-substituted verification is performed through the photovoltaic potential calculation model. Based on the verification results, the rooftop photovoltaic optimization scheme of the building group that meets the constraints is output.
[0094] Specifically, step S200 includes: S210, determining the first [missing information] based on the aforementioned spatial and photovoltaic basic data. Baseline rooftop resource base for each space operation unit Baseline roof deployment rate Reference occlusion constraint coefficient Comprehensive coefficient Unit conversion factor or scene correction factor The baseline roof deployment rate The value range of the reference occlusion constraint coefficient is [0,1]. The value range of the reference occlusion constraint coefficient is [0,1]. The larger the value, the greater the photovoltaic loss caused by building shading. Formed by solar irradiance conditions, component efficiency, system performance ratio, and calculation cycle; S220, calculate the baseline rooftop photovoltaic power generation potential of the space operation unit. The calculation formula is: S230, based on the aforementioned baseline roof deployability rate Roof continuity index Identify deployability-restricted identifiers The calculation formula is: ,in, The diagnostic threshold corresponding to the baseline roof deployability rate. S240, the diagnostic threshold corresponding to the roof continuity index; S240, based on the reference shading constraint coefficient. and building height variation index Determine if obstruction is restricted. The calculation formula is: ,in, The diagnostic threshold corresponding to the reference occlusion constraint coefficient. S250, based on the diagnostic threshold corresponding to the building height variation index; and the occlusion-limited identifier Determine the geometric operation admission type of the spatial operation unit when When, it is determined that the spatial operation unit enters the first type of geometric operation parameter space, when When, it is determined that the spatial operation unit enters the second type of geometric operation parameter space, when and When the spatial operation unit enters the third type of geometric operation parameter space, it is determined that the spatial operation unit simultaneously meets multiple admission conditions, and corresponding multiple types of geometric operation scenarios are generated simultaneously.
[0095] When a spatial operation unit does not meet any of the admission criteria, it is marked as a unit that maintains its current status, or it is only allowed to enter the low-disturbance parameter range. The above admission thresholds are used to limit the search range of candidate scenarios and are not used as fixed urban planning standards. For high-speed rail station areas, along rail lines, or other urban spaces with linear infrastructure barriers, the rail direction, spatial barrier intensity, and governance boundary can also be used as additional background constraints to adjust the range of geometric operation parameters, but they are not necessary conditions for this method.
[0096] like Figure 2As shown, step S300 specifically includes: S310, when the spatial operation unit enters the first type of geometric operation parameter space, it generates a first type of parametric architectural geometric operation to enhance the continuity and boundary regularity of roof patches. The first type of parametric architectural geometric operation includes neighboring roof patch fusion, roof boundary morphological closing operation, boundary smoothing, small-scale concavity and convexity regularization, and inter-building connection body setting. The corresponding operation control parameters include fusion distance threshold. , closed operation radius Boundary smoothing tolerance Minimum retainable component width and minimum concave / convex dimensions, among which, , , The results of morphological closing operations, boundary smoothing, and other graphic processing are used to generate candidate building or roof geometry, but are not directly used as the final planning result. The candidate geometry needs to be further converted into a solid spatial operation that meets the constraints of building boundaries, minimum component size, building spacing, and development capacity. S320, when the spatial operation unit enters the second type of geometric operation parameter space, it generates a second type of parametric building geometry operation to reduce the shading loss caused by the building height field and the solar sensitive direction. The second type of parametric building geometry operation includes building height field smoothing, local building layer cutting, solar sensitive direction retreat, local building outline contraction, and height gradient adjustment. The corresponding operation control parameters include height field smoothing intensity. Layer ratio , yield distance The maximum height adjustment amount and maximum contour contraction ratio are used. Building height field smoothing is used to reduce abrupt height differences between adjacent buildings, which can be achieved by neighborhood weighting, partitioning and hierarchical, or height gradient constraints. Local layer cutting is used to reduce the building parts that contribute significantly to the shading of surrounding deployable roofs. The layer cutting object can be determined based on the building-by-building shadow contribution, energy-weighted shading contribution, or solar sensitive direction. The solar sensitive direction retreat is based on the solar path, main incident direction, and the position of the shaded roof in the target area, and retreats or contracts the specific boundary of the building contour. When the second type of parametric building geometry operation generates a building capacity deficit, step S400 is entered to perform development intensity verification and condition compensation; S330, when the spatial operation unit enters the third type of geometry operation parameter space, the third type of parametric building geometry operation is generated. The first type of parametric building geometry operation and the second type of parametric building geometry operation are executed collaboratively in the same building geometry model. The corresponding operation control parameters include collaborative intensity parameters. and operation sequence parameters, the cooperation strength parameter The operation sequence parameter is used to control the combined intensity of the two types of operations and to distinguish the geometric coupling results generated by different operation sequences. The third type of geometric operation is not a simple addition of the prediction results of the first and second types of operations, but rather continues to perform subsequent operations on the basis that the preceding operations have changed the building geometric model, thereby preserving the coupling effect between the building outline, height field, roof continuity and shading relationship. S340, according to the generated geometric operations and control parameters, the geometric features of the building geometric model in the spatial operation unit are adjusted to obtain the adjusted building geometric model. The geometric features include the building base outline, roof outline, building height, layer cutting position, setback position, roof fusion position, building volume change and geometric effectiveness state.
[0097] like Figure 3 As shown, in S400, the roof resource constraint and development intensity constraint on the building geometry model includes: S410, for the first type of parametric building geometry operation that mainly changes the continuity and boundary regularity of roof patches, calculating the change in the roof resource base before and after the operation. The calculation formula is: At the same time satisfy ,in, This refers to the rooftop resource base of the spatial operation unit after geometric operations. Based on the baseline rooftop resources, S420, where local roof filling, connection, or boundary regularization causes an increase in the roof resource base exceeding the aforementioned roof resource base error. The scenario adaptation process includes reducing operating parameters, adjusting the boundaries of other low-value rooftops to the same area, including the newly added rooftop area in the building capacity change and simultaneously applying development intensity constraints, and marking the scenario as infeasible; S430, define the first The development capacity of each of the aforementioned space operation units is The development capacity includes the total above-ground building area, the equivalent building volume, and the allowable building area converted from the plot ratio and land area, and the corresponding development intensity is calculated. The calculation formula is: ,in, The area is the spatial operation unit area; when the equivalent building volume is used as the development capacity proxy, the calculation formula is as follows: ,in, For the first Within the first of the aforementioned spatial operation units The base area of the building. To correspond to the building height, when using the total above-ground building area, calculations can be performed based on the building footprint area, number of floors, and standard floor height. S440 calculates the development capacity deficit caused by Type II or Type III parametric building geometry operations. The calculation formula is: When satisfied When, capacity compensation is not performed, when When this occurs, low-occlusion sensitive location compensation is activated, whereby... For the first geometric operation The baseline development capacity of each of the aforementioned space operation units, After geometric operations, the first The existing development capacity of the aforementioned space operation units, To allow for a development capacity error, it is preferable to have a tolerance of no more than 5%.
[0098] In step S400, determining the capacity compensation scheme based on the occlusion sensitivity of the candidate receiving locations and updating the building geometry model includes: S450, generating candidate receiving locations for development capacity compensation within the target space operation unit, or within the target space operation unit and its adjacent areas that allow for collaborative adjustment. The candidate receiving position The maximum capacity is Maximum capacity is The candidate acceptance location must meet all physical and planning constraints, including being located within the permitted construction boundary, not encroaching on the road red line or prohibited construction areas, not exceeding the maximum building height, meeting building spacing and fire safety spacing requirements, meeting minimum building component dimensions, and not disrupting the building and roof geometry; S460, for the candidate acceptance location... Apply unit increase in building volume For example, the increase in unit height or the increase in unit building area, in a representative solar state set The candidate receiving positions are calculated below. occlusion sensitivity The calculation formula is: ,in, for Represents the planar irradiance of the component under solar conditions. The candidate receiving position The incremental shaded area resulting from increasing the unit capacity The deployable roof area for evaluation. Penalties for border or prohibited construction areas This is a penalty for exceeding the building height limit. Penalties for building spacing or fire safety spacing , and The corresponding penalty weights and occlusion sensitivity are used to determine the appropriate penalty weights. The smaller the value, the smaller the energy-weighted occlusion impact caused by increasing the development capacity by one unit at the candidate location; S470, aiming to minimize the incremental occlusion impact and capacity transfer cost, constructs a compensation capacity allocation model, expressed by the formula as follows: At the same time satisfy , ,in, To be assigned to the candidate receiving position The compensation capacity, The spatial distance or transfer cost from the capacity-deficient location to the candidate receiving location. For capacity transfer cost weights; S480, according to the occlusion sensitivity. The compensation capacity is allocated to each of the candidate receiving positions from low to high scores. Calculate the development capacity after compensation. The calculation formula is: At the same time satisfy S490, according to the allocated compensation capacity Update the building height and geometry of the candidate site to obtain the updated building geometry model.
[0099] If the additional energy loss caused by compensation at a candidate location exceeds a preset threshold, the allocation is cancelled, and the next candidate location is selected. When no compensation scheme can satisfy the development strength constraint and occlusion constraint, at least one of the following actions is performed: reduce the layer trimming or yielding strength; change the geometric operation parameters; expand the allowable compensation range; or mark the scenario as infeasible.
[0100] Specifically, step S500 includes: S510, re-inputting the building base profile, roof profile, and building height (after geometric operations and capacity compensation) into a photovoltaic potential calculation model with the same calculation caliber as the baseline scenario, to calculate the morphological scenario. Next The potential for rooftop photovoltaic power generation in a single space operation unit The calculation formula is: When the photovoltaic technology parameters and meteorological conditions do not change with the scenario, the comprehensive coefficient The composite coefficient remains unchanged when geometric operations involve changes in roof slope, orientation, or component arrangement. Adjust to scenario-dependent parameters ,in, For the first The rooftop deployment rate after adjusting the spatial operation unit For the first S520, occlusion constraint coefficient adjusted by each spatial operation unit; S520, calculate the morphological scenario. The four types of response tags include: potential absolute increment tags. Potential relative change rate label Deployability response coefficient label Occlusion reduces response coefficient label S530, for each set of parameterized morphological scenarios Construct and save a full-element feature dataset containing morphological features, constraint errors, and photovoltaic response performance; save a morphological dataset containing spatial operation unit identifiers, geometric operation types and parameters, building geometry before and after adjustment, benchmark and optimized indicators and response labels, roof resource base error, development capacity error, geometric disturbance degree, new shading results, planning constraint satisfaction status, and geometric validity status; S540, based on geometric validity, planning and development intensity constraints, and photovoltaic performance gain indicators, analyze the morphological scenario. A comprehensive assessment was conducted, and morphological scenarios that simultaneously met geometric topological compliance, planning control requirements, and positive photovoltaic response benefits were marked as feasible scenarios.
[0101] like Figure 4 As shown, step S600 specifically includes: S610, constructing a multi-dimensional attribute feature set containing morphological geometric parameters, benchmark photovoltaic environment characteristics, three-dimensional spatial layout indicators, and solar environment factors. This multi-dimensional attribute feature set includes geometric operation type, geometric operation parameters, benchmark roof resource base, benchmark roof deployability rate, benchmark shading constraint coefficient, roof continuity and boundary regularity, building height and height variation, building spacing, solar geometric characteristics, planning constraint capacity, spatial operation unit area, and optional terrain, orbital obstruction, or other spatial background variables. A decompositional proxy model corresponding to the three types of parameterized building geometric operations is established. For the first type of parameterized building geometric operation, a first deployability proxy model is established to predict roof deployability changes based on the first type of geometric operation parameters and benchmark roof morphological characteristics. And rooftop resource change proxy model When the first type of operation may cause changes in the roof resource base, a roof resource change proxy model is established; for the second type of parametric building geometry operation, an occlusion response proxy model is established. It is used to predict shading changes based on height adjustment, layer cutting, setback parameters, reference building height field, and solar geometry features; for the third type of parametric building geometry operations, a second deployable proxy model is established. and composite occlusion response proxy model This is used to predict the combined impact of composite morphological reorganization on roof deployability and shading constraints. The surrogate model for the third type of parametric building geometry operation uses the actual recalculated results after the composite operation as training labels, instead of directly adding the prediction results of the first and second types. S620, based on the intermediate response labels predicted by the decomposed surrogate model, the morphological scenario is... The optimized state variables and photovoltaic power generation potential are physically reconstructed under the given conditions, and the calculation formula is as follows: , Apply physical range constraints to make , , This reduces the risk of the surrogate model generating prediction results outside the physical range; S630, based on a multi-dimensional performance index matrix, adapts the corresponding decompositional surrogate model for different geometric operations and response variables; combined with feasibility filtering and an active sampling mechanism driven by uncertainty and error, the decompositional surrogate model is dynamically optimized and iterated in a closed loop.
[0102] Specifically, the surrogate model algorithm is not limited to a single type; it can include linear regression models, random forests, extreme random trees, gradient boosting trees, XGBoost, probability boosting models, neural networks, Gaussian process regression, and other supervised learning models.
[0103] For different geometric operation types and different response variables, candidate models are compared, and the corresponding surrogate model is selected based on the following indicators: cross-validation error; parameter interpolation error within the same spatial operation unit; generalization error across spatial operation units; generalization error across cities or blocks; prediction uncertainty; physical constraint violation rate; and computation time.
[0104] The same surrogate algorithm is not required for the first, second and third types of geometric operations.
[0105] like Figure 5 As shown, in S700, the process of establishing a multi-objective optimization model and solving for non-dominated candidate solutions includes: S710, constructing a morphological perturbation degree composed of geometric changes, used to evaluate the trade-off between the incremental potential of photovoltaic power generation and the degree of building morphological adjustment, the calculation formula is as follows: ,in, To optimize the symmetrical difference in area between the front and rear building or roof outlines, The base building area or roof area, For the first The base area of the building. For the first Changes in the height of the building To accommodate the area of retreat, contraction, or boundary adjustment, The area of the spatial operation unit. To be assigned to a candidate position The compensation capacity, For capacity transfer distance, For scale normalization parameters, , , and The weights are non-negative, where, The morphological perturbation degree only includes geometric changes such as building outline, height, setbacks, and capacity transfer, and does not include photovoltaic revenue variables; S720, establish a bi-objective optimization model with the objective functions of maximizing the incremental potential of photovoltaic power generation and minimizing the morphological perturbation degree, and the calculation formula is as follows. and S730, apply multidimensional space and efficiency constraints to the bi-objective optimization model; S740, use a multi-objective optimization algorithm to solve for a set of non-dominated candidate solutions that simultaneously satisfy the multidimensional space and efficiency constraints and form a non-dominated relationship between the photovoltaic power generation potential increment and the geometric perturbation degree.
[0106] The multidimensional spatial and performance constraints include: spatial boundary constraints, where all building outlines, roof adjustments, setbacks, and capacity compensations are located within a preset allowable construction boundary; and roof resource constraints, where the first type of geometric operation satisfies... Development intensity constraints, operations involving changes in development capacity must satisfy... Building height constraints: The height of each building shall not exceed the upper limit of the planned height and shall not be lower than the preset minimum building height; Building spacing and fire protection constraints: The adjusted spacing between buildings and the fire protection distance shall not be lower than the preset critical value; Obstruction constraints: Meet the requirements. Alternatively, the energy-weighted increase in occlusion loss may not exceed a preset threshold; revenue constraints must be met. Geometric validity constraints: the adjusted building and roof polygons have no self-intersections, no negative areas, no invalid holes, and meet the minimum building component dimensions; capacity bearing constraints: the compensation capacity does not exceed the maximum bearing capacity of the candidate location.
[0107] One or more of the following multi-objective optimization algorithms can be used to solve the problem: multi-objective gray wolf optimization, NSGA-II, multi-objective particle swarm optimization, multi-objective fruit fly or jellyfish search algorithm, MOJS, Latin hypercube candidate selection, Bayesian multi-objective optimization, and other optimization algorithms that can obtain non-dominated solution sets.
[0108] In one implementation, multi-objective gray wolf optimization can be used as the master solver. This choice only indicates its applicability to a specific sample space and parameter budget, and does not constitute an exclusion of other solution methods. The optimization result is a set of candidate solutions that simultaneously satisfy the constraints and form a non-dominated relationship between the photovoltaic power generation potential increment and the morphological perturbation degree.
[0109] From the non-dominated candidate solutions, at least one of the following representative solutions is extracted: technology-priority solution, which prioritizes the solution with the largest increase in power generation potential while satisfying all constraints; low-disturbance solution, which prioritizes the solution with the smallest morphological disturbance while achieving the preset minimum increase in power generation; balanced solution, which selects the solution at the inflection point of the Pareto front or the solution with the smallest normalized distance from the ideal point; and user-preference solution, which selects the solution based on the user's preferences for benefits, disturbances, development intensity, and spatial constraints.
[0110] like Figure 6 As shown, step S800 specifically includes: S810, regenerating the corresponding building base outline, roof outline, building height, and updated building geometric model based on the geometric operation type and parameters of the non-dominant candidate scheme, and re-inputting the building geometric model into the photovoltaic potential calculation model for physical back-substitution recalculation; S820, performing multi-dimensional rigorous verification on the back-substitution recalculation results, including surrogate prediction error verification, resource and development intensity verification, shading and power generation potential verification, planning constraint verification, geometric topology verification, and building-level and spatial operation unit-level consistency verification; S830, triggering an adaptive correction mechanism for candidate schemes that fail the verification, and outputting a building cluster rooftop photovoltaic optimization scheme containing spatial geometry, photovoltaic efficiency, and constraint verification results for candidate schemes that pass the verification, and performing multi-dimensional visualization.
[0111] The multi-dimensional rigorous verification includes: reviewing the surrogate prediction error and comparing the predictions of the surrogate model. , , Obtained by back substitution , , The relative error of the power generation potential prediction is expressed as ,in, To avoid positive numbers with a denominator of zero, when When the preset threshold is exceeded, the candidate scheme is marked as failing the review; resource and development intensity review: verify whether the changes in rooftop resource base and development capacity both meet the allowable error, and verify whether the total compensation capacity matches the capacity deficit and does not exceed the maximum capacity of the candidate location; shading and power generation potential review: verify whether the shading does not worsen after optimization, whether the weighted shading loss of the newly added energy exceeds the preset threshold, and whether the increase in photovoltaic power generation potential is positive; planning constraint review: verify whether the building outline is within the allowable construction boundary, whether the building height exceeds the upper limit, whether the building spacing and fire distance meet the requirements, and whether the setback, connector and capacity compensation locations intrude into the road red line or prohibited construction area; geometric topology review: verify whether the building and roof polygons have self-intersection, overlap, negative area or narrow areas smaller than the minimum component size, and whether the height and outline can construct an effective three-dimensional topological structure; building-level and spatial operation unit-level consistency review: summarize the roof deployable area, shading loss and power generation potential changes of each building in the spatial operation unit, and verify whether the building-level results can be returned to the spatial operation unit-level results, the calculation formula is as follows. When the aggregated error exceeds the preset threshold, check the building identification association, roof allocation, and spatial overlay relationship.
[0112] A second embodiment of the present invention provides a system for optimizing the rooftop photovoltaic potential of building complexes under development intensity constraints, comprising: a data acquisition and unit division module, used to acquire spatial and photovoltaic basic data of a target urban block and divide the target urban block into spatial operation units; a baseline potential calculation and access diagnosis module, used to calculate the baseline rooftop photovoltaic power generation potential of the spatial operation unit based on the spatial and photovoltaic basic data, identify the photovoltaic limitation status of the spatial operation unit in conjunction with diagnostic indicators, and determine the corresponding geometric operation access type; a parametric geometric model adjustment module, used to generate parametric building geometric operations based on the geometric operation access type, adjust the geometric features of the building, and generate an adjusted building geometric model; and a capacity constraint and compensation update module, used to impose rooftop resource constraints and development intensity constraints on the building geometric model, and when there is a development capacity deficit, update the system based on the shading of candidate receiving locations. The system employs a sensitivity-based capacity compensation scheme, updating the building geometry model. A physical recalculation and label generation module inputs the updated building geometry model into a photovoltaic potential calculation model to recalculate roof resources, roof deployability, shading characteristics, and photovoltaic power generation potential, generating response labels. A surrogate model construction and prediction module constructs a mechanism decomposition surrogate model based on the response labels and geometric operation parameters to predict the photovoltaic response results after building geometry adjustments. A multi-objective optimization solution module establishes a multi-objective optimization model incorporating photovoltaic power generation potential increments and geometric perturbation levels based on the prediction results of the surrogate model, solving for non-dominated candidate schemes. A geometry reconstruction and back-substitution verification output module reconstructs the building geometry model for the non-dominated candidate schemes, performs back-substitution verification using the photovoltaic potential calculation model, and outputs a rooftop photovoltaic optimization scheme for the building cluster that meets the constraints based on the verification results.
[0113] The computer program product of the method and apparatus for optimizing the rooftop photovoltaic potential of building complexes under development intensity constraints provided in the embodiments of the present invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods in the preceding method embodiments. For specific implementation, please refer to the method embodiments, which will not be repeated here.
[0114] Specifically, the storage medium can be a general-purpose storage medium, such as a portable disk or hard disk. When the computer program on the storage medium is run, it can execute the above-mentioned optimization method for the potential of rooftop photovoltaics in building clusters under development intensity constraints. This enables the division of urban block space operation units, diagnosis of photovoltaic confinement status, parametric building geometry adjustment, development capacity constraints and capacity compensation, photovoltaic potential response prediction, multi-objective optimization, and back-substitution verification of optimization schemes to obtain a rooftop photovoltaic optimization scheme for building clusters that meets development intensity constraints and other preset constraints.
[0115] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0116] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.
Claims
1. A method for optimizing the rooftop photovoltaic potential of building complexes under development intensity constraints, characterized in that, include: S100: Acquire spatial and photovoltaic basic data of the target city blocks and divide the target city blocks into spatial operation units; S200, based on the space and photovoltaic basic data, calculate the baseline rooftop photovoltaic power generation potential of the space operation unit, identify the photovoltaic confinement status of the space operation unit in combination with diagnostic indicators, and determine the corresponding geometric operation access type; S300, based on the geometric operation admission type, generate parametric building geometric operations, adjust the geometric features of the building, and generate the adjusted building geometric model; S400, apply roof resource constraints and development intensity constraints to the building geometry model. When there is a development capacity deficit, determine a capacity compensation scheme based on the shading sensitivity of the candidate receiving locations and update the building geometry model. S500: Input the updated building geometry model into the photovoltaic potential calculation model, recalculate roof resources, roof deployability, shading characteristics and photovoltaic power generation potential, and generate response labels; S600, based on the response labels and geometric operation parameters, construct a mechanism decomposition surrogate model to predict the photovoltaic response results after building geometry adjustment; S700, Based on the prediction results of the surrogate model, a multi-objective optimization model including the photovoltaic power generation potential increment and the degree of geometric perturbation is established, and the non-dominated candidate solution is obtained by solving the model. S800, reconstruct the building geometry model for the non-inferior candidate scheme, perform back-substitution verification through the photovoltaic potential calculation model, and output the rooftop photovoltaic optimization scheme of the building group that meets the constraints based on the verification results.
2. The method for optimizing the rooftop photovoltaic potential of building complexes under development intensity constraints according to claim 1, characterized in that, The steps in S200 specifically include: S210, based on the aforementioned spatial and photovoltaic basic data, determine the first... Baseline rooftop resource base for each space operation unit Baseline roof deployment rate Reference occlusion constraint coefficient Comprehensive coefficient Unit conversion factor The baseline roof deployment rate The value range of the reference occlusion constraint coefficient is [0,1]. The value range is [0,1]; S220, Calculate the baseline rooftop photovoltaic power generation potential of the space operation unit. The calculation formula is: ; S230, based on the aforementioned baseline roof deployability Roof continuity index Identify deployability-restricted identifiers The calculation formula is: ,in, The diagnostic threshold corresponding to the baseline roof deployment rate. The diagnostic threshold corresponding to the roof continuity index; S240, based on the reference occlusion constraint coefficient and building height variation index Determine if obstruction is restricted. The calculation formula is: ,in, The diagnostic threshold corresponding to the reference occlusion constraint coefficient. The diagnostic threshold corresponding to the building height variation index; S250, based on the deployability-restricted identifier and the occlusion-limited identifier Determine the geometric operation admission type of the spatial operation unit when When, it is determined that the spatial operation unit enters the first type of geometric operation parameter space, when When, it is determined that the spatial operation unit enters the second type of geometric operation parameter space, when and When the spatial operation unit enters the third type of geometric operation parameter space, it is determined that the spatial operation unit enters the third type of geometric operation parameter space. When the spatial operation unit simultaneously meets multiple admission conditions, it simultaneously generates corresponding multiple types of geometric operation scenarios.
3. The method for optimizing the rooftop photovoltaic potential of building complexes under development intensity constraints according to claim 2, characterized in that, The specific steps of S300 include: S310, when the spatial operation unit enters the first type of geometric operation parameter space, it generates a first type of parametric architectural geometric operation to enhance the continuity and boundary regularity of roof patches. The first type of parametric architectural geometric operation includes neighboring roof patch fusion, roof boundary morphological closing operation, boundary smoothing, small-scale concave-convex regularity, and inter-building connection body setting. S320, when the spatial operation unit enters the second type of geometric operation parameter space, it generates a second type of parametric building geometry operation to reduce the shading loss caused by the building height field and the solar sensitive direction. The second type of parametric building geometry operation includes building height field smoothing, local building layer cutting, solar sensitive direction retreat, local building outline contraction, and height gradient adjustment. S330, when the spatial operation unit enters the third type of geometric operation parameter space, it generates the third type of parameterized architectural geometric operation and performs the first type of parameterized architectural geometric operation and the second type of parameterized architectural geometric operation in the same architectural geometric model. S340, based on the generated geometric operation and control parameters, adjust the geometric features of the building geometric model within the spatial operation unit to obtain the adjusted building geometric model. The geometric features include the building base outline, roof outline, building height, layer cutting position, setback position, roof fusion position, building volume change, and geometric effectiveness status.
4. The method for optimizing the rooftop photovoltaic potential of building complexes under development intensity constraints according to claim 3, characterized in that, In S400, the roof resource constraints and development intensity constraints applied to the building geometry model include: S410, for the first type of parametric building geometry operation, calculate the change in the roof resource base before and after the operation. The calculation formula is: At the same time satisfy ,in, This refers to the rooftop resource base of the spatial operation unit after geometric operations. Based on the baseline rooftop resources, This is for the error in the base number of rooftop resources; S420, When local roof filling, connection, or boundary regularization causes an increase in the roof resource base exceeding the roof resource base error. , and perform scenario-adaptive processing; S430, defining the first The development capacity of each of the aforementioned space operation units is The development capacity includes the total above-ground building area, the equivalent building volume, and the allowable building area converted from the plot ratio and land area, and the corresponding development intensity is calculated. The calculation formula is: ,in, The area of the spatial operation unit; S440, Calculate the development capacity deficit caused by Type II or Type III parametric building geometry operations. The calculation formula is: When satisfied When, capacity compensation is not performed, when When this occurs, low-occlusion sensitive location compensation is activated, whereby... For the first geometric operation The baseline development capacity of each of the aforementioned space operation units, After geometric operations, the first The existing development capacity of the aforementioned space operation units, This is the allowable development capacity error.
5. The method for optimizing the rooftop photovoltaic potential of building complexes under development intensity constraints according to claim 4, characterized in that, In S400, determining the capacity compensation scheme based on the occlusion sensitivity of the candidate receiving locations and updating the building geometry model includes: S450, generates candidate receiving locations for development capacity compensation within the target space operation unit. The candidate receiving position The maximum capacity is ; S460, regarding the candidate receiving position Apply unit increase in building volume In the representative set of solar states The candidate receiving positions are calculated below. occlusion sensitivity The calculation formula is: ,in, for Represents the planar irradiance of the component under solar conditions. The candidate receiving position The incremental shaded area resulting from increasing the unit capacity The deployable roof area for evaluation. Penalties for border or prohibited construction areas This is a penalty for exceeding the building height limit. Penalties for building spacing or fire safety spacing , and The corresponding penalty weight; S470, aiming to minimize the impact of incremental occlusion and the cost of capacity transfer, constructs a compensatory capacity allocation model, expressed by the formula: At the same time satisfy , ,in, To be assigned to the candidate receiving position The compensation capacity, The spatial distance or transfer cost from the capacity-deficient location to the candidate receiving location. Weights for capacity transfer costs; S480, according to the described occlusion sensitivity The compensation capacity is allocated to each of the candidate receiving positions from low to high scores. Calculate the development capacity after compensation. The calculation formula is: At the same time satisfy ; S490, according to the allocated compensation capacity Update the building height and geometry of the candidate site to obtain the updated building geometry model.
6. The method for optimizing the rooftop photovoltaic potential of building complexes under development intensity constraints according to claim 3, characterized in that, The specific steps of S500 include: S510 re-enters the building footprint, roof profile, and building height (after geometric operations and capacity compensation) into a photovoltaic potential calculation model with the same calculation caliber as the baseline scenario to calculate the morphological scenario. Next The potential for rooftop photovoltaic power generation in a single space operation unit The calculation formula is: When geometric operations involve changes in roof slope, orientation, or component arrangement, the comprehensive coefficients will be used. Adjust to scenario-dependent parameters ,in, For the first The rooftop deployment rate after adjusting the spatial operation unit For the first The occlusion constraint coefficient after adjustment of each spatial operation unit; S520, Calculate the morphological scenario The four types of response tags include: Potential absolute incremental label , Potential relative change rate label , Deployability Response Coefficient Label , Occlusion reduces response coefficient label ; S530, for each set of parameterized morphological scenarios Construct and save a full-feature dataset containing morphological features, constraint errors, and photovoltaic response performance; S540, based on geometric validity, planning and development intensity constraints, and photovoltaic efficiency gain index, the morphological scenario is... A comprehensive assessment was conducted, and morphological scenarios that simultaneously met geometric topological compliance, planning control requirements, and positive photovoltaic response benefits were marked as feasible scenarios.
7. The method for optimizing the rooftop photovoltaic potential of building complexes under development intensity constraints according to claim 6, characterized in that, The specific steps of S600 include: S610, construct a multi-dimensional attribute feature set including morphological geometric parameters, benchmark photovoltaic environment characteristics, three-dimensional spatial layout indicators and solar environment factors, and establish a decomposition surrogate model corresponding to the three types of parametric building geometric operations; For the first type of parametric building geometry operations, a first deployability proxy model is established. And rooftop resource change proxy model ; For the second type of parametric building geometry operations, an occlusion response proxy model is established. ; For the third type of parametric building geometry operations, a second deployability proxy model is established. and composite occlusion response proxy model ; S620, based on the intermediate response labels predicted by the decompositional surrogate model, the morphological scenario... The optimized state variables and photovoltaic power generation potential are physically reconstructed under the given conditions, and the calculation formula is as follows: , Apply physical range constraints to make , , .
8. The method for optimizing the rooftop photovoltaic potential of building complexes under development intensity constraints according to claim 2, characterized in that, In S700, the process of establishing a multi-objective optimization model and solving for non-dominated candidate solutions includes: S710, a morphological perturbation degree composed of geometric changes, is constructed to evaluate the trade-off between the incremental potential of photovoltaic power generation and the degree of building form adjustment. The calculation formula is as follows: ,in, To optimize the symmetrical difference in area between the front and rear building or roof outlines, The base building area or roof area, For the first The base area of the building. For the first Changes in the height of the building To accommodate the area of retreat, contraction, or boundary adjustment, The area of the spatial operation unit. To be assigned to a candidate position The compensation capacity, For capacity transfer distance, For scale normalization parameters, , , and Non-negative weights; S720 establishes a bi-objective optimization model with the objective functions of maximizing the incremental potential of photovoltaic power generation and minimizing the morphological perturbation. The calculation formula is as follows: and ; S730, apply multidimensional space and performance constraints to the bi-objective optimization model; S740 employs a multi-objective optimization algorithm to obtain a set of non-dominated candidate solutions that simultaneously satisfy the multi-dimensional space and performance constraints and form a non-dominated relationship between the photovoltaic power generation potential increment and geometric perturbation degree.
9. The method for optimizing the rooftop photovoltaic potential of building complexes under development intensity constraints according to claim 6, characterized in that, The specific steps of S800 include: S810, based on the geometric operation type and parameters of the non-inferior candidate scheme, regenerate the corresponding building base outline, roof outline, building height and updated building geometric model, and re-input the building geometric model into the photovoltaic potential calculation model for physical back-substitution recalculation; S820, performs multi-dimensional rigorous verification on the back-substitution recalculation results. The multi-dimensional rigorous verification includes verification of proxy prediction error, verification of resources and development intensity, verification of shading and power generation potential, verification of planning constraints, verification of geometric topology, and verification of consistency between building level and spatial operation unit level. S830 triggers an adaptive correction mechanism for candidate schemes that fail the review, and outputs a rooftop photovoltaic optimization scheme for building complexes that includes spatial geometry, photovoltaic efficiency and constraint verification results for candidate schemes that pass the review, and presents it in a multi-dimensional visualization.
10. A system for optimizing the potential of rooftop photovoltaic power in building complexes under development intensity constraints, characterized in that, include: The data acquisition and unit division module is used to acquire spatial and photovoltaic basic data of the target city blocks and divide the target city blocks into spatial operation units. The baseline potential calculation and access diagnosis module is used to calculate the baseline rooftop photovoltaic power generation potential of the space operation unit based on the space and photovoltaic basic data, identify the photovoltaic confinement status of the space operation unit in combination with diagnostic indicators, and determine the corresponding geometric operation access type. The parametric geometry model adjustment module is used to generate parametric building geometry operations based on the geometry operation admission type, adjust the geometric features of the building, and generate the adjusted building geometry model. The capacity constraint and compensation update module is used to impose roof resource constraints and development intensity constraints on the building geometry model. When there is a development capacity deficit, a capacity compensation scheme is determined based on the shading sensitivity of the candidate receiving locations, and the building geometry model is updated. The physical recalculation and label generation module is used to input the updated building geometry model into the photovoltaic potential calculation model, recalculate roof resources, roof deployability, shading characteristics and photovoltaic power generation potential, and generate response labels. The proxy model construction and prediction module is used to construct a mechanism decomposition proxy model based on the response label and geometric operation parameters, which is used to predict the photovoltaic response results after building geometry adjustment; The multi-objective optimization solution module is used to establish a multi-objective optimization model that includes the photovoltaic power generation potential increment and the degree of geometric perturbation based on the prediction results of the surrogate model, and solve for non-dominated candidate solutions. The geometric reconstruction and back-substitution verification output module is used to reconstruct the building geometric model of the non-inferior candidate scheme, perform back-substitution verification through the photovoltaic potential calculation model, and output the rooftop photovoltaic optimization scheme of the building group that meets the constraints based on the verification results.
