A green and intelligent building planning and design method and system based on multimodal data

Abstract

This invention relates to the field of green and intelligent building planning and design technology, and discloses a method and system for green and intelligent building planning and design based on multimodal data. The method includes: acquiring environmental data of the target mountainous area; performing a micro-topographic tiered foundation construction task; performing an initial building block growth task; performing a three-performance evaluation task based on a multi-objective fuzzy optimization mechanism; and completing the approximate optimization of the final form combination scheme. Compared with existing technologies that cannot effectively adapt to the complex slope aspects and rainfall trends of mountains, especially under typical mountainous environmental conditions, it is difficult to achieve coordinated control of climate response and terrain adaptation, resulting in low integration of building block layout with the environment, poor rainwater drainage efficiency, and obstructed visual corridors. This application, by introducing a terrain abrupt change identification mechanism, a building block growth mechanism, and a multi-objective fuzzy optimization strategy, takes into account both terrain coupling and climate adaptability, thus improving the level of intelligence in the planning and design of buildings in mountainous scenic areas.

Description

Technical Field

[0001] This invention relates to the field of green and intelligent building planning and design technology, and in particular to a green and intelligent building planning and design method and system based on multimodal data. Background Technology

[0002] Currently, the planning and design of buildings in mountainous scenic areas still mainly rely on architects' perception and subjective judgment based on their on-site experience. There is a lack of a systematic quantitative processing mechanism for complex terrain data and climate environmental factors, resulting in poor coupling between the layout of building blocks and actual terrain, sunlight, drainage and other factors, which seriously restricts the green and intelligent development level of scenic area buildings.

[0003] For example, in the existing planning and design process, traditional CAD drawings or topographic contour maps are usually used for preliminary layout, and static spatial arrangement is made with the help of limited building shape parameters. However, factors such as abrupt elevation changes, rapid slope changes, slope aspect heterogeneity, and monsoon-dominated rainfall distribution commonly found in mountainous environments can significantly affect the stability, thermal comfort, and visibility of landscape corridors of building blocks. Existing technologies cannot yet achieve the following tasks: (1) Traditional methods for processing DEM elevation data mostly rely on uniform scale filters, which make it difficult to distinguish between local slope changes and noise interference, leading to an increase in base construction errors. (2) Building shapes cannot respond to the coupled changes of topographic microstructure and climate conditions. Especially in typical mountainous areas, building layout needs to consider rapid rainwater drainage (avoiding waterlogging), good ventilation orientation, and the continuity of landscape corridors. Current methods are mostly based on manual and repeated deliberation, lacking systematic and multi-objective quantitative indicators. (3) Currently, most of the evaluation methods are based on a single objective (such as shading, heat load, etc.) and lack a comprehensive evaluation method that integrates multiple performance indicators such as rainwater drainage efficiency, shading coefficient, and visibility integrity. This results in the design scheme being unable to be intelligently adjusted and optimized under multi-dimensional considerations.

[0004] Therefore, there is an urgent need for a green and intelligent building planning and design method based on multimodal data, which can still achieve multi-objective optimization control of building block growth, performance evaluation and spatial combination in complex terrain and highly climate-sensitive mountainous scenic areas, thereby improving the integration of buildings with the natural environment and enhancing the applicability and responsiveness of green and intelligent building planning and design in actual engineering projects. Summary of the Invention

[0005] To address the aforementioned technical shortcomings, the present invention aims to propose a green and intelligent building planning and design method based on multimodal data. This method addresses the technical problems in existing technologies, such as the inability to effectively adapt to complex slope aspects and rainfall trends in mountainous areas, especially in typical mountainous environments. It also highlights the difficulty in achieving coordinated control of climate response and terrain adaptation, resulting in low integration of building blocks with the environment, poor rainwater drainage efficiency, and obstructed visual corridors.

[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The present invention provides a green and intelligent building planning and design method based on multimodal data.

[0007] The green and intelligent building planning and design method based on multimodal data includes:

[0008] Step S10: Obtain environmental data for the target mountainous area Based on environmental data of the target mountainous area An adaptive elevation noise suppression task is performed using a mechanism based on the combined use of terrain-sensitive area labeling and dynamic filtering, outputting elevation field data T for the target mountainous area. Among them, environmental data of the target mountainous area The data format is elevation DEM data, with x as the horizontal axis and y as the vertical axis;

[0009] Step S20: Based on the elevation field data T of the target mountainous area A slope-guided progressive base partitioning mechanism is used to construct the micro-topographic tiered base, outputting the micro-topographic tiered base sequence H. ;

[0010] Step S30: Based on the micro-topographic stepped base sequence H The initial building block growth task is performed using a terrain factor-driven multi-scale morphological growth mechanism, and the output is a climate response block set. ;

[0011] Step S40: Based on the climate response block set A multi-objective fuzzy optimization mechanism is used to perform the three-performance evaluation task, and the three-performance operator set is output. ;

[0012] Step S50: Based on the three performance operator sets A three-objective approximation optimization task is performed using an alternating iterative mechanism driven by multiple objectives, and the final shape combination scheme is output. .

[0013] Preferably, in step S10, environmental data of the target mountain area is acquired. Based on environmental data of the target mountainous area An adaptive elevation noise suppression task is performed using a mechanism based on the combined use of terrain-sensitive area labeling and dynamic filtering, outputting elevation field data T for the target mountainous area. The steps specifically include:

[0014] Step S101: Obtain environmental data of the target mountainous area Based on environmental data of the target mountainous area A mutation feature operator combining the first and second derivatives of space is used to extract boundary-sensitive regions, outputting a set of mutation region masks. ;

[0015] Step S102: Masking the mutation region set For the region outside the range, a 3×3 mean filter is used for smoothing, and the noise estimation function for the first region is output. ; for the set of masks for mutation regions Within the region, a combination of directional adaptive Sobel and Fourier filters is used for local directional consistency filtering, outputting a noise estimation function for the second region. Based on the noise estimation function of the first region Second region noise estimation function Define the final localization noise estimation function δ(x,y). ;

[0016] Step S103: Based on the regionalized noise estimation function δ(x,y), perform regional difference filtering on the target mountain environmental data D(x,y) using the NumPy and SciPy libraries in Python, and output the target mountain elevation field data T. .

[0017] Preferably, in step S101, the target mountain area environmental data D(x,y) is obtained, wherein the data format of the target mountain area environmental data D(x,y) is elevation DEM data, x is the abscissa, and y is the ordinate; based on the target mountain area environmental data... A mutation feature operator combining the first and second derivatives of space is used to extract boundary-sensitive regions, outputting a set of mutation region masks. The steps specifically include:

[0018] Step S1011: Based on the environmental data of the target mountainous area Calculate the local gradient field and the local gradient field As a first-order topographic abrupt change indicator; based on environmental data of the target mountainous area. Calculate the Laplace operator response and the Laplace operator response As an indicator of second-order curvature abrupt change;

[0019] Step S1012: Construct a catastrophe response function based on the first-order terrain catastrophe index and the second-order curvature catastrophe index using a weighted linear superposition method. , ,in, This is an edge intensity adjustment factor; As a modulatory factor for the sensitivity of landform curvature;

[0020] Step S1013: Set the mutation response threshold And based on the mutation response threshold Mutation response function Constructing a set of mutation region masks using a threshold filtering method , .

[0021] Preferably, in step S20, the elevation field data T of the target mountain area is used as the basis for the calculation. A slope-guided progressive base partitioning mechanism is used to construct the micro-topographic tiered base, outputting the micro-topographic tiered base sequence H. The steps specifically include:

[0022] Step S201: Based on the elevation field data T of the target mountain area The gradient field of the target mountain area was calculated. , Then, based on the gradient field of the target mountain area The slope amplitude of the target mountain area was calculated. , Preset local slope threshold Based on the elevation field data T of the target mountain area China satisfies Regional construction target mountainous base area ; Target mountainous base area Used to eliminate areas with excessively steep mountain slopes;

[0023] Step S202: In the base area of ​​the target mountainous region The area is further subdivided into n equal-slope sub-blocks. For each of the n equal-slope sub-blocks, the slope angle θ(x,y) is extracted. Based on the slope angle θ(x,y), a principal direction set is constructed using the maximum directional consistency clustering principle. Based on the main direction set The base elevation set is generated using slope aspect principal axis projection dimensionality reduction and local minimum elevation estimation; among which, the set of principal directions is used to generate the base elevation set. The steps for generating the base elevation set using the principal axis projection dimensionality reduction and local minimum elevation estimation method specifically include: dividing the n iso-slope sub-blocks according to the principal direction set. Projecting in the direction of the axis to form a one-dimensional projection axis; using the one-dimensional projection axis as the horizontal axis, calculating and extracting the minimum elevation value in each preset equally spaced segment, and finally generating the base elevation set;

[0024] Step S203: Based on the base elevation set, perform micro-hierarchical clustering using the minimum variance fitting principle to output the micro-topographic stepped base sequence H. .

[0025] Preferably, in step S30, based on the micro-topographic tiered base sequence H The initial building block growth task is performed using a terrain factor-driven multi-scale morphological growth mechanism, and the output is a climate response block set. The steps specifically include:

[0026] Step S301: Topographic factor set extraction stage: Based on the micro-topographic tiered base sequence H Construct the set of terrain factors corresponding to each level of the trapezoidal surface j. , ,in, Here is the slope data corresponding to the trapezoidal surface j; Here is the slope aspect data corresponding to the trapezoidal surface j; For the concave / convex trend data corresponding to the trapezoidal surface j; The annual average total sunshine duration data corresponding to surface j is estimated using a pre-set GIS sunshine duration model.

[0027] Step S302: Initial morphological growth stage: targeting the set of terrain factors Set factor weight set Based on the set of terrain factors and factor weight set A multi-factor weighted scoring method based on entropy weighting and hierarchical fusion is used to perform a comprehensive evaluation of terrain suitability, and the terrain suitability score is output. Based on terrain suitability score Implement different growth strategies:

[0028] Step S303: Result Output Stage: After executing different growth strategies, output the climate response block set. .

[0029] Preferably, in step S302, the terrain suitability score is used as the basis for the calculation. The steps for implementing different growth strategies include:

[0030] when At that time, a staggered, stepped growth strategy of equal height was adopted;

[0031] when At that time, a contour-following block array growth strategy that conforms to the curved surface contour is adopted;

[0032] when At that time, a growth strategy was adopted that prioritizes extending the axial direction of the block in the direction of the slope towards the light-facing side;

[0033] when At that time, a growth strategy was adopted that prioritized placing the blocks in areas with high sunlight.

[0034] Preferably, in step S40, based on the climate response block set A multi-objective fuzzy optimization mechanism is used to perform the three-performance evaluation task, and the three-performance operator set is output. The steps specifically include:

[0035] Step S401: Construct the first fuzzy target Second fuzzy target and the third fuzzy target The first fuzzy target Used to represent the target of rainfall dissipation efficiency, the second fuzzy target. Used to represent ventilation infiltration efficiency targets, the third fuzzy target. Used to represent solar radiation receiving targets; based on the first fuzzy target. Second fuzzy target and the third fuzzy target Construct a target vector for energy with three properties;

[0036] Step S402: Obtain the set of climate response blocks The corresponding climate-sensitive factors include average wind speed, rainfall intensity, and sunshine intensity. A climate-sensitive vector is constructed based on the average wind speed, rainfall intensity, and sunshine intensity factors.

[0037] Based on the three energy target vectors and climate sensitivity vectors, a fuzzy hierarchical comprehensive evaluation method is used to calculate fuzzy similarity and output multi-objective matching scores. ;

[0038] Step S403: Based on multi-objective matching score A fuzzy three-performance energy assessment matrix is ​​constructed, and a fuzzy hierarchical relationship evaluation method is used to perform weighted synthesis processing based on the fuzzy three-performance energy assessment matrix, outputting a set of three performance operators. .

[0039] This invention also provides a green intelligent building planning and design system based on multimodal data, comprising:

[0040] The terrain noise adaptive suppression module is used to acquire environmental data of the target mountainous area. Based on environmental data of the target mountainous area An adaptive elevation noise suppression task is performed using a mechanism based on the combined use of terrain-sensitive area labeling and dynamic filtering, outputting elevation field data T for the target mountainous area. Among them, environmental data of the target mountainous area The data format is elevation DEM data, with x as the horizontal axis and y as the vertical axis;

[0041] The micro-topography tiered base construction module is used to construct bases based on the elevation field data T of the target mountainous area. A slope-guided progressive base partitioning mechanism is used to construct the micro-topographic tiered base, outputting the micro-topographic tiered base sequence H. ;

[0042] The terrain-driven initial block generation module is used to generate blocks based on the micro-topography tiered base sequence H. The initial building block growth task is performed using a terrain factor-driven multi-scale morphological growth mechanism, and the output is a climate response block set. ;

[0043] The three-performance fuzzy evaluation module is used for climate response block sets. A multi-objective fuzzy optimization mechanism is used to perform the three-performance evaluation task, and the three-performance operator set is output. ;

[0044] The three-objective iterative optimization module is used to optimize based on three performance operators. A three-objective approximation optimization task is performed using an alternating iterative mechanism driven by multiple objectives, and the final shape combination scheme is output. .

[0045] The present invention also provides a green intelligent building planning and design device based on multimodal data, comprising: a memory, a processor, and a green intelligent building planning and design program based on multimodal data stored in the memory and executable on the processor. When the green intelligent building planning and design program based on multimodal data is executed by the processor, it implements a green intelligent building planning and design method based on multimodal data.

[0046] The present invention also provides a computer program product, including a green and intelligent building planning and design program based on multimodal data, wherein the green and intelligent building planning and design program based on multimodal data implements the green and intelligent building planning and design method based on multimodal data when executed by a processor.

[0047] The beneficial effects of this invention are as follows: By introducing an elevation noise suppression mechanism based on the combination of terrain change sensitive area marking and dynamic filtering, and a slope-guided micro-topographic tiered base construction mechanism, this invention can accurately adapt to complex mountain terrain features and improve the integration of building form with the original landform. It is especially suitable for architectural planning in typical mountain scenic areas such as Jiuzhaigou.

[0048] This invention utilizes a multi-scale morphological growth mechanism and a fuzzy multi-objective optimization strategy to achieve a balance between rainwater drainage efficiency, shading performance, and the integrity of visual corridors. This avoids the performance trade-offs caused by local optimization or reliance on experience-based adjustments in traditional solutions, thereby significantly improving the automation level and overall environmental adaptability of building form design. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 This is a flowchart illustrating the first embodiment of a green and intelligent building planning and design method based on multimodal data according to the present invention.

[0051] Figure 2 This is a schematic diagram of the original elevation field of the first embodiment of a green intelligent building planning and design method based on multimodal data according to the present invention.

[0052] Figure 3 This is a detailed partitioned region diagram of the first embodiment of a green and intelligent building planning and design method based on multimodal data according to the present invention.

[0053] Figure 4 This is a schematic diagram of the building block area before growth, representing a first embodiment of a green intelligent building planning and design method based on multimodal data according to the present invention.

[0054] Figure 5 This is a schematic diagram of the area after the building block has grown, representing a first embodiment of the green intelligent building planning and design method based on multimodal data according to the present invention.

[0055] Figure 6 This is a schematic diagram of the equipment for a green and intelligent building planning and design method based on multimodal data according to the present invention. Detailed Implementation

[0056] 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. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0057] Example 1: As Figure 1 The diagram shown is a flowchart of the first embodiment of the green and intelligent building planning and design method based on multimodal data of the present invention, which presents the first embodiment of the green and intelligent building planning and design method based on multimodal data of the present invention.

[0058] In the first embodiment, the green intelligent building planning and design method based on multimodal data includes:

[0059] Step S10: Obtain environmental data for the target mountainous area Based on environmental data of the target mountainous area An adaptive elevation noise suppression task is performed using a mechanism based on the combined use of terrain-sensitive area labeling and dynamic filtering, outputting elevation field data T for the target mountainous area. Among them, environmental data of the target mountainous area The data format is elevation DEM data, with x as the horizontal axis and y as the vertical axis;

[0060] It should be noted that the "mechanism based on the synergy of terrain change sensitive area labeling and dynamic filtering" refers to identifying terrain change sensitive areas with high elevation differences, strong slope breaks, or drastic changes in aspect by constructing second-order terrain derivative fields such as slope, curvature, and contour density in the elevation DEM data. These areas are then labeled as "high-sensitivity terrain areas" using a dynamic labeling mechanism. Subsequently, a weighted dynamic filtering strategy is introduced for the entire DEM elevation data. Conventional smoothing filtering is used in low-sensitivity areas, while the filter kernel scale and direction are adaptively adjusted in high-sensitivity areas. This removes local noise while preserving the true terrain change information and micro-topographic features to the greatest extent possible.

[0061] Understandably, by integrating the two sub-mechanisms of "terrain structure difference recognition" and "dynamic filtering adaptive control", the problems of edge blurring and landform weakening caused by global mean filtering in traditional DEM processing are effectively avoided. This enhances the accuracy of restoring complex terrain structures in mountainous areas (such as canyon slopes, micro-passes, natural terraces, etc.) and provides more stable and reliable elevation field input data for subsequent tiered base subdivision and shape growth.

[0062] Step S20: Based on the elevation field data T of the target mountainous area A slope-guided progressive base partitioning mechanism is used to construct the micro-topographic tiered base, outputting the micro-topographic tiered base sequence H. ;

[0063] It should be noted that the "slope-guided progressive base partitioning mechanism" refers to first calculating the slope gradient field of each grid cell on the elevation field data, and then normalizing the slope gradient field to form a slope guiding factor field; then constructing initial coarse-grained planar partitioning cells; in each iteration, based on the slope guiding factor field and its local rate of change, identifying areas with large slope variations as "candidate refinement boundary zones", and performing partitioning operations within these zones until the slope variance within all cells is less than a threshold, ultimately forming a terrain-sensitive and adaptive stepped partitioning structure.

[0064] Understandably, by introducing a slope gradient field as a partitioning control guide and combining it with a locally adaptive partitioning threshold control mechanism, the base partitioning structure is closely coupled with changes in landform morphology, significantly improving the basic adaptability of micro-topography to building shape fitting. At the same time, it avoids problems such as "overly dense partitioning" or "insufficient adaptability" that occur in traditional regular grid partitioning, providing a natural and continuous contact surface for subsequent building block seed placement.

[0065] It should be understood that, compared with the static partitioning strategies commonly used in existing methods, which employ fixed-size grid division or vertical or horizontal linear segmentation, the progressive partitioning mechanism adopted in this step has the following advantages: First, the partitioning direction is dominated by the slope gradient, adaptively adjusting the partitioning direction and scale to better conform to the natural terrain trend; second, the introduction of a slope change rate threshold control avoids excessive refinement in areas with gentle terrain; and third, the resulting base structure has elevation continuity and boundary splicing capability, improving the connection stability and constructability of the building seed body.

[0066] For example, such as Figure 2 As shown, for elevation field data of a target mountainous area, a slope gradient field is first generated based on slope gradient calculation, where areas of abrupt terrain change exhibit high grayscale peaks. Then, candidate refined regions are constructed within the boundary zones where slope changes drastically, and local progressive subdivision is performed on them. For example... Figure 3 As shown, a higher resolution grid structure is formed in areas with high slope variations, while coarse-grained subdivision is maintained in areas with gentle terrain, thereby outputting a micro-topographic tiered base sequence with geomorphic sensitivity adaptation characteristics. This structure not only outperforms traditional regular subdivision schemes in terms of boundary continuity and slope response accuracy, but also significantly improves the naturalness of the contact surface and the structural integration stability during building seed generation.

[0067] Step S30: Based on the micro-topographic stepped base sequence H The initial building block growth task is performed using a terrain factor-driven multi-scale morphological growth mechanism, and the output is a climate response block set. ;

[0068] It should be noted that the "multi-scale morphological growth mechanism driven by terrain factors" means that this step does not directly generate homogenized or randomized building blocks on the micro-topographic base. Instead, it introduces multiple sets of terrain factors as driving constraint parameters. In actual execution, the above factor vectors are assigned to each level of micro-topographic base. The starting position and growth mode of the initial building unit are determined according to its weighted result. A dynamic scale advancement strategy is adopted to complete the extended modeling from "growth seed" to "building form".

[0069] Understandably, by introducing a terrain factor-driven mechanism, the spatial distribution of building blocks is highly coupled with terrain features. This approach effectively identifies areas with moderate slope, reasonable aspect, and stable curvature as priority deployment zones, forming initial building forms that meet both climate response requirements and structural feasibility. Compared to traditional uniform deployment methods based on planar grids, this method achieves a dual improvement in terrain adaptability and morphological diversity, making it particularly suitable for generating naturally integrated buildings in complex mountainous scenarios.

[0070] It should be understood that traditional building generation strategies often employ fixed templates or evenly spaced layouts, neglecting the impact of micro-topographical variations on building stability and climate adaptability. This method, however, utilizes a composite driving approach of slope dominance, aspect guidance, and curvature adjustment to create a growth mechanism with clear generation rules and multi-dimensional evaluation constraints: areas with excessively steep slopes automatically inhibit block growth, reducing the risk of structural instability due to terrain instability; blocks in south-facing or main ventilation directions are easier to grow, automatically improving ventilation and lighting efficiency; boundary areas with large curvature changes are considered non-buildable zones, with block edges naturally converging, avoiding manual truncation. This method demonstrates a closed-loop structure of "process-driven, factor-adjusted, and scale-controlled" in the automatic building form generation process, significantly improving the accuracy and engineering practicality of intelligent generation.

[0071] For example, such as Figure 4 As shown, this illustrates the micro-topographical base state "before the building mass grows," containing only naturally undulating landform features; such as... Figure 5 The diagram shows the state of the building block after growth. It can be observed that the building block mainly grows along areas with moderate slope and reasonable slope aspect, while the edges naturally converge to the terrain boundary without abrupt truncation. This block distribution process simulates the morphological evolution process under the constraints of slope and the guidance of direction, demonstrating the precise control of the building space morphology and the terrain integration characteristics of the "multi-scale growth mechanism driven by terrain factors" in this invention.

[0072] Step S40: Based on the climate response block set A multi-objective fuzzy optimization mechanism is used to perform the three-performance evaluation task, and the three-performance operator set is output. ;

[0073] It should be noted that the "three-performance evaluation task" refers to the comprehensive optimization and judgment of three key indicators—"ventilation performance," "lighting performance," and "shading performance"—based on the external shape, orientation angle, and spatial layout of each initial building block. In actual implementation, the above performance indicator functions are mapped to a set of objective functions for a multi-objective fuzzy optimization problem. Fuzzy membership functions are used to express the evaluation strength of each performance's adaptability to the current building block. Finally, a set of three performance operators is output to guide subsequent adjustments to the building blocks through fuzzy weighting or Pareto sorting. This mechanism, by introducing fuzzy logic, can effectively accommodate environmental fluctuations, imprecise inputs, and fuzzy performance boundaries, improving the evaluation model's ability to distinguish differences in building performance under complex climatic scenarios.

[0074] Understandably, this step, driven by multiple performance objectives and incorporating fuzzy optimization principles, introduces an uncertainty reconciliation mechanism into the performance evaluation process, avoiding the fragmented performance evaluation caused by boundary decisions in traditional rigid indicator systems. Especially in complex mountainous environments with drastic changes in solar cycles, conflicting objectives often exist among the three performance metrics (such as ventilation versus shading). Traditional linear superposition or single-objective optimization can easily lead to overfitting of a single performance metric, neglecting the overall adaptability of the building. By using a fuzzy membership model to normalize the performance indicators and dynamically adjusting the weights to center on the optimal adaptation region, a performance guidance vector that better aligns with the building's actual climate response strategy can be output.

[0075] Step S50: Based on the three performance operator sets A three-objective approximation optimization task is performed using an alternating iterative mechanism driven by multiple objectives, and the final shape combination scheme is output. .

[0076] It should be noted that the "multi-objective driven alternating iterative mechanism" in this step refers to the dynamic switching of the dominant objective in each iteration, addressing the potential conflicts among the three performance indicators of building blocks—sunlight and ventilation, thermal stability, and visual landscape—and continuously advancing the structural adjustment and spatial configuration optimization of the block combination based on a multi-objective approximation strategy. This mechanism typically uses a set of three performance operators as the initial evaluation benchmark, introduces a weighted Pareto front approximation and fuzzy weight adjustment strategy, and sequentially executes the local optimum solution of the objective function with one performance as the dominant factor and the others as constraints, ultimately converging to the building form combination with the optimal overall performance.

[0077] Understandably, by employing a multi-objective alternating driving mechanism, this step fully considers the mutual influence between climate response blocks across the three performance dimensions during the optimization process. The dominant performance weights for each round are automatically adjusted based on the feedback from the previous round's evaluation, avoiding overall performance degradation caused by single-objective optimization. Simultaneously, the embedding of fuzzy logic enables flexible coordination between performance objectives, ensuring that the final combined solution exhibits good adaptability under various climate and terrain scenarios.

[0078] For example, taking a residential planning scenario in a mountainous area as an example, the process first prioritizes maximizing sunlight coverage, evaluating the daylight efficiency of the initial block combination scheme, and adjusting the block positions to avoid mutual shading effects. Then, prioritizing thermal stability, the energy consumption levels caused by shading and changes in block thickness are assessed, and structural thickening or skin modification is implemented for south-facing buildings. Next, prioritizing visual landscape openness, the visibility of each residential unit towards the valley or water body is evaluated, and the positions of high-rise blocks are adjusted to avoid view obstruction. After three rounds of iteration, a set of candidate schemes at the Pareto frontier is formed. Finally, based on a fuzzy rule evaluation system, the final form combination scheme with the best overall performance is selected, outperforming traditional single-objective optimization strategies.

[0079] Example 2: Furthermore, the present invention provides a green intelligent building planning and design system based on multimodal data, employing a green intelligent building planning and design method based on multimodal data from the above embodiments, which can solve a technical problem in green intelligent building planning and design based on multimodal data. Compared with the prior art, the beneficial effects of the green intelligent building planning and design system based on multimodal data provided by the present invention are the same as those of the green intelligent building planning and design method based on multimodal data provided in the above embodiments, and other technical features of the green intelligent building planning and design system based on multimodal data are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0080] Example 3: This invention provides a green intelligent building planning and design device based on multimodal data. Please refer to... Figure 6A green intelligent building planning and design device based on multimodal data includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, enable the at least one processor to execute the green intelligent building planning and design method based on multimodal data described in Embodiment 1 above. The green intelligent building planning and design device based on multimodal data in this embodiment of the invention may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), and in-vehicle terminals (e.g., in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. This green intelligent building planning and design device based on multimodal data is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the invention. A green intelligent building planning and design device based on multimodal data may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory 1002 or a program loaded from a storage device 1003 into a random access memory 1004. The random access memory 1004 also stores various programs and data required for the operation of the green intelligent building planning and design device based on multimodal data. The processing unit 1001, the read-only memory 1002, and the random access memory 1004 are interconnected via a bus 1005. An I / O interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 1003 including, for example, magnetic tape, hard disk, etc.; and communication devices 1009. The communication device 1009 allows a green intelligent building planning and design device based on multimodal data to communicate wirelessly or wiredly with other devices to exchange data. While the figure shows a green intelligent building planning and design device based on multimodal data with various systems, it should be understood that implementation or possession of all the systems shown is not required. More or fewer systems may be implemented alternatively.

[0081] Example 4: This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the green intelligent building planning and design method based on multimodal data as described above. The computer program product provided by this invention can solve the technical problem of green intelligent building planning and design based on multimodal data. Compared with the prior art, the beneficial effects of the computer program product provided by this invention are the same as those of the green intelligent building planning and design method based on multimodal data provided in the above embodiments, and will not be repeated here.

[0082] In particular, according to the embodiments disclosed in this invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from read-only memory 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this invention.

[0083] It should be understood that the various parts disclosed in this invention can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics may be combined in any suitable manner in one or more embodiments or examples.

[0084] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A green and intelligent building planning and design method based on multimodal data, characterized in that, The methods include: Step S10: Obtain environmental data for the target mountainous area Based on environmental data of the target mountainous area An adaptive elevation noise suppression task is performed using a mechanism based on the combined use of terrain-sensitive area labeling and dynamic filtering, outputting elevation field data T for the target mountainous area. Among them, environmental data of the target mountainous area The data format is elevation DEM data, with x as the horizontal axis and y as the vertical axis; specifically, it includes: Obtain environmental data of the target mountainous area Based on environmental data of the target mountainous area A mutation feature operator combining the first and second derivatives of space is used to extract boundary-sensitive regions, outputting a set of mutation region masks. ; For mutation region mask set For the region outside the range, a 3×3 mean filter is used for smoothing, and the noise estimation function for the first region is output. ; for the set of masks for mutation regions Within the region, a combination of directional adaptive Sobel and Fourier filters is used for local directional consistency filtering, outputting a noise estimation function for the second region. Based on the noise estimation function of the first region Second region noise estimation function Define the final localization noise estimation function δ(x,y). ; Based on the regionalized noise estimation function δ(x,y), the NumPy and SciPy libraries in Python are used to perform regional difference filtering on the environmental data D(x,y) of the target mountain area, and output the elevation field data T of the target mountain area. ; Step S20: Based on the elevation field data T of the target mountainous area A slope-guided progressive base partitioning mechanism is used to construct the micro-topographic tiered base, outputting the micro-topographic tiered base sequence H. ; Step S30: Based on the micro-topographic stepped base sequence H The initial building block growth task is performed using a terrain factor-driven multi-scale morphological growth mechanism, and the output is a climate response block set. ; Step S40: Based on the climate response block set A multi-objective fuzzy optimization mechanism is used to perform the three-performance evaluation task, and the three-performance operator set is output. ; Step S50: Based on the three performance operator sets A three-objective approximation optimization task is performed using an alternating iterative mechanism driven by multiple objectives, and the final shape combination scheme is output. .

2. The green intelligent building planning and design method based on multimodal data as described in claim 1, characterized in that, In step S101, the target mountain area environmental data D(x,y) is obtained, wherein the data format of the target mountain area environmental data D(x,y) is elevation DEM data, x is the abscissa, and y is the ordinate; based on the target mountain area environmental data... A mutation feature operator combining the first and second derivatives of space is used to extract boundary-sensitive regions, outputting a set of mutation region masks. The steps specifically include: Step S1011: Based on the environmental data of the target mountainous area Calculate the local gradient field and the local gradient field As a first-order topographic abrupt change indicator; based on environmental data of the target mountainous area. Calculate the Laplace operator response and the Laplace operator response As an indicator of second-order curvature abrupt change; Step S1012: Construct a catastrophe response function based on the first-order terrain catastrophe index and the second-order curvature catastrophe index using a weighted linear superposition method. , ,in, This is an edge intensity adjustment factor; As a modulatory factor for the sensitivity of landform curvature; Step S1013: Set the mutation response threshold And based on the mutation response threshold Mutation response function Constructing a set of mutation region masks using a threshold filtering method , .

3. The green intelligent building planning and design method based on multimodal data as described in claim 1, characterized in that, In step S20, based on the elevation field data T of the target mountainous area A slope-guided progressive base partitioning mechanism is used to construct the micro-topographic tiered base, outputting the micro-topographic tiered base sequence H. The steps specifically include: Step S201: Based on the elevation field data T of the target mountain area The gradient field of the target mountain area was calculated. , Then, based on the gradient field of the target mountain area The slope amplitude of the target mountain area was calculated. , Preset local slope threshold Based on the elevation field data T of the target mountain area China satisfies Regional construction target mountainous base area ; Target mountainous base area Used to eliminate areas with excessively steep mountain slopes; Step S202: In the base area of ​​the target mountainous region The area is further subdivided into n equal-slope sub-blocks. For each of the n equal-slope sub-blocks, the slope angle θ(x,y) is extracted. Based on the slope angle θ(x,y), a principal direction set is constructed using the maximum directional consistency clustering principle. Based on the main direction set The base elevation set is generated using slope aspect principal axis projection dimensionality reduction and local minimum elevation estimation; among which, the set of principal directions is used to generate the base elevation set. The steps for generating the base elevation set using the principal axis projection dimensionality reduction and local minimum elevation estimation method specifically include: dividing the n iso-slope sub-blocks according to the principal direction set. Projecting in the direction of the axis to form a one-dimensional projection axis; using the one-dimensional projection axis as the horizontal axis, calculating and extracting the minimum elevation value in each preset equally spaced segment, and finally generating the base elevation set; Step S203: Based on the base elevation set, perform micro-hierarchical clustering using the minimum variance fitting principle to output the micro-topographic stepped base sequence H. .

4. The green intelligent building planning and design method based on multimodal data as described in claim 1, characterized in that, In step S30, based on the micro-topographic tiered basement sequence H The initial building block growth task is performed using a terrain factor-driven multi-scale morphological growth mechanism, and the output is a climate response block set. The steps specifically include: Step S301: Topographic factor set extraction stage: Based on the micro-topographic tiered base sequence H Construct the set of terrain factors corresponding to each level of the trapezoidal surface j. , ,in, Here is the slope data corresponding to the trapezoidal surface j; Here is the slope aspect data corresponding to the trapezoidal surface j; For the concave / convex trend data corresponding to the trapezoidal surface j; The annual average total sunshine duration data corresponding to surface j is estimated using a pre-set GIS sunshine duration model. Step S302: Initial morphological growth stage: targeting the set of terrain factors Set factor weight set Based on the set of terrain factors and factor weight set A multi-factor weighted scoring method based on entropy weighting and hierarchical fusion is used to perform a comprehensive evaluation of terrain suitability, and the terrain suitability score is output. Based on terrain suitability score Implement different growth strategies: Step S303: Result Output Stage: After executing different growth strategies, output the climate response block set. .

5. The green intelligent building planning and design method based on multimodal data as described in claim 4, characterized in that, In step S302, the terrain suitability score is used as the basis for the calculation. The steps for implementing different growth strategies include: when At that time, a staggered, stepped growth strategy of equal height was adopted; when At that time, a contour-following block array growth strategy that conforms to the curved surface contour is adopted; when At that time, a growth strategy was adopted that prioritizes extending the axial direction of the block in the direction of the slope towards the light-facing side; when At that time, a growth strategy was adopted that prioritized placing the blocks in areas with high sunlight.

6. The green intelligent building planning and design method based on multimodal data as described in claim 1, characterized in that, In step S40, based on the climate response block set A multi-objective fuzzy optimization mechanism is used to perform the three-performance evaluation task, and the three-performance operator set is output. The steps specifically include: Step S401: Construct the first fuzzy target Second fuzzy target and the third fuzzy target The first ambiguous target Used to represent the target of rainfall dissipation efficiency, the second fuzzy target. Used to represent ventilation infiltration efficiency targets, the third fuzzy target. Used to represent solar radiation receiving targets; based on the first fuzzy target. Second fuzzy target and the third fuzzy target Construct a target vector for energy with three properties; Step S402: Obtain the climate response block set The corresponding climate-sensitive factors include average wind speed, rainfall intensity, and sunshine intensity. A climate-sensitive vector is constructed based on the average wind speed, rainfall intensity, and sunshine intensity factors. Based on the three energy target vectors and climate sensitivity vectors, a fuzzy hierarchical comprehensive evaluation method is used to calculate fuzzy similarity and output multi-objective matching scores. ; Step S403: Based on multi-objective matching score A fuzzy three-performance energy assessment matrix is ​​constructed, and a fuzzy hierarchical relationship evaluation method is used to perform weighted synthesis processing based on the fuzzy three-performance energy assessment matrix, outputting a set of three performance operators. .

7. A green and intelligent building planning and design system based on multimodal data, applied to the green and intelligent building planning and design method based on multimodal data as described in any one of claims 1 to 6, characterized in that, The green and intelligent building planning and design system based on multimodal data includes: The terrain noise adaptive suppression module is used to acquire environmental data of the target mountainous area. Based on environmental data of the target mountainous area An adaptive elevation noise suppression task is performed using a mechanism based on the combined use of terrain-sensitive area labeling and dynamic filtering, outputting elevation field data T for the target mountainous area. Among them, environmental data of the target mountainous area The data format is elevation DEM data, with x as the horizontal axis and y as the vertical axis; specifically, it includes: Obtain environmental data of the target mountainous area Based on environmental data of the target mountainous area A mutation feature operator combining the first and second derivatives of space is used to extract boundary-sensitive regions, outputting a set of mutation region masks. ; For mutation region mask set For the region outside the range, a 3×3 mean filter is used for smoothing, and the noise estimation function for the first region is output. ; for the set of masks for mutation regions Within the region, a combination of directional adaptive Sobel and Fourier filters is used for local directional consistency filtering, outputting a noise estimation function for the second region. Based on the noise estimation function of the first region Second region noise estimation function Define the final localization noise estimation function δ(x,y). ; Based on the regionalized noise estimation function δ(x,y), the NumPy and SciPy libraries in Python are used to perform regional difference filtering on the environmental data D(x,y) of the target mountain area, and output the elevation field data T of the target mountain area. ; The micro-topography tiered base construction module is used to construct bases based on the elevation field data T of the target mountainous area. A slope-guided progressive base partitioning mechanism is used to construct the micro-topographic tiered base, outputting the micro-topographic tiered base sequence H. ; The terrain-driven initial block generation module is used to generate blocks based on the micro-topography tiered base sequence H. The initial building block growth task is performed using a terrain factor-driven multi-scale morphological growth mechanism, and the output is a climate response block set. ; The three-performance fuzzy evaluation module is used for climate response block sets. A multi-objective fuzzy optimization mechanism is used to perform the three-performance evaluation task, and the three-performance operator set is output. ; The three-objective iterative optimization module is used to optimize based on three performance operators. A three-objective approximation optimization task is performed using an alternating iterative mechanism driven by multiple objectives, and the final shape combination scheme is output. .

8. A green intelligent building planning and design device based on multimodal data, characterized in that, The green intelligent building planning and design equipment based on multimodal data includes: a memory, a processor, and a green intelligent building planning and design program based on multimodal data stored in the memory and executable on the processor. When the green intelligent building planning and design program based on multimodal data is executed by the processor, it implements a green intelligent building planning and design method based on multimodal data according to any one of claims 1 to 6.

9. A computer program product, characterized in that, The computer program product includes a green and intelligent building planning and design program based on multimodal data. When the green and intelligent building planning and design program based on multimodal data is executed by a processor, it implements a green and intelligent building planning and design method based on multimodal data as described in any one of claims 1 to 6.

Images