A method and system for optimizing a layout of a special-shaped curtain wall structure

Abstract

The present application relates to the field of architectural design, and particularly relates to a special-shaped curtain wall structure layout optimization method and system. A target special-shaped space to be subjected to curtain wall structure layout is acquired, and first curtain wall layout information is randomly generated, wherein the first curtain wall layout information comprises curtain wall structure information; a first special-shaped space approximation degree is calculated and acquired according to the first curtain wall layout information and the target special-shaped space; a first cleaning feasibility parameter and a second cleaning feasibility parameter are acquired by performing cleaning feasibility analysis according to the first curtain wall layout information and the first special-shaped space approximation degree; a first layout score is calculated according to the first special-shaped space approximation degree, the first cleaning feasibility parameter and the second cleaning feasibility parameter, and optimization of the curtain wall layout information is performed to acquire optimal curtain wall layout information. The problems of difficult layout optimization and high later cleaning cost of the special-shaped curtain wall caused by the complex shape are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of architectural design, and in particular to a method and system for optimizing the layout of a special-shaped curtain wall structure. BACKGROUND

[0002] With the development of the construction industry, more and more buildings are designed with special shapes, which have large planar dimensions, multiple curtain wall systems, and complex spatial modeling. To address these characteristics, the industry has gradually applied technologies such as prefabrication, unitization, and informatization to optimize curtain wall processes and construction methods. At the same time, automated tools such as drones have been used for curtain wall cleaning to improve maintenance efficiency. However, the complex spatial form of special-shaped spaces makes it difficult for traditional curtain wall layout design to accurately fit the space edges, leading to problems such as material waste and insufficient structural stability. Moreover, conventional curtain wall layout design does not fully consider cleaning feasibility, especially in terms of the operating characteristics of tools such as drones, resulting in unreasonable cleaning intervals and low cleaning rates, which increases the cost of later maintenance. In summary, the existing technology has technical problems such as difficulty in layout optimization and high cost of later cleaning due to the complex form of special-shaped curtain walls. SUMMARY

[0003] The present application provides a method and system for optimizing the layout of a special-shaped curtain wall structure to address the technical problems of difficulty in layout optimization and high cost of later cleaning due to the complex form of special-shaped curtain walls in the prior art.

[0004] The technical solution of the present application to solve the above technical problems is as follows:

[0005] In a first aspect, the present application provides a method for optimizing the layout of a special-shaped curtain wall structure, comprising: obtaining a target special-shaped space to be subjected to curtain wall structure layout, and randomly generating first curtain wall layout information, wherein the first curtain wall layout information includes curtain wall structure information; calculating a first special-shaped space approximation degree based on the first curtain wall layout information and the target special-shaped space; performing cleaning feasibility analysis based on the first curtain wall layout information and the first special-shaped space approximation degree to obtain first and second cleaning feasibility parameters; calculating a first layout score based on the first special-shaped space approximation degree, the first cleaning feasibility parameter, and the second cleaning feasibility parameter, optimizing the curtain wall layout information, and obtaining optimal curtain wall layout information.

[0006] Optionally, obtaining a target special-shaped space to be subjected to curtain wall structure layout and randomly generating first curtain wall layout information comprises: obtaining a target special-shaped space to be subjected to curtain wall structure layout, wherein the target special-shaped space includes a special-shaped space edge; and randomly generating first curtain wall layout information within the target special-shaped space, wherein the first curtain wall layout information includes curtain wall structure information, and the curtain wall structure information includes a plurality of curtain wall edges.

[0007] Optionally, the first special-shaped space approximation degree is calculated according to the first curtain wall layout information and the target special-shaped space, including: calculating distances between a plurality of curtain wall edges in the first curtain wall layout information and a special-shaped space edge in the target special-shaped space to obtain a plurality of first curtain wall distances; and calculating the first special-shaped space approximation degree of the first curtain wall layout information according to the plurality of first curtain wall distances.

[0008] The first special-shaped space approximation degree of the first curtain wall layout information is calculated according to the plurality of first curtain wall distances, including: calculating a mean value of the plurality of first curtain wall distances to obtain a first average curtain wall distance; calculating a maximum fluctuation amplitude of the plurality of first curtain wall distances, compensating the first average curtain wall distance to obtain a first compensated curtain wall distance; obtaining a standard curtain wall distance; and calculating a similarity between the first compensated curtain wall distance and the standard curtain wall distance to obtain the first special-shaped space approximation degree.

[0009] Optionally, the cleaning feasibility analysis is performed according to the first curtain wall layout information and the first special-shaped space approximation degree to obtain first cleaning feasibility parameters and second cleaning feasibility parameters, including: obtaining a plurality of perpendicular line directions of a plurality of curtain wall edges away from the target special-shaped space according to the plurality of curtain wall edges in the first curtain wall layout information; extending the plurality of perpendicular line directions by a preset distance to obtain a plurality of curtain wall cleaning positions of the plurality of curtain wall edges; calculating distances between every two adjacent curtain wall cleaning positions to obtain a plurality of cleaning interval distances, and calculating a mean value to obtain an average cleaning interval distance; inputting the average cleaning interval distance into a first cleaning classification table to obtain the first cleaning feasibility parameters; and performing the cleaning feasibility analysis according to the first special-shaped space approximation degree to obtain the second cleaning feasibility parameters.

[0010] The first cleaning classification table is constructed by: collecting a sample cleaning interval distance set and collecting control accuracies under different sample cleaning interval distances according to historical data of curtain wall cleaning by using unmanned aerial vehicles to label a sample first cleaning feasibility parameter set; and constructing a mapping relationship between the sample cleaning interval distance set and the sample first cleaning feasibility parameter set to obtain the first cleaning classification table.

[0011] The second cleaning feasibility parameters are obtained by performing the cleaning feasibility analysis according to the first special-shaped space approximation degree, including: obtaining a second cleaning classification table, wherein the second cleaning classification table is constructed based on a mapping relationship between a sample special-shaped space approximation degree set and a sample second cleaning feasibility parameter set in historical curtain wall cleaning record data, and the sample second cleaning feasibility parameter includes a cleaning rate; and inputting the first special-shaped space approximation degree into the second cleaning classification table to classify and obtain the second cleaning feasibility parameters.

[0012] Optionally, according to the first special-shaped space approximation degree, the first cleaning feasibility parameter and the second cleaning feasibility parameter, a first layout score is calculated, optimization of the curtain wall layout information is performed, and optimal curtain wall layout information is obtained, including: according to the first special-shaped space approximation degree, the first cleaning feasibility parameter and the second cleaning feasibility parameter, a first layout score is calculated; curtain wall layout information is randomly generated, a layout score is calculated, iterative optimization is performed, and optimal curtain wall layout information is obtained.

[0013] In a second aspect, the present application provides a special-shaped curtain wall structure layout optimization system, comprising:

[0014] A curtain wall layout information acquisition module is configured to acquire a target special-shaped space to be subjected to curtain wall structure layout and randomly generate first curtain wall layout information, wherein the first curtain wall layout information comprises curtain wall structure information.

[0015] A special-shaped space approximation degree calculation module is configured to calculate a first special-shaped space approximation degree according to the first curtain wall layout information and the target special-shaped space.

[0016] A curtain wall cleaning difficulty evaluation module is configured to perform cleaning feasibility analysis according to the first curtain wall layout information and the first special-shaped space approximation degree, and obtain a first cleaning feasibility parameter and a second cleaning feasibility parameter.

[0017] A curtain wall layout information optimization module is configured to calculate a first layout score according to the first special-shaped space approximation degree, the first cleaning feasibility parameter and the second cleaning feasibility parameter, perform optimization of the curtain wall layout information, and obtain optimal curtain wall layout information.

[0018] By implementing the present application, the target special-shaped space to be subjected to curtain wall structure layout can be acquired, and first curtain wall layout information can be randomly generated to provide basic data and an initial scheme for subsequent optimization. The random generation can cover more potential layout possibilities and avoid optimization limitations caused by a single initial scheme.

[0019] By implementing the present application, the first special-shaped space approximation degree can be calculated according to the first curtain wall layout information and the target special-shaped space, the matching degree of the curtain wall layout and the special-shaped space can be quantified, the layout can be ensured to meet the design requirements in terms of form, and material waste or unreasonable structure caused by a large deviation between the layout and the space form can be reduced.

[0020] By implementing the present application, the first curtain wall layout information and the first special-shaped space approximation degree are used for cleaning feasibility analysis, the first cleaning feasibility parameter and the second cleaning feasibility parameter are obtained, the operation ability of the cleaning tool such as the unmanned aerial vehicle is considered from the actual feasibility of the cleaning operation, the influence of the layout on the cleaning efficiency is evaluated in advance, and the subsequent cleaning difficulty or high cost caused by the layout design is avoided.

[0021] By implementing the present application, the first curtain wall layout information and the first special-shaped space approximation degree are used for cleaning feasibility analysis, the first cleaning feasibility parameter and the second cleaning feasibility parameter are obtained, the operation ability of the cleaning tool such as the unmanned aerial vehicle is considered from the actual feasibility of the cleaning operation, the influence of the layout on the cleaning efficiency is evaluated in advance, and the subsequent cleaning difficulty or high cost caused by the layout design is avoided.

[0022] In summary, by implementing the present application, the design rationality and the construction and maintenance efficiency of the special-shaped curtain wall structure can be effectively improved, and the problems of difficult layout optimization and high later cleaning cost caused by the complex shape of the special-shaped curtain wall are solved. BRIEF DESCRIPTION OF DRAWINGS

[0023] Figure 1 A flowchart of a special-shaped curtain wall structure layout optimization method provided by the present application is shown.

[0024] Figure 2 A structural diagram of a special-shaped curtain wall structure layout optimization system provided by the present application is shown.

[0025] In the drawings, the components represented by the numbers are as follows:

[0026] The curtain wall layout information acquisition module 11, the special-shaped space approximation degree calculation module 12, the curtain wall cleaning difficulty evaluation module 13, and the curtain wall layout information optimization module 14. DETAILED DESCRIPTION

[0027] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0028] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0029] In the description of this invention, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0030] Example 1, as Figure 1 As shown, this embodiment of the invention provides a method for optimizing the layout of irregularly shaped curtain wall structures, including:

[0031] S100: Obtain the target irregular space for the curtain wall structure layout to be carried out, and randomly generate the first curtain wall layout information, wherein the first curtain wall layout information includes curtain wall structure information.

[0032] S200: Calculate the first irregular space approximation degree based on the first curtain wall layout information and the target irregular space;

[0033] S300: Based on the first curtain wall layout information and the first irregular space approximation degree, perform a cleaning feasibility analysis to obtain the first cleaning feasibility parameter and the second cleaning feasibility parameter;

[0034] S400: Based on the first irregular space approximation degree, the first cleaning feasibility parameter, and the second cleaning feasibility parameter, calculate the first layout score, optimize the curtain wall layout information, and obtain the optimal curtain wall layout information.

[0035] In step S100 of this application embodiment, the target irregular space for the curtain wall structure layout is obtained, and first curtain wall layout information is randomly generated, including:

[0036] Obtain the target irregular space for the curtain wall structure layout, wherein the target irregular space includes the edge of the irregular space;

[0037] generate first curtain wall layout information in the target special-shaped space, wherein the first curtain wall layout information comprises curtain wall structure information, and the curtain wall structure information comprises a plurality of curtain wall edges.

[0038] In the embodiments of the present application, step S100 is a basic preparation link for the optimization of the special-shaped curtain wall structure layout, and the core purpose is to provide initial data and space boundary reference for subsequent layout evaluation and optimization. Specifically, the target special-shaped space containing special-shaped space edges needs to be obtained to determine the range and form constraints of the curtain wall layout, so as to ensure that the subsequently generated curtain wall layout is always within the preset target special-shaped space, and to avoid the layout exceeding the space boundary or being out of line with the space form.

[0039] The first curtain wall layout information containing a plurality of curtain wall edges is randomly generated to provide initial scheme samples for the optimization process. The random generation method can cover more potential layout possibilities, reduce the optimization limitations caused by single initial scheme or subjective experience limitations, and lay a foundation for finding a global optimal scheme through iteration subsequently.

[0040] In the implementation process, first, the target special-shaped space needs to be obtained, and the key feature of the target special-shaped space is to contain clear special-shaped space edges, that is, the boundary contour of the space, such as the special-shaped contour line of the building facade.

[0041] Suppose that the appearance of a certain science and technology museum building is an irregular curved surface model, and the outer facade of the science and technology museum is an irregular space formed by a plurality of curved surfaces. Then, the target special-shaped space of the science and technology museum can be obtained by obtaining the outer facade three-dimensional model of the science and technology museum through BIM or other building three-dimensional modeling software. The three-dimensional model should contain clear special-shaped space edges, that is, the boundary contour line of the irregular curved surface, such as the top edge, the bottom edge, and the connecting edge with the main structure of the building, etc. These edges define the space range of the curtain wall layout.

[0042] At this time, the target special-shaped space is the irregular curved surface area of the outer facade of the science and technology museum, and its core feature is to contain all the special-shaped space edges described above as the boundary reference for the subsequent curtain wall layout.

[0043] Then, the first curtain wall layout information needs to be randomly generated. Within the determined target special-shaped space range, the initial curtain wall layout scheme, that is, the first curtain wall layout information, is randomly generated. The core of the first curtain wall layout information is the curtain wall structure information, which contains a plurality of curtain wall edges, that is, the boundary lines of the curtain wall, such as the splicing edges of the unit curtain wall, the division edges of different curtain wall systems, etc. The random generation process can randomly distribute the positions, angles, and other parameters of the curtain wall units in the space through an algorithm, so as to form an initial curtain wall edge combination as a starting point for subsequent evaluation.

[0044] For example, in the above-mentioned science museum example, the science museum adopts a glass curtain wall system, and the curtain wall layout needs to be spliced by multiple unit glass curtain walls. The edges of each unit are the curtain wall edges.

[0045] The specific random generation process can be that the positions and angles of the unit glass curtain walls are randomly distributed in the above-mentioned irregular curved surface, i.e., the target special-shaped space, by an algorithm. For example, the vertex coordinates of each glass unit are randomly determined so that all the units are located within the curved surface range, and the splicing positions of adjacent units form the curtain wall edges, such as the side edges, top edges, or bottom edges of the glass units.

[0046] The finally generated first curtain wall layout information can be an initial layout composed of 200 glass units, and the edges of each unit are randomly distributed in the special-shaped space, part of the edges are close to the edges of the special-shaped space, and part of the edges are located in the middle of the space. Assuming that each unit has four edges, there are about 800 curtain wall edges in total.

[0047] In step S200 of the embodiment of the present application, a first special-shaped space approximation degree is calculated and obtained according to the first curtain wall layout information and the target special-shaped space, including:

[0048] According to the special-shaped space edges in the target special-shaped space, the distances between the multiple curtain wall edges in the first curtain wall layout information and the special-shaped space edges are calculated to obtain multiple first curtain wall distances.

[0049] According to the multiple first curtain wall distances, a first special-shaped space approximation degree of the first curtain wall layout information is calculated and obtained.

[0050] In the embodiment of the present application, this step is a basic link for calculating the first special-shaped space approximation degree, and the core purpose is to provide original data support for subsequent evaluation of the fitting degree of the curtain wall layout and the special-shaped space by quantifying the distances between the curtain wall edges and the target special-shaped space edges. Specifically, by obtaining the distances of the multiple curtain wall edges to the special-shaped space edges, the matching accuracy of the initial curtain wall layout in the space form can be intuitively reflected, which is a prerequisite for subsequent calculation of the first special-shaped space approximation degree to measure the layout fitting degree, and ensures that the subsequent optimization process can judge whether the layout meets the modeling requirements of the special-shaped space based on objective distance data.

[0051] In the implementation process, the target special-shaped space obtained in step S100 and the generated first curtain wall layout information need to be used as the basis to determine the two key elements involved in the calculation:

[0052] The "special-shaped space edge" of the target special-shaped space is the boundary contour of the space, such as the curved boundary line of the building facade, the angular line, etc. The "multiple curtain wall edges" in the first curtain wall layout information are the structural boundaries of the curtain wall itself, such as the splicing edge of the unit curtain wall, the edge of the glass / metal plate, etc.

[0053] Then, the distance between the multiple curtain wall edges in the first curtain wall layout information and the special-shaped space edge needs to be calculated.

[0054] Specifically, the distance between each curtain wall edge in the first curtain wall layout information and the target special-shaped space edge needs to be calculated respectively. For example, if the special-shaped space edge is an irregular curve, such as the curved boundary of the science museum facade, and the curtain wall edge is the side of a glass unit, such as a straight line or a curve, the shortest distance from each point on the curtain wall edge to the special-shaped space edge is calculated by a space geometry algorithm, such as the shortest distance formula from a point to a line, the minimum distance calculation method between curves, and the minimum value is taken as the first curtain wall distance between the curtain wall edge and the special-shaped space edge.

[0055] For example, the minimum distance between the curtain wall edge 1 and the special-shaped space edge is 14.14 cm, the minimum distance between the curtain wall edge 2 and the special-shaped space edge is 22.36 cm, the minimum distance between the curtain wall edge 3 and the special-shaped space edge is 14.14 cm, and so on.

[0056] Repeat the above process to obtain multiple first curtain wall distances corresponding to all curtain wall edges.

[0057] The above steps convert the abstract "layout fitting degree" into specific numerical data by quantifying the spatial distance between each curtain wall edge and the special-shaped space edge, and provide a key basis for subsequent calculation of the first special-shaped space approximation degree, i.e. the matching degree of the overall layout and the special-shaped space.

[0058] In step S200 of the embodiments of the present application, the first special-shaped space approximation degree of the first curtain wall layout information is calculated according to the multiple first curtain wall distances, including:

[0059] Calculate the mean value of the multiple first curtain wall distances to obtain the first average curtain wall distance;

[0060] Calculate the maximum fluctuation amplitude of the multiple first curtain wall distances, compensate for the first average curtain wall distance, and obtain the first compensated curtain wall distance;

[0061] Obtain the standard curtain wall distance;

[0062] Calculate the similarity between the first compensated curtain wall distance and the standard curtain wall distance to obtain the first special-shaped space approximation degree.

[0063] In the embodiments of the present application, the core purpose of calculating the first special-shaped space approximation degree is to convert a plurality of dispersed first curtain wall distances into a quantitative index that can comprehensively reflect the fitting degree of the overall layout of the curtain wall to the target special-shaped space. By calculating the mean value of a plurality of first curtain wall distances, compensating the fluctuation amplitude, and comparing with the standard curtain wall distance, the precise evaluation of the layout fitting degree is realized, and a quantifiable reference basis is provided for subsequent optimization.

[0064] In the implementation process, first, the mean value of the plurality of first curtain wall distances needs to be calculated to obtain the first average curtain wall distance. For example, taking the plurality of first curtain wall distances 14.14 cm, 22.36 cm, and 14.14 cm obtained in the foregoing as an example, the arithmetic mean value thereof can be calculated to obtain the first average curtain wall distance, i.e., the first average curtain wall distance=(14.14+22.36+14.14) / 3=16.88 cm. This value reflects the average distance between the edge of the curtain wall and the edge of the special-shaped space, and the smaller the value, the higher the overall fitting degree.

[0065] Then, the maximum fluctuation amplitude of the plurality of first curtain wall distances needs to be calculated to compensate the first average curtain wall distance and obtain the first compensated curtain wall distance.

[0066] Among them, the maximum fluctuation amplitude refers to the difference between the maximum value and the minimum value in the plurality of first curtain wall distances, which is used to measure the dispersion degree of the distance. For example, in the above example, the maximum fluctuation amplitude=22.36-14.14=8.22 cm.

[0067] The maximum fluctuation amplitude of the plurality of first curtain wall distances is used to compensate the first average curtain wall distance by a preset algorithm, and the purpose of compensation is to reduce the influence of extreme values on the overall evaluation, so that the result is more consistent with the actual layout.

[0068] For example, the method of “first curtain wall distance mean value-maximum fluctuation amplitude×weight” is adopted, assuming that the weight is 0.2, then the calculation method of the first compensated curtain wall distance can be first compensated curtain wall distance=16.88-8.22×0.2≈16.88-1.64=15.24 cm.

[0069] Further, the standard curtain wall distance needs to be obtained, which is an ideal curtain wall distance value preset according to the project design requirements, structural safety, construction feasibility, etc., i.e., the optimal average distance between the edge of the curtain wall and the edge of the special-shaped space. For example, for a certain special-shaped glass curtain wall, the standard curtain wall distance is set to 15 cm.

[0070] Finally, the similarity between the first compensated curtain wall distance and the standard curtain wall distance needs to be calculated to obtain the first irregular space approximation degree. The similarity between the first compensated curtain wall distance and the standard curtain wall distance represents the degree of closeness between them. Specifically, the calculation method is: First irregular space approximation degree = 1 - (Absolute difference between the first compensated curtain wall distance and the standard curtain wall distance) / Standard curtain wall distance. For example, in the above example, the first irregular space approximation degree = (1 - |15.24 - 15|) / 15 = 1 - 0.016 = 0.984. The closer this first irregular space approximation degree is to 1, the higher the similarity between the first compensated curtain wall distance and the standard curtain wall distance, i.e., the higher the approximation degree.

[0071] In step S300 of this application embodiment, a cleaning feasibility analysis is performed based on the first curtain wall layout information and the first irregular space approximation degree to obtain a first cleaning feasibility parameter and a second cleaning feasibility parameter, including:

[0072] Based on the multiple curtain wall edges within the first curtain wall layout information, obtain multiple perpendicular directions of the multiple curtain wall edges away from the target irregular space;

[0073] Extend a preset distance along multiple vertical lines to obtain multiple curtain wall cleaning locations along multiple curtain wall edges;

[0074] Calculate the distance between every two adjacent curtain wall cleaning locations to obtain multiple cleaning interval distances, and calculate the average cleaning interval distance by taking the mean.

[0075] Input the average cleaning interval distance into the first cleaning classification table to obtain the first cleaning feasibility parameter;

[0076] Based on the first irregular space approximation degree, a cleaning feasibility analysis is performed to obtain the second cleaning feasibility parameter.

[0077] The core purpose of this step is to provide foundational data for obtaining initial cleaning feasibility parameters. Specifically, it involves analyzing the spatial orientation of the curtain wall edge to determine the operating positions of cleaning tools such as drones and the distances between adjacent positions, thereby assessing the ease of cleaning operations. By calculating the average cleaning interval distance, the uniformity of the cleaning position distribution is quantified, providing a crucial basis for determining whether cleaning tools such as drones can operate efficiently.

[0078] During implementation, it is first necessary to obtain multiple vertical directions of the curtain wall edges away from the target irregular space. Since the curtain wall edge is the boundary of the curtain wall structure, such as the splicing edge of the glass unit, the vertical direction away from the target irregular space refers to the direction that is perpendicular to the curtain wall edge and points to the outside of the building, that is, the side where the cleaning tool is working.

[0079] For example, if a curtain wall edge is a horizontal straight line, assuming it extends along the X-axis direction, with coordinates y = 3, x ∈ [2, 5], the perpendicular direction of the edge is the direction perpendicular to the X-axis, i.e. the Y-axis direction; since the target special-shaped space is the internal range of the building facade, the "away" direction is the positive direction of the Y-axis, which points to the outside of the building, and the cleaning tool can work on this side.

[0080] Then, for each curtain wall edge in the first curtain wall layout information, the perpendicular direction away from the target special-shaped space is determined by a geometric algorithm, and a plurality of direction data are obtained, such as "along the positive direction of the Y-axis", "outward at an angle of 60° with the X-axis", etc.

[0081] Then, the distance between each two adjacent curtain wall cleaning positions needs to be calculated to obtain a plurality of cleaning interval distances, and the average cleaning interval distance is calculated. The preset distance is a fixed value set according to the working radius of the cleaning tool, such as the optimal cleaning distance of 1.5 meters for a drone, i.e. the distance between the drone and the curtain wall surface needs to be kept at 1.5 meters to work efficiently.

[0082] Taking the horizontal straight line curtain wall edge (y = 3, x ∈ [2, 5]) in the above example as an example, the perpendicular direction of the edge is the positive direction of the Y-axis, and after extending 1.5 meters along this direction, the points on the edge, such as (2, 3), (3.5, 3), and (5, 3), correspond to the cleaning position coordinates (2, 4.5), (3.5, 4.5), and (5, 4.5), respectively. These points constitute the curtain wall cleaning positions of the edge, and in order to simplify the calculation process, the extension point of the midpoint of the edge can be taken as the representative, i.e. (3.5, 4.5), as the curtain wall cleaning position of the edge.

[0083] Repeat the above operation for all curtain wall edges, and each edge corresponds to a cleaning position, and finally a plurality of curtain wall cleaning positions are obtained, such as 10 cleaning positions corresponding to 10 curtain wall edges.

[0084] Then, the distance between each two adjacent curtain wall cleaning positions needs to be calculated to obtain a plurality of cleaning interval distances, and the average cleaning interval distance is calculated. The cleaning interval distance refers to the straight line distance between two adjacent curtain wall cleaning positions. Assuming that 5 adjacent cleaning positions are obtained through the above steps, with coordinates A(3.5, 4.5), B(7, 5.0), C(10, 4.8), D(13, 5.2), and E(16, 4.6), the distance between each two adjacent positions is calculated.

[0085] The distance between A and B is: √[(7-3.5) 2 +(5.0-4.5) 2 ]≈√(12.25+0.25)=√12.5≈3.54 meters. The distance between B and C is: √[(10-7) 2+ (4.8-5.0) 2 ] = V(9 + 0.04) = V9.04 = 3.01 meters. The distance between C and D is: V[(13-10) 2 + (5.2-4.8) 2 ] = V(9 + 0.16) = V9.16 = 3.03 meters. The distance between D and E is: V[(16-13) 2 + (4.6-5.2) 2 ] = V(9 + 0.36) = V9.36 = 3.06 meters.

[0086] Through the above steps, a plurality of cleaning interval distances are obtained, i.e. 3.54 meters, 3.01 meters, 3.03 meters, and 3.06 meters. The average cleaning interval distance is the mean value of the above cleaning interval distances, specifically (3.54 + 3.01 + 3.03 + 3.06) ÷ 4 = 12.64 ÷ 4 = 3.16 meters.

[0087] This step converts the operation space requirement of the cleaning tool into quantitative data, i.e. the average cleaning interval distance, by determining the cleaning position, calculating the cleaning interval distance and the mean value, which provides a key basis for subsequent query of the first cleaning classification table and is a basic link for evaluating the cleaning feasibility.

[0088] Further, the average cleaning interval distance needs to be input into the first cleaning classification table to obtain the first cleaning feasibility parameter.

[0089] In step S300 of the embodiment of the present application, the construction step of the first cleaning classification table comprises:

[0090] According to historical data of using a UAV to clean a curtain wall, a sample cleaning interval distance set is collected, and control accuracy under different sample cleaning interval distances is collected to obtain a sample first cleaning feasibility parameter set;

[0091] A mapping relationship between the sample cleaning interval distance set and the sample first cleaning feasibility parameter set is constructed to obtain the first cleaning classification table.

[0092] In the embodiment of the present application, the core purpose of constructing the first cleaning classification table is to construct a standard reference tool for quickly judging the cleaning feasibility. The corresponding relationship between the cleaning interval distance and the UAV cleaning control accuracy is established through historical data, so as to convert the quantitative index of the average cleaning interval distance obtained through the foregoing steps into the first cleaning feasibility parameter which can be directly used to evaluate the cleaning feasibility. The essence is to use historical experience data to form a standardized judgment basis, so as to ensure that the subsequent analysis of the curtain wall layout cleaning feasibility is objective and accurate, and to avoid the deviation of subjective judgment.

[0093] In the implementation process, optionally, first, the cleaning interval distance data under different curtain wall layouts needs to be extracted from the historical operation records of using unmanned aerial vehicles to clean the curtain wall. For example, the average cleaning interval distance of each operation in the past 100 unmanned aerial vehicle cleaning operations is collected to form a data set, such as [2.5 meters, 3.0 meters, 3.2 meters, …, 4.0 meters], as the sample cleaning interval distance set.

[0094] Then, for each sample cleaning interval distance, the control accuracy of the unmanned aerial vehicle at the cleaning interval distance is synchronously collected, that is, whether the unmanned aerial vehicle can stably maintain the distance from the curtain wall and accurately move to the cleaning position, which can be measured by indicators such as operation error rate and collision risk, and the control accuracy is labeled as a quantitative cleaning feasibility parameter, such as a value between 0 and 1, 1 representing the highest control accuracy and the strongest cleaning feasibility, and 0 representing the inability to complete cleaning. For example, the specific first cleaning feasibility parameter evaluation method can be evaluated by the unmanned aerial vehicle operator after actually operating the unmanned aerial vehicle to clean, and the specific evaluation process is not described here.

[0095] For example, when the cleaning interval distance is 3.0 meters, the cleaning effect is good, and the first cleaning feasibility parameter is labeled as 0.9; when the interval distance is 4.5 meters, the cleaning effect is poor, and the first cleaning feasibility parameter is labeled as 0.3.

[0096] Further, the corresponding relationship between the sample cleaning interval distance and the sample first cleaning feasibility parameter needs to be established through data fitting or statistical analysis. For example, if the historical data shows that when the cleaning interval distance is 2.8-3.2 meters, the control accuracy of the unmanned aerial vehicle is above 0.8, the cleaning interval distance in this interval can be mapped to the first cleaning feasibility parameter 0.8-1.0; when the cleaning interval distance is 3.3-3.7 meters, the control accuracy is reduced to 0.5-0.7, and the corresponding first cleaning feasibility parameter is 0.5-0.7; when the cleaning interval distance exceeds 4.0 meters, the control accuracy is less than 0.3, and the corresponding first cleaning feasibility parameter is 0-0.3.

[0097] In step S300 of the embodiment of the present application, the first irregular space approximation degree is used to analyze the cleaning feasibility to obtain a second cleaning feasibility parameter, including:

[0098] A second cleaning classification table is obtained, wherein the second cleaning classification table is constructed based on the mapping relationship between the sample irregular space approximation degree set and the sample second cleaning feasibility parameter set in the historical curtain wall cleaning record data, and the sample second cleaning feasibility parameter includes a cleaning rate;

[0099] The first irregular space approximation degree is input into the second cleaning classification table, and a second cleaning feasibility parameter is obtained by classification.

[0100] In the embodiments of the present application, the core purpose of performing the cleaning feasibility analysis to obtain the second cleaning feasibility parameter is to further evaluate the cleaning feasibility based on the fitting degree of the curtain wall layout and the target special-shaped space, i.e., the first special-shaped space approximation degree. The first special-shaped space approximation degree is converted into a quantitative cleaning efficiency index through the second cleaning classification table, so as to obtain the second cleaning feasibility parameter. The function is to supplement the evaluation dimension from the perspective of the influence of the layout form on the cleaning effect, and together with the first cleaning feasibility parameter based on the control precision of the interval distance, to form a comprehensive judgment of the cleaning feasibility, so as to ensure that the subsequent layout optimization can take into account the convenience of cleaning operation and the final cleaning effect.

[0101] Firstly, the second cleaning classification table needs to be obtained. The second cleaning classification table is a mapping tool constructed based on historical data, and the core is to establish the corresponding relationship between the sample special-shaped space approximation degree and the sample second cleaning feasibility parameter.

[0102] Optionally, the "sample special-shaped space approximation degree" of different curtain wall layouts, i.e., the fitting degree of the historical layout and the special-shaped space at that time, which is usually in the range of 0-1, and the corresponding sample second cleaning feasibility parameter, such as the cleaning rate, i.e., the ratio of the actual cleaning area to the total area, which is in the range of 0-100%, can be extracted from the past curtain wall cleaning records.

[0103] Then, the correlation between the approximation degree and the cleaning rate is obtained through statistical analysis of the historical data. For example:

[0104] When the sample special-shaped space approximation degree is greater than or equal to 0.9, the layout is highly fitted with the space, the form is regular, the cleaning tool can cover most of the area, and the sample cleaning rate is mostly 90%-100%;

[0105] When the sample special-shaped space approximation degree is between 0.7 and 0.9, the fitting degree is good, there is a small local deviation, and the cleaning rate is mostly 70%-90%;

[0106] When the sample special-shaped space approximation degree is between 0.5 and 0.7, the fitting degree is general, there is a significant form deviation, and the cleaning rate is mostly 50%-70%;

[0107] When the sample special-shaped space approximation degree is less than 0.5, the fitting degree is poor, the form is chaotic, and the cleaning rate is mostly less than 50%.

[0108] Then, the first special-shaped space approximation degree calculated in step S200, such as 0.984 in the foregoing example, is input into the second cleaning classification table to match the corresponding cleaning rate range. For example, the first special-shaped space approximation degree is 0.984, which falls in the interval of 0.9-1.0, and according to the classification table, the corresponding second cleaning feasibility parameter, i.e., the cleaning rate, is 90%-100%. If a more accurate value is needed, the interval mean, such as 95%, can be further taken. Alternatively, the cleaning rate classification in the second cleaning classification table can be refined to obtain a more accurate second cleaning feasibility parameter value, and the basic logic is the same as the foregoing steps, which will not be described herein again.

[0109] In step S400 of the embodiment of the present application, a first layout score is calculated according to the first special-shaped space approximation degree, the first cleaning feasibility parameter, and the second cleaning feasibility parameter, the curtain wall layout information is optimized, and optimal curtain wall layout information is obtained, including:

[0110] The first layout score is calculated and obtained according to the first special-shaped space approximation degree, the first cleaning feasibility parameter, and the second cleaning feasibility parameter.

[0111] The curtain wall layout information is continuously randomly generated, the layout score is calculated, the iteration optimization is performed, and the optimal curtain wall layout information that converges is obtained.

[0112] In the embodiment of the present application, the core purpose of step S400 is to find the optimal curtain wall layout scheme that takes into account the space adaptability and the convenience of later maintenance by comprehensively evaluating the multi-dimensional indexes of the curtain wall layout.

[0113] In the implementation process, first, the first layout score is calculated according to the first special-shaped space approximation degree, the first cleaning feasibility parameter, and the second cleaning feasibility parameter. Specifically, the first layout score can be calculated by weighted summation or the like. The first layout score range can be 0-100 points, and the higher the score, the better the layout.

[0114] First, the importance of each parameter is determined according to the actual demand to set the parameter weight.

[0115] For example, in a certain implementation scenario, the weight of the first special-shaped space approximation degree is set to 40%; the weight of the first cleaning feasibility parameter is set to 30%; and the weight of the second cleaning feasibility parameter is set to 30%.

[0116] With the values in the above example, assume that the first special-shaped space approximation degree is 0.984; the first cleaning feasibility parameter, based on a cleaning interval distance of 3.16 meters, is matched from the first cleaning classification table as 0.8; the second cleaning feasibility parameter is based on the first special-shaped space approximation degree of 0.984, and the cleaning rate matched from the second cleaning classification table, i.e., the second cleaning feasibility parameter, is 95%, converted to 0.95.

[0117] Therefore, for example, the first layout score at this time can be first layout score = 0.984 x 40 + 0.8 x 30 + 0.95 x 30 = 39.36 + 24 + 28.5 = 91.86 points.

[0118] Further, it is necessary to continue to randomly generate curtain wall layout information, calculate the layout score, iterate and optimize, and obtain the optimal curtain wall layout information of optimization convergence.

[0119] Optionally, the second, third, …, and N curtain wall layout information can be randomly generated in the target special-shaped space, and the specific generation manner is the same as that in step S100. Each curtain wall layout information includes a plurality of curtain wall edges.

[0120] Then, for each new curtain wall layout, the first special-shaped space approximation degree, the first cleaning feasibility parameter, and the second cleaning feasibility parameter are calculated in turn, and the layout score is calculated in the same manner. For example, the second layout score is 85 points, and the third layout score is 93 points.

[0121] When the layout scores generated continuously for multiple times fluctuate less than a preset threshold, such as less than ±0.5 points, and reach the highest score, it is considered that the optimization converges, and the layout with the highest layout score is taken as the optimal curtain wall layout information. For example, the curtain wall layout information with a layout score of 95 points is taken as the optimal curtain wall layout.

[0122] This step quantifies the overall advantages and disadvantages of the layout by comprehensive scoring, and constantly approaches the optimal solution by iterative optimization. Finally, the obtained scheme can not only highly adapt to the special-shaped space modeling, but also meet the cleaning requirements, realizes the dual goals of reasonable curtain wall form and efficient maintenance, and improves the scientificity and practicality of the special-shaped curtain wall layout design.

[0123] As shown in FIG. 2, based on the same inventive concept of the special-shaped curtain wall structure layout optimization method provided in embodiment one, the present embodiment also provides a special-shaped curtain wall structure layout optimization system, which comprises: Figure 2 The curtain wall layout information acquisition module 11 is configured to acquire a target special-shaped space to be subjected to curtain wall structure layout, and randomly generate first curtain wall layout information, wherein the first curtain wall layout information comprises curtain wall structure information.

[0124]

[0125] ​The special-shaped space approximation degree calculation module 12 is configured to calculate a first special-shaped space approximation degree according to the first curtain wall layout information and the target special-shaped space.

[0126] The curtain wall cleaning difficulty assessment module 13 is configured to perform cleaning feasibility analysis according to the first curtain wall layout information and the first special-shaped space approximation degree, and obtain first and second cleaning feasibility parameters.

[0127] The curtain wall layout information optimization module 14 is configured to calculate a first layout score according to the first special-shaped space approximation degree, the first cleaning feasibility parameter and the second cleaning feasibility parameter, and optimize the curtain wall layout information to obtain optimal curtain wall layout information.

[0128] Further, the curtain wall layout information acquisition module 11 includes the following execution steps:

[0129] The target special-shaped space to be arranged with the curtain wall structure is acquired, wherein the target special-shaped space includes a special-shaped space edge; and the first curtain wall layout information is randomly generated in the target special-shaped space, wherein the first curtain wall layout information includes curtain wall structure information, and the curtain wall structure information includes a plurality of curtain wall edges.

[0130] Further, the special-shaped space approximation degree calculation module 12 includes the following execution steps:

[0131] The first special-shaped space approximation degree of the first curtain wall layout information is calculated according to the plurality of first curtain wall distances, including: calculating distances between the plurality of curtain wall edges in the first curtain wall layout information and the special-shaped space edge in the target special-shaped space according to the special-shaped space edge in the target special-shaped space, to obtain a plurality of first curtain wall distances; and calculating the first special-shaped space approximation degree of the first curtain wall layout information according to the plurality of first curtain wall distances.

[0132] The first special-shaped space approximation degree of the first curtain wall layout information is calculated according to the plurality of first curtain wall distances, including:

[0133] The mean value of the plurality of first curtain wall distances is calculated to obtain a first average curtain wall distance;

[0134] The maximum fluctuation amplitude of the plurality of first curtain wall distances is calculated, and the first average curtain wall distance is compensated to obtain a first compensated curtain wall distance;

[0135] A standard curtain wall distance is acquired;

[0136] The similarity between the first compensated curtain wall distance and the standard curtain wall distance is calculated to obtain the first special-shaped space approximation degree.

[0137] Further, the curtain wall cleaning difficulty assessment module 13 includes the following execution steps:

[0138] According to a plurality of curtain wall edges in the first curtain wall layout information, a plurality of perpendicular line directions of the plurality of curtain wall edges away from the target special-shaped space are obtained;

[0139] According to the plurality of perpendicular line directions, a plurality of curtain wall cleaning positions of the plurality of curtain wall edges are obtained by extending a preset distance;

[0140] The distance between each two adjacent curtain wall cleaning positions is calculated to obtain a plurality of cleaning interval distances, and the average cleaning interval distance is obtained by calculating the average;

[0141] The average cleaning interval distance is input into the first cleaning classification table to obtain a first cleaning feasibility parameter;

[0142] According to the first special-shaped space approximation degree, a cleaning feasibility analysis is performed to obtain a second cleaning feasibility parameter.

[0143] The construction of the first cleaning classification table includes:

[0144] According to historical data of curtain wall cleaning by using a drone, a sample cleaning interval distance set is collected, and control accuracy under different sample cleaning interval distances is collected to obtain a sample first cleaning feasibility parameter set by labeling;

[0145] A mapping relationship between the sample cleaning interval distance set and the sample first cleaning feasibility parameter set is constructed to obtain the first cleaning classification table.

[0146] According to the first special-shaped space approximation degree, a cleaning feasibility analysis is performed to obtain a second cleaning feasibility parameter, including:

[0147] A second cleaning classification table is obtained, wherein the second cleaning classification table is constructed based on a mapping relationship between a sample special-shaped space approximation degree set and a sample second cleaning feasibility parameter set in historical curtain wall cleaning record data, and the sample second cleaning feasibility parameter includes a cleaning rate;

[0148] The first special-shaped space approximation degree is input into the second cleaning classification table to obtain the second cleaning feasibility parameter by classification.

[0149] Further, the curtain wall layout information optimization module 14 includes the following execution steps:

[0150] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0151] Those skilled in the art will appreciate that embodiments of the present application can be readily used as software, hardware, or a combination of software and hardware. In a software embodiment, the methods can be tangibly embodied in a machine-readable storage medium having stored thereon instructions that can be used to program a computer to perform any of the methods. The software implementation can be initialized by loading and executing a set of instructions arranged to perform one of the methods into the computer's memory. Alternatively, hard-wired circuitry can be used in place of, or in combination with, software instructions. Thus, the

[0152] The present application is described in reference to the drawings using a flowchart and / or a block diagram of the method, apparatus (system) and computer program product according to embodiments of the application. It will be understood that each block of the flowchart and / or block diagram, and combinations of blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in one or more of the flowchart or block diagram block or blocks. Figure 1 one or more functions specified in one or more of the flowchart or block diagram block or blocks.

[0153] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in one or more of the flowchart or block diagram block or blocks. Figure 1 one or more functions specified in one or more of the flowchart or block diagram block or blocks.

[0154] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in one or more of the flowchart or block diagram block or blocks. Figure 1 one or more functions specified in one or more of the flowchart or block diagram block or blocks.

[0155] While preferred embodiments of the application have been described, modifications and variations can be apparent to those skilled in the art once aware of the general underlying concepts.

[0156] Obviously, numerous modifications and variations of the present application are possible in light of the above teachings. It is therefore to be understood that within the scope of the application, modifications and variations of this application can be made by those skilled in the art upon celebrating the above teaching and once the general underlying concepts are appreciated.

Claims

1. A method for optimizing a layout of a curtain wall structure of a special shape, characterized in that, The method comprises: obtaining a target irregular space to be arranged with a curtain wall structure, and randomly generating first curtain wall layout information, wherein the first curtain wall layout information comprises curtain wall structure information; obtaining a first irregular space approximation degree according to the first curtain wall layout information and the target irregular space; performing cleaning feasibility analysis according to the first curtain wall layout information and the first irregular space approximation degree to obtain first cleaning feasibility parameters and second cleaning feasibility parameters; calculating a first layout score according to the first irregular space approximation degree, the first cleaning feasibility parameters and the second cleaning feasibility parameters, optimizing the curtain wall layout information, and obtaining optimal curtain wall layout information; performing cleaning feasibility analysis according to the first curtain wall layout information and the first irregular space approximation degree to obtain first cleaning feasibility parameters and second cleaning feasibility parameters, comprising: obtaining a plurality of perpendicular line directions of a plurality of curtain wall edges in the first curtain wall layout information away from the target irregular space; extending a plurality of curtain wall cleaning positions of the plurality of curtain wall edges by a preset distance according to the plurality of perpendicular line directions; calculating the distance between every two adjacent curtain wall cleaning positions to obtain a plurality of cleaning interval distances, and calculating the average cleaning interval distance; inputting the average cleaning interval distance into a first cleaning classification table to obtain the first cleaning feasibility parameters; performing cleaning feasibility analysis according to the first irregular space approximation degree to obtain the second cleaning feasibility parameters; the construction steps of the first cleaning classification table comprise: collecting a sample cleaning interval distance set and collecting control accuracy under different sample cleaning interval distances according to historical data of curtain wall cleaning by an unmanned aerial vehicle, and labeling to obtain a sample first cleaning feasibility parameter set; constructing a mapping relationship between the sample cleaning interval distance set and the sample first cleaning feasibility parameter set to obtain the first cleaning classification table; performing cleaning feasibility analysis according to the first irregular space approximation degree to obtain the second cleaning feasibility parameters, comprising: obtaining a second cleaning classification table, wherein the second cleaning classification table is constructed based on a mapping relationship between a sample irregular space approximation degree set and a sample second cleaning feasibility parameter set in historical curtain wall cleaning record data, and the sample second cleaning feasibility parameter comprises a cleaning rate; inputting the first irregular space approximation degree into the second cleaning classification table to classify and obtain the second cleaning feasibility parameter.

2. The method of claim 1, wherein, obtaining a target irregular space to be arranged with a curtain wall structure, and randomly generating first curtain wall layout information, comprising: obtaining a target irregular space to be arranged with a curtain wall structure, wherein the target irregular space comprises an irregular space edge; randomly generating first curtain wall layout information in the target irregular space, wherein the first curtain wall layout information comprises curtain wall structure information, and the curtain wall structure information comprises a plurality of curtain wall edges.

3. The method of claim 1, wherein, obtaining a first irregular space approximation degree according to the first curtain wall layout information and the target irregular space, comprising: calculating the distance between the plurality of curtain wall edges in the first curtain wall layout information and the irregular space edge according to the irregular space edge in the target irregular space to obtain a plurality of first curtain wall distances; According to a plurality of first curtain wall distances, a first special-shaped space approximation degree of first curtain wall layout information is calculated.

4. The method of claim 3, wherein, According to a plurality of first curtain wall distances, a first special-shaped space approximation degree of first curtain wall layout information is calculated, comprising: The mean value of the plurality of first curtain wall distances is calculated to obtain a first average curtain wall distance; The maximum fluctuation amplitude of the plurality of first curtain wall distances is calculated, and the first average curtain wall distance is compensated to obtain a first compensated curtain wall distance; A standard curtain wall distance is obtained; The similarity of the first compensated curtain wall distance and the standard curtain wall distance is calculated to obtain a first special-shaped space approximation degree.

5. The method of claim 1, wherein, According to the first special-shaped space approximation degree, a first cleaning feasibility parameter and a second cleaning feasibility parameter, a first layout score is calculated, the curtain wall layout information is optimized, and optimal curtain wall layout information is obtained, comprising: According to the first special-shaped space approximation degree, a first cleaning feasibility parameter and a second cleaning feasibility parameter, a first layout score is calculated; Continue to randomly generate curtain wall layout information, calculate the layout score, and iteratively optimize to obtain optimal curtain wall layout information that converges.

6. A layout optimization system for irregularly shaped curtain wall structures, characterized in that, The system comprises: A curtain wall layout information acquisition module is configured to acquire a target special-shaped space for curtain wall structure layout, and randomly generate first curtain wall layout information, wherein the first curtain wall layout information comprises curtain wall structure information; A special-shaped space approximation degree calculation module is configured to calculate a first special-shaped space approximation degree according to the first curtain wall layout information and the target special-shaped space; A curtain wall cleaning difficulty evaluation module is configured to perform cleaning feasibility analysis according to the first curtain wall layout information and the first special-shaped space approximation degree, and obtain a first cleaning feasibility parameter and a second cleaning feasibility parameter; A curtain wall layout information optimization module is configured to calculate a first layout score according to the first special-shaped space approximation degree, a first cleaning feasibility parameter and a second cleaning feasibility parameter, optimize the curtain wall layout information, and obtain optimal curtain wall layout information; According to the first curtain wall layout information and the first special-shaped space approximation degree, a first cleaning feasibility parameter and a second cleaning feasibility parameter are obtained by performing cleaning feasibility analysis, comprising: According to a plurality of curtain wall edges in the first curtain wall layout information, a plurality of vertical line directions of the plurality of curtain wall edges away from the target special-shaped space are obtained; A plurality of curtain wall cleaning positions of the plurality of curtain wall edges are obtained by extending a plurality of vertical line directions by a preset distance; The distance between every two adjacent curtain wall cleaning positions is calculated to obtain a plurality of cleaning interval distances, and the mean value is calculated to obtain an average cleaning interval distance; The average cleaning interval distance is input into a first cleaning classification table to obtain a first cleaning feasibility parameter; According to the first special-shaped space approximation degree, a second cleaning feasibility parameter is obtained by performing cleaning feasibility analysis; The construction steps of the first cleaning classification table comprise: According to historical data of curtain wall cleaning by a drone, a sample cleaning interval distance set is collected, and control accuracy under different sample cleaning interval distances is collected to obtain a sample first cleaning feasibility parameter set; A mapping relationship between the sample cleaning interval distance set and the sample first cleaning feasibility parameter set is constructed to obtain a first cleaning classification table; According to the first special-shaped space approximation degree, a cleaning feasibility analysis is performed to obtain a second cleaning feasibility parameter, including: A second cleaning classification table is obtained, wherein the second cleaning classification table is constructed based on a mapping relationship between a set of sample special-shaped space approximation degrees and a set of sample second cleaning feasibility parameters in historical curtain wall cleaning record data, and the sample second cleaning feasibility parameter includes a cleaning rate; The first special-shaped space approximation degree is input into the second cleaning classification table, and a second cleaning feasibility parameter is classified and obtained.

Images