AI-Based Building Design Optimization System and Method

Abstract

The present invention relates to the field of artificial intelligence technology, specifically an AI-based building design optimization system and method. The system includes: a visual parameter acquisition module, an interface adjustment determination module, a behavior path construction module, a space layout replacement module, and a sequence adaptation and update module. In the present invention, by extracting the line-of-sight access length and angle offset value of the functional area boundary, a visual difference sequence is constructed to clarify the spatial visual connection difference, identify the line-of-sight obstacle and uneven perception area, combine the functional attributes and wayfinding path to construct an interface adjustment list, improve the pertinence and efficiency of structural adjustment, establish a weight structure diagram through the behavior call frequency and the path segment change ratio to reflect the spatial behavior association, screen the layout optimization plan in combination with the connection relationship and the path expectation difference, enhance the logical consistency and behavior matching degree, identify the behavior sequence deviation and update the operation path diagram, realize the dynamic adaptation of the functional area sequence and the reconstruction of the behavior logic, and improve the spatial responsiveness and usage rationality.

Description

Technical Field

[0001] The present invention relates to the technical field of artificial intelligence, and in particular to an AI-based building design optimization system and method. Background Art

[0002] The technical field of artificial intelligence involves the research and application of theories, methods, and technologies related to computer science and artificial intelligence, mainly covering aspects such as machine learning, deep learning, natural language processing, and computer vision. The core content of artificial intelligence is to enable computer systems to automatically execute tasks and provide decision support by simulating and extending human intelligence. Its application scope is extensive, including speech recognition, image processing, robotics, intelligent recommendation, autonomous driving, etc. This technical field is committed to improving the automation degree, accuracy, and efficiency of systems in specific tasks through the innovation of artificial intelligence algorithms and models.

[0003] Among them, a building design optimization system refers to the automatic generation and optimization of building design schemes based on artificial intelligence technology. The technical matters of this patent theme mainly cover data analysis and processing in the building design process, intelligent reasoning of design parameters, and optimization decisions, etc. Specifically, machine learning algorithms are used to analyze design data, artificial intelligence models such as deep neural networks are used to predict and optimize design schemes, and the models are iteratively adjusted according to feedback information to improve design quality and efficiency. This system promotes the automation and intelligence of building design through the combination of data input and intelligent reasoning, and solves the complexity and efficiency problems in the traditional manual design process.

[0004] The prior art relies on static modeling in visual element analysis, lacks systematic identification of differences in line-of-sight accessibility, resulting in errors in spatial wayfinding perception. During the interface adjustment process, the functional attributes and path order cannot be effectively associated, and interface modification often falls into structural isolation, lacking overall logical support. The construction of behavioral paths lacks comprehensive weight analysis of call frequency and path continuity, and the path settings do not match the actual usage logic well. During the spatial layout adjustment process, the connection relationship is not combined to maintain control of the difference from the path expectation, and the replacement scheme is prone to deviate from the usage habits. The adjustment of the functional area sorting under changing scenarios lacks a weight deviation judgment mechanism, resulting in insufficient behavioral sequence response ability and poor spatial adaptability. Summary of the Invention

[0005] The purpose of the present invention is to solve the deficiencies existing in the prior art, and to propose an AI-based building design optimization system and method.

[0006] To achieve the above purpose, the present invention adopts the following technical solutions: The AI-based building design optimization system includes:

[0007] The visual parameter acquisition module obtains all the boundary surfaces of functional areas in the indoor layout of a building, acquires the line-of-sight access lengths and angular offset values in both forward and reverse directions, calculates the average values respectively and performs difference normalization and superposition to obtain a sequence of visual difference values between functional areas;

[0008] The interface adjustment determination module calls the items exceeding the threshold in the sequence of visual difference values between functional areas, judges the enclosure, light transmittance and the number of operable components of the interface, marks the material replacement items and structural opening items, associates the sorting position of the interface in the wayfinding path with the usage attributes of the functional area, and establishes an operation sequence for all the suggested items to obtain an interface reset operation list;

[0009] The behavior path construction module extracts the call frequencies of functional areas based on the interface reset operation list, calculates the frequency difference and change ratio of path segments, and marks the combinations exceeding the path coherence threshold to obtain a path node weight structure diagram;

[0010] The space arrangement replacement module calls the functional area combinations in the path node weight structure diagram, calculates the spatial grid distance and the difference from the path expectation, filters the replacement schemes with unchanged connection relationships to obtain a feasible arrangement replacement sequence.

[0011] As a further solution of the present invention, the sequence of visual difference values between functional areas specifically includes line-of-sight accessibility difference, angular offset difference, and difference normalization index. The interface reset operation list includes material replacement items, structural opening items, and operation sorting labels. The path node weight structure diagram specifically refers to path node call frequency, path coherence weight, and node combination mark. The feasible arrangement replacement sequence includes spatial grid distance evaluation value, path expectation deviation value, and connection relationship retention label.

[0012] As a further solution of the present invention, the visual parameter acquisition module includes:

[0013] The boundary extraction sub-module acquires all the boundary surfaces of functional areas in the indoor layout of a building, converts the outer contour lines of functional areas in the two-dimensional plane image into boundary vertex coordinate data, uses the contour boundary closure condition to check the integrity of the acquired boundary, and obtains all the functional area boundary line segment information according to the connection relationship between the spatial contours of different functional areas to generate functional area boundary line segment data;

[0014] The line-of-sight parameter calculation sub-module calls the functional area boundary line segment data, sets multiple equally spaced monitoring points in both forward and reverse directions of the boundary line segment respectively, projects the line of sight to each monitoring point respectively, acquires the line-of-sight access length without penetrating other boundaries, and records the angular offset value relative to the boundary normal, calculates the average length and average angular offset of all forward lines of sight, as well as the average length and average angular offset of all reverse lines of sight, and obtains the average values of two-way line-of-sight parameters;

[0015] The visual difference generation sub-module calls the mean value of the bidirectional line-of-sight parameters, refers to the influence of the difference in line-of-sight parameters between the multi-functional areas on spatial perception, and introduces a line-of-sight difference index. Adopt the formula:

[0016] ;

[0017] Calculate the visual difference coefficient between the multi-functional areas to form a sequence of visual difference values for the functional areas;

[0018] Among them, represents the line-of-sight difference index, represents the average length of the forward line of sight of the th functional area, represents the average length of the reverse line of sight of the th functional area, represents the threshold constant used in the line-of-sight difference calculation of the th functional area, is an adjustment constant used to maintain the non-zero state of the denominator, represents the weight factor of the line-of-sight parameters of the th functional area in the difference calculation, represents the total number of functional areas.

[0019] As a further solution of the present invention, the interface adjustment determination module includes:

[0020] The difference item screening sub-module, based on the sequence of visual difference values between the functional areas, calls a preset visual difference threshold, compares and screens each item in the difference value sequence one by one, obtains the difference items in the difference value sequence that exceed the visual difference threshold, determines the spatial distribution block where the difference items are located in the functional area, calculates the number of difference items and their corresponding interface unit numbers, and generates a set of difference items exceeding the threshold;

[0021] The interface characteristic judgment sub-module calls the set of difference items exceeding the threshold, obtains the enclosure parameter, light transmittance parameter and the number of opening elements of the corresponding interface unit, judges whether the interface is enclosed according to the enclosure parameter, judges whether there is a light-transmitting structure by using the light transmittance parameter, judges the openable state by the number of opening elements, calculates the interface transparency characteristic value and performs weighted integration with the usage frequency coefficient, and adopts the formula:

[0022] ;

[0023] Calculate the interface transparency difference value, call the transparency difference threshold to judge whether it is an item to be adjusted, and obtain the interface item to be adjusted attribute;

[0024] Among them, represents the transparency difference value of the th interface, represents the The number of difference items corresponding to an interface, indicating the th closure parameter value of the th item in the th interface, representing the corresponding light transmittance parameter, being the light intensity coverage coefficient of the th item in the th interface, being the number of activatable elements corresponding to the th difference item in the th interface, being the usage frequency of the th interface, being the activity value of the functional area where the

[0025] th interface is located;

[0026] As a further solution of the present invention, the behavior path construction module includes:

[0027] The interface operation reset sub-module extracts the functional area name, trigger time, and usage frequency corresponding to each operation in the list based on the interface reset operation list. After classifying by functional area, it obtains the call frequency of the multi-functional area in the complete interaction process, and summarizes and sorts the call frequencies to generate the functional area call frequency value;

[0028] The path segment difference calculation sub-module calls the functional area call frequency value to obtain the call frequency difference between consecutive path segments, using the formula:

[0029] ;

[0030] Calculate the path segment stability score, and determine the continuity of the multi-path segment based on this score to obtain the path segment frequency change rate;

[0031] Among them, represents the path segment stability score value, k represents the kth path segment, N represents the total number of path segments to be evaluated, represents the call frequency of the starting functional area of the kth path segment, represents the call frequency of the ending functional area of the kth path segment, Denotes the absolute value of the difference in call frequency between the start and end points of the k-th path segment. Represents the decay coefficient of the operation interval time of the path segment. Represents the operation interval time corresponding to the k-th path segment. Denotes the decay factor based on the operation interval time. Represents the weighting factor of the k-th path segment, used to adjust the weight of the path segment in the overall stability score. Denotes the influence value of the call frequency at the end point of the k-th path segment after weighting. Represents the adjustment parameter.

[0032] The coherence structure marking sub-module calls the frequency change rate of the path segment, judges item by item with the path coherence threshold according to the frequency change ratio in the path segment combination, performs marking processing on the path segment combination whose comparison value is greater than the path coherence threshold, maps the marked path segment to the path node structure diagram, and assigns corresponding weights to multiple nodes to generate a path node weight structure diagram.

[0033] As a further solution of the present invention, the spatial arrangement replacement module includes:

[0034] The path weight extraction sub-module extracts the connection side length, node number, and corresponding functional area combination between multiple path nodes based on the path relationship marked in the path node weight structure diagram, constructs a path paragraph matrix according to the node connection order, and reassigns its weight value with the physical position parameter and combination relationship parameter between nodes to generate a path paragraph weight matrix.

[0035] The expected difference calculation sub-module calls the path paragraph weight matrix, calculates the functional adaptability difference between nodes according to the node function attribute and the actual distance, and uses the formula:

[0036] ;

[0037] Calculates the functional adaptability deviation of multiple nodes and establishes a functional path expected deviation value.

[0038] Among them, Represents the functional adaptability deviation value between path nodes. Represents the Actual physical distance between the connected nodes in the -th segment of the path. Represents the functional difference level value between the functional area combinations corresponding to the nodes in the -th segment of the path. Represents the structural strength weight value of the functional area combination in the nodes connected in the -th segment of the path. Represents the participation level of the nodes in the -th segment of the path in the functional connection network. Represents the upper limit of the count of all paragraphs in the path, represents the total number of nodes participating in the weight evaluation within the path segment, represents the actual physical distance between the connected nodes in the represents the functional difference level value between the functional area combinations corresponding to the nodes in the

[0039] The alternative sequence screening sub-module counts the connection numbers, judges the structural preservation, and sorts the offset values for all replaceable path combinations according to the expected deviation value of the functional path and the screening condition that the node connection relationship in the path segment is not reduced, screens out the path combinations with reduced number of connection edges, sorts and selects the structural alternative sequences with the top-ranked offset values, and generates a feasible layout alternative sequence.

[0040] As a further solution of the present invention, the system further includes:

[0041] The sequence adaptation and update module adjusts the functional area order according to the functional area order adjustment plan in the feasible layout alternative sequence, extracts the weight distribution table of the functional area behavior order corresponding to the current context label in the marked exhibition and teaching scenarios, calculates the deviation value between the actual layout order and the weight sorting, and if the deviation value is greater than the functional area sorting offset threshold, marks the set of functional areas that need to be updated, generates an order update operation queue corresponding to the current context label, and obtains a functional area sorting update path diagram;

[0042] The functional area sorting update path diagram specifically includes sorting offset marking, updating the functional area set, and context adaptation order.

[0043] As a further solution of the present invention, the sequence adaptation and update module includes:

[0044] The context label extraction sub-module adjusts the functional area order according to the functional area order adjustment plan in the feasible layout alternative sequence, obtains the marked exhibition and teaching scenario data, calls the weight distribution table of the functional area behavior order corresponding to the context label, extracts the matching items between the context label and the functional area behavior, determines the weight arrangement order according to the matching relationship, and generates a context label weight order table;

[0045] The behavior order deviation calculation sub-module calls the current actual layout order according to the context label weight order table, calculates the difference arrangement position between the actual order and the weight order, and uses the formula:

[0046] ;

[0047] Calculates the behavior order offset index, judges whether it is greater than the benchmark according to the functional area sorting offset threshold, screens the functional areas that need to be adjusted, and generates a set of functional areas to be updated;

[0048] Among them, represents the behavior sequence offset index, represents the position of the th functional area in the actual layout, represents the position of the th functional area in the weight sorting, is the visibility coefficient of the th functional area, represents the total number of functional areas;

[0049] The sequence update path generation sub-module extracts the arrangement information of the to-be-updated functional area set in the current sequence and its target position in the weight sorting according to the to-be-updated functional area set, constructs a sorting replacement path according to the differential order adjustment method, forms an adjustment action queue and sorts it, combines the actual site feasibility comparison structure required for sequence adjustment, filters the feasible operation path, and obtains the functional area sorting update path diagram.

[0050] An AI-based building design optimization method, the AI-based building design optimization method is executed based on the above-mentioned AI-based building design optimization system, and includes the following steps:

[0051] S1: Obtain all the boundary surfaces of the functional areas in the building interior layout, collect the sight line access length and angle offset values, calculate the access difference and angle difference between each combination and superimpose them after normalization, and generate a visual difference value sequence between the functional areas;

[0052] S2: Compare the visual difference threshold based on the visual difference value sequence between the functional areas, extract the combinations that exceed the threshold and call their enclosure, light transmittance and the number of openable components, judge the interface material replacement and structural opening requirements, establish the interface and path association in combination with the guide path sorting and functional attributes, arrange the suggestion items in the judgment order, and generate an interface reset operation list;

[0053] S3: Call the functional area numbers in the interface reset operation list, obtain the call frequency of each functional area and calculate the path segment frequency difference and ratio, and generate a path node weight structure diagram;

[0054] S4: Call the functional area combinations in the path node weight structure diagram, calculate their spatial grid distance and path expectation difference, filter the combinations with unchanged connection relationships, and obtain a feasible layout replacement sequence;

[0055] S5: Extract the behavior sequence weight distribution table under the context label according to the feasible layout replacement sequence, calculate the current sorting deviation value and filter the functional areas with excessive deviation, and generate a functional area sorting update path diagram.

[0056] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0057] In the present invention, by extracting the line-of-sight access length and angle offset value of the functional area boundary, constructing a visual difference value sequence, and clarifying the actual differences in the visual connectivity of the space, it helps to identify the line-of-sight obstacle positions and the areas with uneven perception. Based on the visual difference results, combined with the functional attributes and the wayfinding path sequence, an interface adjustment operation list is constructed to improve the pertinence and execution efficiency of the interface structure adjustment. In path construction, a weight structure diagram is established through the behavior call frequency and the path segment change ratio to accurately reflect the spatial correlation relationship of the usage behavior. The space layout replacement combines the connection relationship and the path expected difference to screen and optimize the solutions, enhancing the logical consistency and behavior matching degree of the layout adjustment. The identification of the behavior sequence deviation and the generation of the updated operation path diagram realize the dynamic adaptation of the functional area sequence and the reconstruction of the behavior logic in the context, improving the spatial responsiveness and the rationality of use. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 is the system flow chart of the present invention;

[0059] Figure 2 is the flow chart of the visual parameter acquisition module of the present invention;

[0060] Figure 3 is the flow chart of the interface adjustment determination module of the present invention;

[0061] Figure 4 is the flow chart of the behavior path construction module of the present invention;

[0062] Figure 5 is the flow chart of the space layout replacement module of the present invention;

[0063] Figure 6 is the flow chart of the sequence adaptation and update module of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0064] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only for explaining the present invention and are not used to limit the present invention.

[0065] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present invention. In addition, in the description of the present invention, "a plurality of" means two or more unless otherwise specifically defined.

[0066] Example 1: Please refer to Figure 1 , the present invention provides a technical solution: The AI-based building design optimization system includes:

[0067] The visual parameter acquisition module acquires all the boundary surfaces of the functional areas in the building interior layout, collects the line-of-sight access lengths and angle offset values in both forward and reverse directions, calculates the average values respectively and performs difference normalization superposition to obtain the visual difference value sequence between functional areas;

[0068] The interface adjustment determination module calls the items exceeding the threshold in the visual difference value sequence between functional areas, judges the interface closure, light transmittance and the number of operable components, marks the material replacement items and structural opening items, associates the sorting position of the interface in the wayfinding path with the usage attributes of the functional areas, and establishes an operation sequence for all the recommended items to obtain the interface reset operation list;

[0069] The behavior path construction module extracts the functional area call frequencies based on the interface reset operation list, calculates the path segment frequency difference and change ratio, and marks the combinations exceeding the path coherence threshold to obtain the path node weight structure diagram;

[0070] The space arrangement replacement module calls the functional area combinations in the path node weight structure diagram, calculates the space grid distance and the path expectation difference, and screens the replacement schemes with unchanged connection relationships to obtain the feasible arrangement alternative sequence;

[0071] The sequence adaptation and update module adjusts the functional area order according to the functional area order adjustment scheme in the feasible arrangement alternative sequence, extracts the weight distribution table of the functional area behavior order corresponding to the current situation label in the marked exhibition and teaching scenarios, calculates the deviation value between the actual arrangement order and the weight sorting, and if the deviation value is greater than the functional area sorting offset threshold, marks the set of functional areas that need to be updated, generates an order update operation queue corresponding to the current situation label, and obtains the functional area sorting update path diagram.

[0072] The visual difference value sequence between functional areas specifically refers to the line-of-sight accessibility difference, angle offset difference, and difference normalization index. The interface reset operation list includes material replacement items, structural opening items, and operation sorting labels. The path node weight structure diagram specifically refers to the path node call frequency, path coherence weight, and node combination mark. The feasible arrangement alternative sequence includes space grid distance evaluation value, path expectation deviation value, and connection relationship retention label. The functional area sorting update path diagram specifically includes sorting offset mark, updated functional area set, and situation adaptation order.

[0073] Please refer to Figure 2 , the visual parameter acquisition module includes:

[0074] The boundary extraction sub-module collects all the boundary surfaces of the functional areas in the indoor layout of the building, converts the outer contour lines of the functional areas in the two-dimensional plane image into boundary vertex coordinate data, uses the closed condition of the contour boundary to check the integrity of the collected boundary, and obtains all the boundary line segment information of the functional areas according to the connection relationship between the spatial contours of different functional areas, generating the boundary line segment data of the functional areas;

[0075] The boundary extraction sub-module collects the boundary surfaces of all the functional areas in the indoor layout of the building. This boundary collection process is first carried out in the two-dimensional plane image of the actual indoor layout (for example, the CAD two-dimensional layout of an office building). For the outer contour lines of the two-dimensional images of each functional area, extraction is carried out. For example, for areas such as the office area, meeting room, corridor, and rest area, extraction is carried out one by one. Among them, for each area contour line, point-by-point scanning is carried out, and the coordinate positions of each corner point on the functional area contour line are marked and numbered in sequence. For example, the corner points of the office area are A(3.5m, 2.0m), B(7.0m, 2.0m), C(7.0m, 5.0m), and D(3.5m, 5.0m). Similarly, for the contour points of the meeting room E(7.2m, 2.0m), F(10.0m, 2.0m), G(10.0m, 5.0m), and H(7.2m, 5.0m), similar processing is carried out, thus obtaining a specific two-dimensional vertex coordinate data array. Secondly, according to the closed condition of the contour boundary, the integrity of all the functional area contour boundaries is checked one by one. By judging the closure of the coordinates at both ends of each boundary line segment of each functional area, that is, whether the start and end coordinates are the same. For example, for the office area, starting from the starting point A, passing through B, C, D in sequence and returning to the starting point A, a coordinate closure comparison is carried out once to ensure the integrity of the contour. Then, based on the connection relationship between the spatial contours of different functional areas, that is, judging whether the boundaries of adjacent functional areas are collinear. By comparing the boundary coordinates of adjacent areas point by point, the common line segments are determined (for example, the boundary points B and E, C and H between the office area and the meeting room are collinear). After the connection relationship is confirmed, the complete boundary line segment data sets of each functional area are stored respectively, completing the standardization processing of the boundary information, and finally generating the boundary line segment data set of the functional areas, in the form of {office area:[(A, B), (B, C), (C, D), (D, A)], meeting room:[(E, F), (F, G), (G, H), (H, E)]}.

[0076] The line-of-sight parameter calculation sub-module calls the boundary line segment data of the functional areas, sets multiple equally spaced monitoring points in the positive and negative directions of the boundary line segments respectively, projects the line of sight for each monitoring point respectively, collects the line-of-sight reach length under the condition of not penetrating other boundaries, and records its angular offset value relative to the boundary normal. Calculate the average length and average angular offset of all the positive line-of-sight, as well as the average length and average angular offset of all the negative line-of-sight, obtaining the average value of the two-way line-of-sight parameters;

[0077] The line-of-sight parameter calculation sub-module calls the data of the boundary line segments in the functional area. Taking the boundary line segment data of the actual office area as an example, for the line segment (A, B), a monitoring point is set every 0.5 m in its forward direction (A→B) and reverse direction (B→A). For example, there are 6 monitoring points in total, namely A1(4.0 m, 2.0 m), A2(4.5 m, 2.0 m) to A6(6.5 m, 2.0 m). Each monitoring point projects a line of sight into the functional area until it encounters other boundary line segments, and the line-of-sight length of each monitoring point is calculated. For example, the line-of-sight length of A1 is 3.0 m, that of A2 is 3.0 m, and similarly, the line-of-sight length of A6 is measured to be 3.0 m. Then, the average length of the forward line of sight is calculated to be 3.0 m through the length data. At the same time, the angle offset value between the line of sight of each monitoring point and the normal direction of the boundary line segment is recorded. For example, the offset angles of each monitoring point are measured to be 5°, 6°, 5°, 4°, 5°, 4° in sequence, so as to calculate the average angle offset of the forward line of sight to be 4.83°. Repeat the above process to measure the monitoring points B1 to B6 in the reverse direction (B→A), and the measured average length is 2.8 m and the average offset angle is 5.16°. Then repeat the above process for all boundary line segments in the functional area, and finally calculate and integrate to obtain the average value of the two-way line-of-sight parameters of this functional area.

[0078] The visual difference generation sub-module calls the average value of the two-way line-of-sight parameters, and refers to the influence of the difference in line-of-sight parameters between multi-functional areas on spatial perception, and introduces a line-of-sight difference index Adopt the formula:

[0079] ;

[0080] Calculate the visual difference coefficient between multi-functional areas to form a sequence of visual difference values of the functional area;

[0081] Among them, represents the line-of-sight difference index, represents the average length of the forward line of sight of the th functional area, represents the average length of the reverse line of sight of the th functional area, represents the threshold constant used in the line-of-sight difference calculation of the th functional area, is an adjustment constant used to maintain the non-zero state of the denominator, represents the weight factor of the line-of-sight parameters of the th functional area in the difference calculation, represents the total number of functional areas.

[0082] The visual difference generation sub-module calls the average value of the two-way line-of-sight parameters between functional areas. Taking 3 functional areas of office area, meeting room and corridor as an example, the parameters are shown in Table 1:

[0083] Table 1 Weight table of line-of-sight parameters of functional areas;

[0084] ;

[0085] Calculate the visual difference index for each functional area respectively :

[0086] Office area:

[0087] ;

[0088] Conference room:

[0089] ;

[0090] Corridor:

[0091] ;

[0092] According to the above results, obtain the sequence of visual difference values for functional areas: ;

[0093] Furthermore, accumulate the visual difference indexes of each functional area according to the formula to obtain the visual difference coefficient between multi-functional areas (i.e., the total value of overall visual difference): ;

[0094] This visual difference coefficient reflects the degree of spatial perception difference generated by the line-of-sight drop between functional areas in the overall space. The higher the value, the weaker the visual coherence of the space and the stronger the sense of difference; conversely, a low difference coefficient indicates that the visual transition of the space is relatively natural.

[0095] Both this sequence of visual difference values and the total value can be used as quantitative evaluation bases for optimizing spatial layout, adjusting channels, or enhancing visual permeability, further improving the functional adaptability and humanistic perception performance of buildings or interior spaces.

[0096] The innovation of the formula lies in that by introducing the line-of-sight parameter difference index , the subjective visual perception is quantified into objective numerical values, clearly indicating the spatial positions or orientations that can be optimized in the spatial layout of different functional areas, and improving the accuracy and applicability of the overall layout design.

[0097] Please refer to Figure 3 , the interface adjustment determination module includes:

[0098] The difference item screening sub-module, based on the sequence of visual difference values between functional areas, calls the preset visual difference threshold, compares and screens each item in the difference value sequence one by one, obtains the difference items in the difference value sequence that exceed the visual difference threshold, determines the spatial distribution block where the difference items are located in the functional area, calculates the number of difference items and their corresponding interface unit numbers, and generates a set of difference items exceeding the threshold;

[0099] The difference item screening sub-module operates based on the sequence of visual difference values in the functional area. Specifically, it calls the calculated sequence of visual difference metrics (such as 0.4 for the office area, 0.429 for the meeting room, and 0.5 for the corridor), and sets a visual difference threshold. The threshold is set with reference to the statistical distribution of this visual difference metric in actual projects. For example, through the statistical analysis of the visual difference values of the indoor layout of office buildings, the normal range of the visual difference metric is obtained as [0.1, 0.6]. If it exceeds 0.35, it is determined that the visual difference is obvious. Therefore, the threshold is set to 0.35. Next, a comparison operation of the difference value sequence is carried out item by item, that is, the difference metrics of the three functional areas of the office area (0.4), the meeting room (0.429), and the corridor (0.5) are compared with the set threshold of 0.35 one by one. It is determined that all three data exceed the threshold (for example: 0.4 > 0.35 in the office area). The specific spatial distribution positions of the three difference values are determined as the office area interface unit (interface numbers one and two), the meeting room interface unit (interface number three), and the corridor interface unit (interface number four), and the difference items with the determined positions are counted. The total number of difference items is counted as 3 items, and finally, a set of difference items exceeding the threshold is formed with the interface unit number as the identifier. For example: set of difference items exceeding the threshold = [interface one (office area), interface two (meeting room), interface four (corridor)].

[0100] The interface characteristic judgment sub-module calls the set of difference items exceeding the threshold, obtains the enclosure parameter, light transmittance parameter, and the number of operable elements of the corresponding interface unit, and judges whether the interface is enclosed according to the enclosure parameter, judges whether there is a light-transmitting structure using the light transmittance parameter, judges the operable state through the number of operable elements, calculates the interface transparency characteristic value and performs weighted integration with the usage frequency coefficient, using the formula:

[0101] ; Calculate the interface transparency difference value, call the transparency difference threshold to judge whether it is an item to be adjusted, and obtain the interface attribute item to be adjusted;

[0102] Among them, represents the transparency difference value of the th interface, represents the number of difference items corresponding to the th interface, represents the enclosure parameter value of the th item in the th interface, represents the corresponding light transmittance parameter, is the light intensity coverage coefficient of the th item, is the number of operable elements corresponding to the th item in the th interface, is the usage frequency of the th interface, is the activity value of the functional area where the th interface is located, is the sorting number of the interface in the wayfinding path, is the reference wayfinding sorting number of the functional area;

[0103] Taking Interface 1 as an example, the total area measured on site is 12 ㎡, and the area of the airtight glass curtain wall is 10 ㎡. Then the sealing parameter is: ;

[0104] The light transmittance parameter is single-layer glass, light transmittance: ;

[0105] Light intensity coverage coefficient (measured coverage rate 75%): ;

[0106] Number of operable components: ;

[0107] Assume that the interface only contains one difference item ( ), the first part of the calculation is:

[0108] ;

[0109] Interface usage frequency: ;

[0110] The second part of the calculation is: ;

[0111] Then the transparency difference value of Interface 1 is: ;

[0112] The total area of Interface 3 is measured to be 10 ㎡, and the area of the airtight material is 7 ㎡. Then the sealing parameter is: ;

[0113] Double-layer insulating glass is adopted, light transmittance: ;

[0114] The measured light intensity coverage area is 70%, so: ;

[0115] The operable components are 1 window and 1 door leaf: ;

[0116] Assume there is only one difference item ( ), the first part of the calculation is:

[0117] ;

[0118] The usage frequency and functional area parameters are: ;

[0119] The second part is calculated as: ;

[0120] The three - transparency difference value of Interface Three is: ;

[0121] The total area of Interface Four is 15㎡, and the air - impermeable part is 9㎡: ;

[0122] The light - transmissivity is frosted glass, and the light - transmittance: ;

[0123] The measured light intensity coverage is 60%: ;

[0124] The operable component is 1 window: 1;

[0125] Assume that only one difference item is included ( ), the first part is calculated as: ;

[0126] The usage frequency and the functional area parameters are: ;

[0127] The second part is calculated as: ;

[0128] The three - transparency difference value of Interface Four is: ;

[0129] Finally, the three - transparency difference values of Interface One, Three, and Four are respectively:

[0130] ;

[0131] ;

[0132] ;

[0133] Compare the difference values of each interface with the transparency difference threshold to determine whether it is an item to be adjusted, and further obtain the interface adjustment attribute items.

[0134] The reset operation list generation sub - module calls the interface adjustment attribute items, obtains the way - finding path sorting numbers and the functional area usage attributes of the corresponding interfaces, constructs a list of the sequence of operations based on the sorting numbers and usage attributes, arranges the content of the material items to be replaced and the structural opening items included in sequence, establishes the interface replacement operation sequence, and obtains the interface reset operation list.

[0135] The property items to be adjusted in the reset operation list generation sub-module call interface (Interface 1, Interface 3, and Interface 4) are obtained by getting the guiding path sorting numbers of each interface (the sorting number of Interface 1 is 2, the sorting number of Interface 3 is 4, and the sorting number of Interface 4 is 5) and the usage properties of the functional areas (the usage property of the office area is daily office, the usage property of the meeting room is intermittent use, and the usage property of the corridor is passage connection). Based on the ascending order of the sorting numbers and the importance of the usage properties (office area > meeting room > corridor), the reset operation sequence is determined: Interface 1 → Interface 3 → Interface 4. Further analyze the material items or structural opening items to be replaced for each interface. Taking Interface 1 as an example, since the calculated transparency difference value is too high, the interface material needs to be replaced with high-transparency double-layer glass or the structural opening needs to be increased (such as an opening window); for Interface 3, due to its high degree of closure, 2 opening components need to be added; for Interface 4, due to its low activity, 1 opening component needs to be added appropriately. Specify the specific replacement content to form a detailed operation step sequence, that is, the reset operation list is: [Interface 1: Replace the double-layer transparent glass, replace 1 opening window; Interface 3: Increase 2 window openings; Interface 4: Increase 1 opening door], and generate the interface reset operation list.

[0136] Please refer to Figure 4 , the behavior path construction module includes:

[0137] Based on the interface reset operation list, the interface operation reset sub-module extracts the functional area name, trigger time, and usage frequency corresponding to each operation in the list. After classifying by functional area, it obtains the call frequency of the multi-functional area in the complete interaction process, and summarizes and sorts the call frequencies to generate the functional area call frequency value;

[0138] Based on the interface reset operation list, the interface operation reset sub-module first extracts three pieces of data for each operation item from the list, namely the name of the functional area involved in the corresponding operation, the specific trigger time, and the usage frequency of the functional area. For example, the operation record of Interface 1 includes that the name of the functional area is the office area, the trigger time is 08:00 - 18:00, and the usage frequency is 50 times per day. The corresponding functional area of Interface 3 is the meeting room, the trigger times are 09:00 - 12:00 and 14:00 - 17:00, and the usage frequency is 30 times per day. The corresponding functional area of Interface 4 is the corridor, the trigger time is open 24 hours a day, and the usage frequency is 100 times per day. Then, the extracted data for each functional area are classified respectively, and the usage frequencies of different interfaces under the same functional area are added and merged. For example, the usage frequency of the office area is only 50 times per day for Interface 1, the frequency of the meeting room in Interface 3 is 30 times per day, and the frequency of the corridor in Interface 4 is 100 times per day. Then, cross-functional area frequency comparison and sorting operations are performed, that is, the merged usage frequency values are directly sorted from high to low. After sorting, the corridor (100 times per day) is ranked first, the office area (50 times per day) is ranked second, and the meeting room (30 times per day) is ranked third. In this way, a sequence of functional area call frequency values is generated. Finally, the functional area call frequency values are recorded as: [corridor 100, office area 50, meeting room 30].

[0139] The path segment difference calculation sub-module calls the functional area call frequency value to obtain the call frequency difference between consecutive path segments, using the formula:

[0140] ;

[0141] Calculate the path segment stability score, and determine the continuity of multiple path segments based on this score to obtain the path segment frequency change rate;

[0142] Among them, represents the path segment stability score value, k represents the k-th path segment, N represents the total number of path segments to be evaluated, represents the call frequency of the starting functional area of the k-th path segment, represents the call frequency of the ending functional area of the k-th path segment, represents the absolute value of the call frequency difference between the starting point and the ending point of the k-th path segment, represents the attenuation coefficient of the path segment operation interval time, represents the operation interval time corresponding to the k-th path segment, represents the attenuation factor based on the operation interval time, represents the weighting factor of the k-th path segment, which is used to adjust the weight of the path segment in the overall stability score, represents the influence value after weighting the call frequency of the ending point of the k-th path segment, represents the adjustment parameter;

[0143] Taking the three - segment continuous path of "corridor → office area → meeting room" as an example, calculate the stability score of the path segment.

[0144] Table 2 Statistical table of path segment parameters;

[0145] ;

[0146] The parameter settings are as follows:

[0147] (attenuation coefficient);

[0148] (coherence adjustment parameter).

[0149] Substitute into the formula for calculation:

[0150] The numerator part: ;

[0151] The denominator part: ;

[0152] The stability score result: ;

[0153] Calculate the stability score of the path segment , indicating that there is a certain degree of fluctuation in the call frequency between this three - segment path combination. If compared with the preset path coherence threshold (for example, 0.5):

[0154] When , it indicates that there are obvious frequency differences between the path segments, belonging to a path segment combination with large fluctuations;

[0155] When , it indicates that the path segment switching is stable and has good continuity.

[0156] In this example , so it is judged that the coherence between the path segments is weak. It is recommended to improve the path connection experience through methods such as path guidance optimization and buffer space setting in actual applications.

[0157] This path segment difference calculation method quantifies and expresses the path segment continuity difference by introducing the attenuation factor of the operation interval time and the weight modeling of the node frequency difference, and can be used as a decision - making basis for space layout optimization, guiding path generation, and usage heat analysis, enhancing the adaptability and logical coherence of the multi - path segment layout design.

[0158] The coherence structure marking sub-module calls the frequency change rate of the path segment. According to the frequency change ratio in the path segment combination, it judges item by item with the path coherence threshold. For the path segment combination whose comparison value is greater than the path coherence threshold, it performs marking processing, maps the marked path segment to the path node structure diagram, and assigns corresponding weights to multiple nodes to generate a path node weight structure diagram.

[0159] The coherence structure marking sub-module calls the frequency change rate of the path segment 0.574. It performs specific numerical comparison operations on the frequency change rate of the path segment, that is, compares the frequency change rate of the path segment item by item with the path coherence threshold (the threshold 0.5 is set according to the historical record of the usage frequency of the spatial structure and the survey results of the subjective comfort of users). If the previous result 0.574 > 0.5, it is determined that the frequency change rate of the path segment combination (corridor → office area → meeting room) exceeds the threshold, and this path segment combination is clearly marked, and then the mapping operation of the path node structure diagram is performed. That is, according to the spatial positions of the three nodes of the corridor, office area, and meeting room, the corresponding path segment in the node structure diagram is clearly marked as a path segment with obvious fluctuations, and at the same time, corresponding node weights are assigned to the nodes. For example, according to the function area call frequency value sequence, the corridor node weight is assigned 1.0 (highest frequency), the office area node weight is 0.8 (medium frequency), and the meeting room node weight is 0.6 (lowest frequency). Finally, a path node weight structure diagram with path segment coherence markings and node weights is generated.

[0160] Please refer to Figure 5 , the space layout replacement module includes:

[0161] Based on the path relationships marked in the path node weight structure diagram, the path weight extraction sub-module extracts the connection side lengths, node numbers, and corresponding functional area combinations between multiple path nodes, constructs a path paragraph matrix according to the node connection order, and reassigns its weight values with the physical position parameters and combination relationship parameters between nodes to generate a path paragraph weight matrix;

[0162] Based on the path relationships annotated in the path node weight structure diagram, the path weight extraction sub-module first extracts, one by one, the combined information of the physical connection side lengths between path nodes, the node numbers themselves, and the corresponding functional areas from the structure diagram. Specifically, it calculates the actual connection distance between nodes after two-dimensional processing of the node physical coordinates. For example, the physical distance from node number one (corridor) to node number two (office area) is confirmed to be 12 meters through on-site measurement or CAD drawings, and the physical distance from node number two to node number three (meeting room) is confirmed to be 15 meters. At the same time, it extracts the combined relationship of the corresponding functional areas of the nodes. For example, the combined definition of the functional area corresponding to node one is the traffic connection area, node two is the daily office functional area, and node three is the temporary office functional area. Next, it constructs a path paragraph matrix section by section according to the actual connection order of the nodes, and assigns and resets the weight values of the matrix based on the physical position parameters (i.e., the measured physical distances) between the nodes and the parameters of the functional combination relationship. For example, a relatively high weight value of 0.9 - 1.0 is assigned for a physical distance within 10 meters, a medium weight value of 0.7 - 0.89 for a distance of 10 - 20 meters, and a relatively low weight value of 0.5 - 0.69 for a distance exceeding 20 meters. The functional combination relationship is assigned according to the similarity of functional properties. A high functional similarity (such as between the office area and the meeting room) is assigned a weight of 0.9, a medium similarity (such as between the corridor and the office area) is assigned a weight of 0.7, and a low similarity is assigned a weight of 0.5. The path weight values of 0.7 and 0.9 are respectively assigned to path 01→02 and path 02→03. Finally, the path paragraph weight matrix obtained is as follows: ;

[0163] The expected difference calculation sub-module calls the path paragraph weight matrix, calculates the functional adaptability difference between nodes according to the node functional attributes and the actual distance, and uses the formula:

[0164] ;

[0165] Calculate the functional adaptability deviation of multiple nodes and establish the expected deviation value of the functional path;

[0166] Among them, represents the functional adaptability deviation value between path nodes, represents the actual physical distance between the th section of connected nodes in the path, represents the functional difference level value between the functional area combinations corresponding to the th section of nodes in the path, represents the structural strength weight value of the functional area combination in the th section of connected nodes in the path, represents the participation level of the th section of nodes in the functional connection network, represents the counting upper limit of all sections in the path, Represents the total number of nodes participating in the weight evaluation within the path segment, Represents the actual physical distance between the connected nodes in the th segment of the path, represents the functional difference level value between the functional area combinations corresponding to the nodes in the

[0167] The expected difference calculation sub-module calls the above path segment weight matrix, calculates the functional adaptability difference between nodes based on the node functional attributes and the actual physical distance. Taking the actual path segment as an example, where the actual physical distance is obtained by on-site measurement, the distance of path 01→02 is 12 meters, and the distance of 02→03 is 15 meters. The functional difference level value is determined by expert scoring. When the difference is significant, the value ranges from 0.8 to 1.0; when the difference is medium, it ranges from 0.5 to 0.79; when the difference is low, it ranges from 0.1 to 0.49. For example, the functional difference between the corridor and the office area is medium, taking 0.6, and the difference between the office area and the meeting room is low, taking 0.4. The weight value of the functional combination structure strength is obtained through on-site evaluation of the structural connection stability. The reinforced concrete frame structure is assigned a value of 0.9, the lightweight partition wall structure is 0.7, and the glass curtain wall structure is 0.6. For example, 01→02 is a reinforced concrete structure with a weight of 0.9, and 02→03 is a lightweight partition wall structure with a weight of 0.7. The node participation level is determined by the number of times the actual node participates in the path segment per day on average. When the participation is high (more than 80 times), the value is 0.9; when the participation is medium (50 - 79 times), it takes 0.7; when the participation is low (less than 50 times), it takes 0.5. For example, the participation of 01→02 is 100 times, taking 0.9, and the participation of 02→03 is 60 times, taking 0.7. The parameters are shown in Table 3 as follows:

[0168] Table 3 Node Functional Adaptability Parameter Table;

[0169] ;

[0170] According to the above data, substitute into the formula to calculate the functional adaptability deviation value :

[0171] ;

[0172] Substitute the specific data:

[0173] ;

[0174] The result shows that the functional adaptability deviation value of the path node is 4.0199, and the value is greater than the preset deviation benchmark (usually set between 1 - 3), indicating that there is an obvious deviation in the functional adaptability of this path.

[0175] The advantage of the formula is that by introducing the combination calculation of the functional difference level, structural strength, and node participation, it effectively quantifies the functional adaptability deviation between nodes.

[0176] The alternative sequence screening sub-module counts the connection numbers, judges the structural retention, and sorts the deviation values for all replaceable path combinations according to the expected deviation value of the functional path and the screening condition that the node connection relationship in the path segment is not reduced. It screens out the path combinations with reduced connection edges, sorts and selects the top-ranked structural alternative sequences of the deviation values, and generates the feasible layout alternative sequences.

[0177] The alternative sequence screening sub-module performs screening operations based on the calculated expected deviation value of the functional path 4.0199. First, it counts the node connection numbers for each of all replaceable path combinations one by one. For example, if the original path has 2 connection segments, the replacement combination must maintain a connection number not less than that of the original path (2 segments). If the number of connection segments in the replacement combination is less than 2 segments, it is directly excluded. Subsequently, it makes a clear judgment on the structural retention of the path, that is, it judges whether the replacement path combination maintains the node connection relationship without reduction. For example, there must be an effective connection between node one and node two. If the replacement path does not maintain this connection, the replacement path is directly excluded. Then, it calculates the deviation values of the path combinations that meet the above two conditions respectively, and sorts them in descending order according to the calculated functional adaptability deviation values. For example, for the alternative path A (node one → node four → node two → node three), the functional deviation value is 4.5, and for path B (node one → node five → node two → node three), the deviation value is 3.8. After sorting, the deviation value of path A ranks first and that of path B ranks second. According to the sorting results, it generates the feasible layout alternative sequences, that is, generates the structural alternative sequences as [A: node one → node four → node two → node three, B: node one → node five → node two → node three].

[0178] Please refer to Figure 6 , the sequence adaptation and update module includes:

[0179] The scenario label extraction sub-module adjusts the plan according to the functional area sequence in the feasible layout alternative sequences, obtains the marked exhibition and teaching scenario data, calls the weight distribution table of the functional area behavior sequence corresponding to the scenario label, extracts the matching items of the scenario label and the functional area behavior, and determines the weight arrangement order according to the matching relationship to generate the scenario label weight order table;

[0180] The scenario label extraction sub-module adjusts the functional area order in the alternative sequence according to the feasibility arrangement. First, it extracts the scenario data pre-labeled for different functional areas such as the exhibition area and the teaching area, obtains the specific scenario labels corresponding to the functional areas from on-site or layout data. For example, the exhibition area corresponds to "visual guidance" and "immersive experience", and the teaching area corresponds to "interactive demonstration" and "concentrated explanation" and other labels. Then it calls the functional area behavior order weight distribution table corresponding to the scenario labels, and determines the label matching degree by comparing each obtained scenario label with the labels in the weight distribution table one by one. For example, the weight of the exhibition area label "visual guidance" is 0.9, the weight of "immersive experience" is 0.85, the weight of the teaching area label "interactive demonstration" is 0.8, and the weight of "concentrated explanation" is 0.95. And it sorts and reorganizes the labels in descending order of label weight. For example, for the teaching area (concentrated explanation 0.95, interactive demonstration 0.8), and the exhibition area (visual guidance 0.9, immersive experience 0.85). It rearranges the functional area behavior order corresponding to the sorted scenario labels as teaching area → exhibition area, and finally generates a scenario label weight order table, that is: [teaching area (weight 0.95), exhibition area (weight 0.90)].

[0181] The behavior order deviation calculation sub-module calls the current actual arrangement order according to the scenario label weight order table, calculates the difference arrangement positions between the actual order and the weight order, and uses the formula:

[0182] ;

[0183] Calculate the behavior order deviation index, judge whether it is greater than the benchmark according to the functional area sorting deviation threshold, screen the functional areas that need to be adjusted in position, and generate a set of functional areas to be updated;

[0184] Among them, represents the behavior order deviation index, represents the position of the th functional area in the actual arrangement, represents the position of the th functional area in the weight sorting, is the visibility coefficient of the th functional area, represents the total number of functional areas;

[0185] The behavior order deviation calculation sub-module calls the generated scenario label weight order table above, and compares it item by item with the current actual functional area arrangement order. For example, the current actual arrangement order is: exhibition area (position 1), teaching area (position 2), and the weight order is: teaching area (position 1), exhibition area (position 2). Then it clearly extracts the actual arrangement positions (exhibition area position 1, teaching area position 2) and the weight sorting positions (Exhibition area location 2, teaching area location 1), then call the visibility coefficient of the functional area (The actual visible area of ​​the functional area accounts for the total layout area, for example, the exhibition area accounts for 0.7, and the teaching area accounts for 0.8), and the difference between the actual position and the weighted position of each functional area is calculated in turn, and then divided by the corresponding visibility coefficient, and brought into the formula to calculate the behavior order offset index Calculation:

[0186] ;

[0187] The actual calculation example is as follows:

[0188] ;

[0189] The calculated behavior order offset index of 3.6033 is compared with the preset functional area sorting offset threshold (generally between 2.0 and 3.0 based on user feedback statistics), and the result is 3.6033>3.0. Therefore, it is determined that the actual arrangement order of the exhibition area and the teaching area needs to be adjusted, and the set of functional areas that need to be adjusted is determined to be: [exhibition area, teaching area]. This result shows that there is a significant difference between the current actual layout order and the expected weight sorting.

[0190] The benefit of the formula is that, by comparing the actual position with the weighted position and introducing the visibility coefficient, the degree of deviation between the actual layout order of the functional areas and the expected order of the situation can be clarified.

[0191] The sequential update path generation submodule extracts the arrangement information of the functional areas to be updated in the current order and their target positions in the weight sorting according to the set of functional areas to be updated, constructs a sorting alternative path according to the difference order adjustment method, forms an adjustment action queue and sorts it, and screens feasible operation paths in combination with the actual site feasibility comparison structure required for the sequential adjustment to obtain the functional area sorting update path diagram.

[0192] The sequential update path generation sub-module, based on the set of functional areas to be updated determined above (the exhibition area and the teaching area), first specifically extracts the position information of each functional area in the current actual layout sequence (the exhibition area is currently in position 1 and the teaching area is in position 2), as well as their target positions in the context label weight sequence table (the target position of the exhibition area is 2 and the target position of the teaching area is 1), clearly determines the position difference between the actual position and the target position (the exhibition area needs to be adjusted backward by 1 position and the teaching area needs to be adjusted forward by 1 position), then determines the specific position adjustment actions according to the numerical size of the difference position. For example, the teaching area is adjusted from the 2nd position to the 1st position, and the exhibition area is adjusted from the 1st position to the 2nd position, forming a specific sorting replacement path plan, and arranging all adjustment actions in sequence to form a clear functional area adjustment action queue: [(teaching area: position 2 → position 1), (exhibition area: position 1 → position 2)]. Next, this action queue is compared item by item with the feasibility of the actual site space. For example, it is judged whether there are structural obstacles on the path of the teaching area moving from position 2 to position 1. If not, it is determined as a feasible operation path. The adjustment of the exhibition area is the same. Finally, a clear functional area sorting update path diagram is formed, clearly indicating that the new functional area order after adjustment is the teaching area (position 1) and the exhibition area (position 2).

[0193] Table 4 Functional area behavior sequence adjustment parameter table;

[0194] ;

[0195] As shown in Table 4, the specific parameters and operations required for the functional area order adjustment are clearly given. The data in the table can be directly used to generate and execute the functional area sorting update path diagram during the actual implementation process.

[0196] The AI-based building design optimization method is executed based on the above AI-based building design optimization system and includes the following steps:

[0197] S1: Obtain all the boundary surfaces of the functional areas in the building interior layout, collect the line-of-sight access length and angle offset values, calculate the access difference and angle difference between each combination and normalize and superimpose them to generate a sequence of visual difference values between functional areas;

[0198] S2: Compare the line-of-sight access difference threshold based on the sequence of visual difference values between functional areas, extract the combinations that exceed the threshold and call their enclosure, light transmittance and the number of operable components, judge the need for interface material replacement and structural opening, establish the association between the interface and the path in combination with the wayfinding path sorting and functional attributes, arrange the suggestions in the judgment order, and generate an interface reset operation list;

[0199] S3: Call the function area numbers in the interface reset operation list, obtain the call frequencies of each function area, calculate the frequency difference and ratio of the path segments, and generate a path node weight structure diagram;

[0200] S4: Call the function area combinations in the path node weight structure diagram, calculate their spatial grid distances and path expectation differences, and screen out the combinations with unchanged connection relationships to obtain a feasible layout alternative sequence;

[0201] S5: Extract the behavior sequence weight distribution table under the scenario label according to the feasible layout alternative sequence, calculate the current sorting deviation value, and screen out the function areas with excessive deviation to generate a function area sorting update path diagram.

[0202] The above are only the preferred embodiments of the present invention, and do not limit the present invention in other forms. Any person skilled in the art may use the technical content disclosed above to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention still belong to the protection scope of the technical solution of the present invention.

Claims

1. An AI-based building design optimization system, characterized in that, The system includes: The visual parameter acquisition module obtains the boundary surfaces of all functional areas in the indoor layout of the building, acquires the line-of-sight access lengths and angular offset values in both forward and reverse directions, calculates the mean values respectively and performs difference normalization superposition to obtain a sequence of visual difference values between functional areas; The interface adjustment determination module calls the items exceeding the threshold in the sequence of visual difference values between functional areas, judges the enclosure, light transmittance and the number of operable components of the interface, marks the material replacement items and structural opening items, associates the sorting position of the interface in the wayfinding path with the usage attributes of the functional areas, and establishes an operation sequence for all suggested items to obtain an interface reset operation list; The behavior path construction module extracts the call frequencies of functional areas based on the interface reset operation list, calculates the frequency differences and change ratios of path segments, marks the combinations exceeding the path coherence threshold to obtain a path node weight structure diagram; The space arrangement replacement module calls the functional area combinations in the path node weight structure diagram, calculates the space grid distance and the path expectation difference, filters the replacement schemes with unchanged connection relationships to obtain a feasible arrangement replacement sequence; The sequence adaptation and update module adjusts the functional area order according to the functional area order adjustment scheme in the feasible arrangement replacement sequence, extracts the weight distribution table of the functional area behavior order corresponding to the current context label in the marked exhibition and teaching scenarios, calculates the deviation value between the actual arrangement order and the weight sorting. If the deviation value is greater than the functional area sorting offset threshold, mark the set of functional areas that need to be updated, generate an order update operation queue corresponding to the current context label to obtain a functional area sorting update path diagram; The functional area sorting update path diagram specifically includes sorting offset marking, updating the functional area set, and context adaptation order.

2. The AI-based building design optimization system according to claim 1, wherein The sequence of visual difference values between functional areas specifically includes line-of-sight accessibility difference, angular offset difference, and difference normalization index. The interface reset operation list includes material replacement items, structural opening items, and operation sorting labels. The path node weight structure diagram specifically refers to path node call frequency, path coherence weight, and node combination marking. The feasible arrangement replacement sequence includes space grid distance evaluation value, path expectation deviation value, and connection relationship retention label.

3. The AI-based building design optimization system according to claim 2, wherein The visual parameter acquisition module includes: The boundary extraction sub-module acquires the boundary surfaces of all functional areas in the indoor layout of the building, converts the outer contour lines of the functional areas in the two-dimensional plane image into boundary vertex coordinate data, uses the contour boundary closure condition to check the integrity of the acquired boundary, and obtains all functional area boundary line segment information according to the connection relationship between the spatial contours of different functional areas to generate functional area boundary line segment data; The line-of-sight parameter calculation sub-module calls the functional area boundary line segment data, sets multiple equally spaced monitoring points in both forward and reverse directions of the boundary line segment respectively, projects the line of sight for each monitoring point respectively, acquires the line-of-sight access length without penetrating other boundaries, and records its angular offset value relative to the boundary normal, calculates the average length and average angular offset of all forward line of sights, as well as the average length and average angular offset of all reverse line of sights to obtain the mean values of two-way line-of-sight parameters; The visual difference generation sub-module calls the mean value of the bidirectional line-of-sight parameters, refers to the influence of the difference in line-of-sight parameters between multi-functional areas on spatial perception, and introduces a line-of-sight difference index Adopt the formula: ； Calculate the visual difference coefficient between multiple functional areas to form a sequence of visual difference values of functional areas; Among them, represents the visual difference index, represents the average length of the forward line of sight of the th functional area, represents the average length of the backward line of sight of the th functional area, represents the threshold constant used in the calculation of the line of sight difference of the th functional area, is an adjustment constant used to maintain the denominator non-zero, represents the weight factor of the line of sight parameters of the th functional area in the difference calculation, represents the total number of functional areas.

4. The AI-based building design optimization system according to claim 3, characterized in that The interface adjustment determination module includes: The difference item screening sub-module, based on the visual difference value sequence of the function interval, calls a preset visual difference threshold, compares and screens each item in the difference value sequence one by one, obtains the difference items in the difference value sequence that exceed the visual difference threshold, determines the spatial distribution block where the difference items are located in the function area, calculates the number of difference items and their corresponding interface unit numbers, and generates a set of difference items exceeding the threshold; The interface characteristic judgment sub-module calls the set of difference items exceeding the threshold, obtains the sealing parameter, light transmittance parameter and the number of opening elements of the corresponding interface unit, judges whether the interface is sealed according to the sealing parameter, judges whether there is a light-transmitting structure by using the light transmittance parameter, judges the openable state by the number of opening elements, calculates the interface transparency characteristic value and performs weighted integration with the usage frequency coefficient, using the formula: ； Calculate the interface transparency difference value, call the transparency difference threshold to judge whether it is an item to be adjusted, and obtain the interface attribute items to be adjusted; Among them, represents the transparency difference value of the th interface, represents the number of difference items corresponding to the th interface, represents the enclosure parameter value of the th item in the th interface, represents the corresponding light transmittance parameter, is the light intensity coverage coefficient of the th item, is the number of operable elements corresponding to the th item of the difference items in the th interface, is the usage frequency of the th interface, is the activity value of the functional area where the th interface is located, is the sorting number of this interface in the wayfinding path, is the reference wayfinding sorting number of the functional area; The reset operation list generation sub-module calls the interface attribute items to be adjusted, obtains the wayfinding path sorting number and the function area usage attribute of the corresponding interface, constructs a list of operation precedence relationships based on the sorting number and usage attribute, arranges the content of the material items to be replaced and the structural opening items included in sequence, establishes the interface replacement operation sequence, and obtains the interface reset operation list.

5. The AI-based building design optimization system according to claim 4, wherein, The behavior path construction module includes: The interface operation reset sub-module, based on the interface reset operation list, extracts the function area name, trigger time and usage frequency corresponding to each operation in the list, classifies them by function area, obtains the call frequency of the multi-functional area in the complete interaction process, and summarizes and sorts the call frequencies to generate the function area call frequency value; The path segment difference calculation sub-module calls the function area call frequency value, obtains the call frequency difference between consecutive path segments, using the formula: ； Calculate the path segment stability score, and judge the continuity of the multi-path segment according to this score to obtain the path segment frequency change rate; Among them, represents the stability score value of the path segment, k represents the k-th path segment, and N represents the total number of path segments to be evaluated. represents the call frequency of the start functional area of the k-th path segment. represents the call frequency of the end functional area of the k-th path segment. represents the absolute value of the difference in call frequency between the start and end of the k-th path segment. represents the decay coefficient of the operation interval time of the path segment. represents the operation interval time corresponding to the k-th path segment. represents the decay factor based on the operation interval time. represents the weighting factor of the k-th path segment, which is used to adjust the weight of the path segment in the overall stability score. represents the influence value after weighting the call frequency of the end of the k-th path segment. represents the adjustment parameter. The coherence structure marking sub-module calls the path segment frequency change rate, judges item by item with the path coherence threshold according to the frequency change ratio in the path segment combination, performs marking processing on the path segment combination with the comparison value greater than the path coherence threshold, maps the marked path segments to the path node structure diagram, and assigns corresponding weights to multiple nodes to generate the path node weight structure diagram.

6. The AI-based building design optimization system according to claim 5, characterized in that The space layout replacement module includes: The path weight extraction sub-module, based on the path relationships marked in the path node weight structure diagram, extracts the connection side lengths, node numbers and corresponding function area combinations between multiple path nodes, constructs a path paragraph matrix according to the node connection order, and reassigns its weight value with the physical position parameter and combination relationship parameter between nodes to generate a path paragraph weight matrix; The expected difference calculation sub-module calls the path paragraph weight matrix, calculates the functional adaptability difference between nodes according to the node function attribute and the actual distance, using the formula: ； Calculate the functional adaptability deviation of multiple nodes and establish the expected deviation value of the functional path; Among them, represents the functional adaptability deviation value between path nodes, represents the actual physical distance between the connection nodes in the th segment of the path, represents the functional difference level value between the functional area combinations corresponding to the th segment of nodes in the path, represents the structural strength weight value of the functional area combination in the th segment of the connected nodes in the path, represents the participation level of the th segment of nodes in the functional connection network, represents the counting upper limit of all paragraphs in the path, represents the total number of nodes participating in the weight evaluation within the path segment; The alternative sequence screening sub-module counts the connection numbers, judges the structure preservation, and sorts the offset values for all replaceable path combinations according to the screening conditions that the expected deviation value of the functional path and the node connection relationship in the path segment are not reduced, screens out the path combinations with reduced connection edges, sorts and selects the top-ranked structural alternative sequences of the offset values, and generates a feasible layout alternative sequence.

7. The AI-based building design optimization system according to claim 1, wherein The sequence adaptation and update module includes: The scenario label extraction sub-module adjusts the plan according to the functional area order in the feasible layout alternative sequence, obtains the marked exhibition and teaching scenario data, calls the weight distribution table of the functional area behavior order corresponding to the scenario label, extracts the matching items of the scenario label and the functional area behavior, determines the weight arrangement order according to the matching relationship, and generates a scenario label weight order table; The behavior order deviation calculation sub-module calls the current actual arrangement order according to the scenario label weight order table, calculates the difference arrangement position between the actual order and the weight order, and uses the formula: ； Calculate the behavior order offset index, judge whether it is greater than the benchmark according to the functional area sorting offset threshold, screen the functional areas that need to be adjusted, and generate a set of functional areas to be updated; Among them, represents the behavior sequence offset index, represents the position of the th functional area in the actual layout, represents the position of the th functional area in the weight sorting, is the visibility coefficient of the th functional area, represents the total number of functional areas; The order update path generation sub-module extracts the arrangement information in the current order and its target position in the weight sorting according to the set of functional areas to be updated, constructs a sorting alternative path according to the difference order adjustment method, forms and sorts an adjustment action queue, combines the actual site feasibility comparison structure required for order adjustment, screens the feasible operation paths, and obtains a functional area sorting update path diagram.

8. AI-based building design optimization method, characterized in that, Execute according to the AI-based building design optimization system described in any one of claims 1-7, including the following steps: S1: Obtain all functional area boundary surfaces in the building interior layout, collect the sight line access length and angle offset values, calculate the access difference and angle difference between each combination and superimpose them after normalization, and generate a visual difference value sequence between functional areas; S2: Compare the sight line access difference threshold based on the visual difference value sequence between functional areas, extract the combinations that exceed the threshold and call their enclosure, light transmittance and the number of operable components, judge the interface material replacement and structural opening requirements, establish the interface and path association in combination with the guide path sorting and functional attributes, arrange the suggestions in the judgment order, and generate an interface reset operation list; S3: Call the functional area numbers in the interface reset operation list, obtain the call frequency of each functional area and calculate the frequency difference and ratio of the path segments, and generate a path node weight structure diagram; S4: Call the functional area combinations in the path node weight structure diagram, calculate their spatial grid distance and path expectation difference, and screen the combinations with unchanged connection relationships to obtain a feasible layout alternative sequence; S5: Extract the weight distribution table of the behavior order under the scenario label according to the feasible layout alternative sequence, calculate the current sorting deviation value and screen the functional areas with excessive deviation, and generate a functional area sorting update path diagram.

Images