An intelligent design system and method for architectural decoration

Abstract

The application relates to the technical field of architectural decoration design, in particular to an architectural decoration intelligent design system and method, which comprises the following modules: an intelligent perception analysis module, which is used for collecting space characteristic information and environmental parameters; generating a fusion feature matrix based on the space characteristic information and the environmental parameters; an AI design engine module, which is in communication connection with the intelligent perception analysis module and is used for receiving the fusion feature matrix; generating a preliminary design scheme based on the fusion feature matrix; a multi-dimensional optimization evaluation module, which is in communication connection with the AI design engine module and is used for evaluating the preliminary design scheme in terms of functionality, aesthetics and economy; generating an optimized design scheme based on the evaluation results; and an immersive visual interaction module, which is in communication connection with the multi-dimensional optimization evaluation module and is used for receiving the optimized design scheme; generating a three-dimensional visual model of the optimized design scheme; and providing an interactive design adjustment interface, so that the design cycle can be significantly shortened and the labor cost can be reduced.

Description

Technical Field

[0001] This invention relates to the field of architectural decoration design technology, and in particular to an intelligent architectural decoration design system and method. Background Technology

[0002] As people's living standards continue to improve, their demands for living environments are also increasing. Architectural decoration design, as an important means of improving living environments, is receiving more and more attention. Traditional architectural decoration design mainly relies on the experience and creativity of designers. While this method can meet basic needs, it still has many shortcomings in terms of efficiency, accuracy, and personalization.

[0003] In recent years, with the development of computer technology, various architectural decoration design software programs have emerged. These programs offer functions such as 3D modeling and rendering, improving design efficiency to some extent. However, current design software still has many limitations. First, they lack intelligent perception of space and environment, making it impossible to comprehensively and accurately acquire and analyze the basic data required for design. Second, these programs typically only provide preset templates and limited design elements, making it difficult to meet users' personalized needs. Furthermore, existing software has limited capabilities in design optimization, often requiring designers to manually make multiple adjustments to achieve satisfactory results.

[0004] Furthermore, current design methodologies and tools have shortcomings in areas such as collaborative design and version management. In large projects, multiple designers often encounter problems such as information asynchrony and version conflicts when working simultaneously, severely impacting work efficiency. At the same time, existing design systems lack adaptive learning capabilities and cannot draw experience from past projects, making it difficult to continuously improve design quality.

[0005] Finally, current design methods also face challenges in cost control and material selection. Designers often need to manually calculate and adjust to ensure that the design is within budget, which is not only time-consuming and labor-intensive but also prone to errors. Summary of the Invention

[0006] In view of the above problems, there is an urgent need for an intelligent system that can comprehensively improve the efficiency and quality of architectural decoration design. This invention aims to solve these problems existing in the prior art and provide an intelligent architectural decoration design system and method.

[0007] This invention proposes an intelligent design system for architectural decoration, comprising:

[0008] The intelligent sensing and analysis module is used for:

[0009] Collect spatial feature information and environmental parameters;

[0010] Based on the spatial feature information and environmental parameters, a fusion feature matrix is ​​generated;

[0011] The AI ​​design engine module, which is communicatively connected to the intelligent perception and analysis module, is used for:

[0012] Receive the fused feature matrix;

[0013] Based on the fused feature matrix, a preliminary design scheme is generated;

[0014] The multi-dimensional optimization evaluation module, which communicates with the AI ​​design engine module, is used for:

[0015] The preliminary design scheme was evaluated in terms of functionality, aesthetics, and economy.

[0016] Based on the evaluation results, an optimized design scheme is generated;

[0017] The immersive visualization interaction module, which is communicatively connected to the multi-dimensional optimization evaluation module, is used for:

[0018] Receive the optimized design scheme;

[0019] Generate a 3D visualization model of the optimized design scheme;

[0020] Provides an interactive design adjustment interface.

[0021] Preferably, the intelligent sensing and analysis module includes:

[0022] Spatial feature extraction unit, used for:

[0023] Collect point cloud data;

[0024] Based on the point cloud data, a spatial feature matrix is ​​generated using a topological persistence operator and a random projection matrix;

[0025] The environmental feature fusion unit, which is communicatively connected to the spatial feature extraction unit, is used for:

[0026] Receive the spatial feature matrix and environmental sensor data;

[0027] The spatial feature matrix and environmental sensor data are fused using a quantum convolutional neural network to generate a fused feature matrix.

[0028] Preferably, the AI ​​design engine module includes:

[0029] User requirement mapping unit, used for:

[0030] Receive the fused feature matrix;

[0031] Based on chaos theory and fractal geometry, a nonlinear mapping between user needs and environmental characteristics is established to generate a user needs matrix.

[0032] The design scheme generation unit, which is communicatively connected to the user requirement mapping unit, is used for:

[0033] Receive the user demand matrix;

[0034] Generative adversarial networks are constructed based on group theory and Lie algebras to generate preliminary design schemes.

[0035] Preferably, the multi-dimensional optimization evaluation module employs functional analysis and variational methods for optimization evaluation, including:

[0036] Objective function building unit, used for:

[0037] An optimization objective function is constructed based on functional, aesthetic, and economic indicators;

[0038] The constraint setting unit is used for:

[0039] Set constraints such as design space, material selection, and cost limitations;

[0040] The optimization solution unit is communicatively connected to the objective function construction unit and the constraint setting unit, and is used for:

[0041] Based on the objective function and constraints, the optimal design scheme is solved using the variational method.

[0042] Preferably, the immersive visual interaction module includes:

[0043] 3D rendering unit, used for:

[0044] Receive the optimized design scheme;

[0045] Generate high-quality 3D rendered images using ray tracing technology;

[0046] The VR interaction unit, communicatively connected to the 3D rendering unit, is used for:

[0047] A virtual reality scene is constructed based on the 3D rendered image;

[0048] Offers immersive spatial navigation and real-time design adjustments;

[0049] The AR preview unit, communicatively connected to the 3D rendering unit, is used for:

[0050] The 3D rendered image is overlaid onto the actual environment;

[0051] It provides real-time preview and gesture interaction features on mobile devices.

[0052] As a preferred option, it also includes:

[0053] The knowledge base module, which communicates with the AI ​​design engine module, is used for:

[0054] Store historical design cases and design rules;

[0055] Provide knowledge support for the AI ​​design engine module;

[0056] The adaptive learning module, which is communicatively connected to the knowledge base module, is used for:

[0057] Analyze user feedback and design effectiveness;

[0058] Update the design knowledge and rules in the knowledge base module.

[0059] As a preferred option, it also includes:

[0060] The BIM integration module, which communicates with the AI ​​design engine module and the multi-dimensional optimization evaluation module, is used for:

[0061] The optimized design scheme is then converted into a BIM model.

[0062] Enables data interaction with the Building Information Model.

[0063] Preferably, the AI ​​design engine module also includes:

[0064] Style matching unit, used for:

[0065] Design style identification based on deep learning algorithms;

[0066] Select matching design elements from the style element library;

[0067] The material recommendation unit, which is communicatively connected to the style matching unit, is used for:

[0068] Recommend suitable decorative materials based on the selected design style and environmental characteristics;

[0069] Optimize the selection by considering material performance, cost, and environmental indicators.

[0070] Preferably, the multi-dimensional optimization evaluation module further includes:

[0071] Collaborative design units are used for:

[0072] Supports multi-user remote collaborative design;

[0073] Real-time synchronization of design modifications and evaluation results;

[0074] The version management unit, which is communicatively connected to the collaborative design unit, is used for:

[0075] Record the evolution history of the design scheme;

[0076] Supports comparison and rollback between different versions.

[0077] A smart design method for architectural decoration, based on the aforementioned system, includes the following steps:

[0078] S1 collects spatial feature information and environmental parameters through the intelligent sensing and analysis module;

[0079] S2, Based on the spatial feature information and environmental parameters, a spatial feature matrix is ​​generated using a topological persistence operator and a random projection matrix;

[0080] S3, use a quantum convolutional neural network to fuse the spatial feature matrix and environmental sensor data to generate a fused feature matrix;

[0081] S4, based on chaos theory and fractal geometry, establishes a nonlinear mapping between user needs and environmental characteristics to generate a user needs matrix.

[0082] S5. Using a generative adversarial network based on group theory and Lie algebra, a preliminary design scheme is generated based on the user demand matrix.

[0083] S6. Through the multi-dimensional optimization evaluation module, functional analysis and variational methods are used to evaluate the functionality, aesthetics and economy of the preliminary design scheme.

[0084] S7. Based on the evaluation results, generate an optimized design scheme;

[0085] S8, Use the immersive visualization interaction module to generate a 3D visualization model of the optimized design scheme;

[0086] S9 provides an interactive design adjustment interface, allowing users to make real-time modifications;

[0087] S10 uses an adaptive learning module to analyze user feedback and design effectiveness, and then updates the design knowledge base.

[0088] Specifically, the beneficial effects of the present invention are mainly reflected in the following aspects:

[0089] From a macro perspective, this invention achieves comprehensive intelligence and automation in the architectural decoration design process. The system integrates functions across the entire process, from spatial perception and needs analysis to scheme generation and optimization evaluation, significantly improving design efficiency and quality. Its advantages are particularly evident when handling complex, large-scale design projects, significantly shortening the design cycle and reducing labor costs.

[0090] In terms of system architecture, this invention adopts a modular design, with close collaboration between various functional modules to form an efficient and flexible design ecosystem. The intelligent perception and analysis module provides accurate basic data for subsequent design; the AI ​​design engine module generates personalized design solutions based on this data; the multi-dimensional optimization and evaluation module ensures that the design solutions are optimized in multiple dimensions such as functionality, aesthetics, and economy; and the immersive visual interaction module provides users with an intuitive and convenient design experience. This modular architecture not only improves the maintainability and scalability of the system but also lays the foundation for future functional upgrades and technological iterations.

[0091] From a technological innovation perspective, this invention employs cutting-edge algorithms and methods in several key stages. For example, topological persistence operators and random projection matrices are used in spatial feature extraction, quantum convolutional neural networks are employed in environmental feature fusion, and generative adversarial networks based on group theory and Lie algebras are applied in design scheme generation. These innovative algorithms not only improve the system's performance but also significantly expand its capabilities, enabling it to handle more complex and diverse design tasks.

[0092] This invention also excels in resolving technical contradictions. For example, there is often a conflict between personalized design and high efficiency, but this system successfully improves design efficiency while ensuring personalization through intelligent demand analysis and rapid solution generation. Similarly, this system achieves a balance between ensuring design quality and controlling costs through multi-dimensional optimization evaluation and intelligent material recommendation.

[0093] Another significant advantage of this invention is its adaptive learning capability. By continuously analyzing user feedback and design results, the system can continuously optimize its design strategies and knowledge base, achieving continuous performance improvement. This characteristic of becoming smarter with use makes the long-term value of this system far exceed that of traditional design methods and tools.

[0094] In terms of collaborative design, this invention effectively solves the problems of information synchronization and version control in multi-person collaboration through collaborative design units and version management units, greatly improving the efficiency and quality of team work. This is especially important for the management of large and complex projects.

[0095] Finally, the innovations in visualization and interaction in this invention also bring about a significant improvement in user experience. Through immersive VR / AR technology, users can intuitively experience and adjust design schemes, which not only improves the accuracy of the design but also enhances user participation and satisfaction.

[0096] In summary, the intelligent architectural decoration design system and method of the present invention have achieved a qualitative leap in many aspects such as efficiency, quality, personalization, and collaboration, bringing revolutionary progress to the field of architectural decoration design, and are expected to become a new standard and paradigm in the industry. Attached Figure Description

[0097] Figure 1 This is a logical block diagram of the overall system of the present invention. Detailed Implementation

[0098] See Figure 1 The present invention discloses an intelligent architectural decoration design system comprising an intelligent perception and analysis module 1, an AI design engine module 2, a multi-dimensional optimization and evaluation module 3, and an immersive visualization and interaction module 4. These modules work together to achieve intelligent management of the entire process from spatial perception to the final design scheme.

[0099] The intelligent sensing and analysis module 1 is used to collect spatial feature information and environmental parameters, and generate a fused feature matrix based on this information. Specifically, this module can collect geometric information of the building space through devices such as 3D laser scanners and depth cameras, while using various sensors to collect environmental parameters such as illumination, temperature and humidity, and acoustic characteristics. In a preferred embodiment, spatial feature information may include room dimensions, door and window positions, wall structure, etc., while environmental parameters may include natural light intensity, indoor temperature distribution, noise level, etc.

[0100] The intelligent perception and analysis module 1 employs innovative mathematical methods to process and fuse this information. First, a spatial feature matrix is ​​generated using topological persistence operators and random projection matrices. This method enables the system to effectively extract and represent the geometric and topological features of space, providing crucial foundational information for subsequent design processes.

[0101] Next, the intelligent sensing and analysis module 1 uses a quantum convolutional neural network to fuse the spatial feature matrix and environmental sensor data to generate a fused feature matrix. This quantum computing-based method can more effectively process high-dimensional data and capture the complex relationships between spatial features and environmental parameters.

[0102] AI design engine module 2 communicates with intelligent perception and analysis module 1 to receive the fused feature matrix and generate preliminary design schemes based on it. The core of this module is a design generation system based on the latest artificial intelligence technology. In one embodiment, AI design engine module 2 first uses chaos theory and fractal geometry to establish a nonlinear mapping between user needs and environmental characteristics, generating a user need matrix. This method can capture the complex nonlinear relationship between user needs and environmental characteristics, providing a foundation for generating personalized design schemes.

[0103] Next, AI Design Engine Module 2 uses a generative adversarial network based on group theory and Lie algebras to generate preliminary design schemes. This generative method based on advanced mathematical theory can create more innovative and diverse design schemes.

[0104] The multi-dimensional optimization evaluation module 3 communicates with the AI ​​design engine module 2 to perform functional, aesthetic, and economic evaluations of the preliminary design scheme and generate an optimized design scheme based on the evaluation results. This module employs an innovative multi-objective optimization algorithm that can simultaneously evaluate and optimize across multiple dimensions. This method comprehensively considers multiple design goals and constraints to obtain an optimized design scheme that achieves a balance in functionality, aesthetics, and economy.

[0105] Through this innovative system architecture and algorithm design, the intelligent architectural decoration design system of this invention enables an efficient, personalized, and high-quality design process. The system can not only accurately perceive and analyze spatial characteristics and environmental parameters, but also generate design schemes that meet user needs based on this information and optimize them across multiple dimensions. This approach significantly improves design efficiency while ensuring the quality and personalization of the design schemes.

[0106] In practical applications, this system can be flexibly adjusted according to the specific needs of architectural decoration projects. For example, for different types of architectural spaces (such as residential, office, and commercial spaces), the system can adjust its parameters and optimization objectives to adapt to different design requirements. Simultaneously, the system's adaptive learning capability enables it to continuously learn and improve from historical projects, further enhancing design quality and efficiency.

[0107] In summary, this invention provides an innovative intelligent design system and method for architectural decoration. By combining advanced mathematical theory and artificial intelligence technology, it achieves intelligent management throughout the entire process, from spatial perception to the final design scheme. This not only significantly improves design efficiency but also provides new possibilities for creating more personalized and high-quality architectural decoration designs.

[0108] The intelligent perception and analysis module 1 of the present invention further includes a spatial feature extraction unit 11 and an environmental feature fusion unit 12. These two units work together to realize the transformation process from raw data to a fused feature matrix.

[0109] The spatial feature extraction unit 11 is primarily responsible for collecting point cloud data and generating a spatial feature matrix based on this data. In a preferred embodiment of the invention, the point cloud data can be collected using devices such as a 3D laser scanner or a depth camera. These devices can quickly and accurately capture spatial geometric information, including room dimensions, wall structure, and door and window positions.

[0110] The spatial feature extraction unit 11 employs innovative mathematical methods to process these point cloud data. Specifically, it uses a topological persistence operator and a random projection matrix to generate the spatial feature matrix. This process can be represented by the following formula:

[0111]

[0112] In this formula, S represents the generated spatial feature matrix, T is the topological persistence operator, P is the input point cloud data matrix, and R is the random projection matrix.

[0113] Suppose we have 3D scan data of a room, a point cloud data matrix It is an n×3 matrix, where each row represents the three-dimensional coordinates (x, y, z) of a point. For example:

[0114]

[0115] Topological persistence operator The calculation can be performed using the following steps:

[0116] 1. Calculate the Betty number β k (P):

[0117] For a one-dimensional structure (such as the walls inside a room), β0 represents the number of connected components, and β1 represents the number of cycles. Assuming there are two walls and a doorway in the room, then β0 = 1 (the entire room is a connected component), and β1 = 1 (the doorway forms a cycle).

[0118] 2. Compute the topological persistence operator:

[0119]

[0120] In this example, we assume that only β0 and β1 are considered:

[0121]

[0122] 3. The random projection matrix R can be a 3×2 matrix used to project 3D point cloud data onto a 2D plane:

[0123]

[0124] 4. Calculate the spatial characteristic matrix S:

[0125]

[0126] In this example, because Therefore, S will also be a zero matrix.

[0127] When designing a residence, the system can determine the optimal furniture layout by analyzing a 3D model of the room. For example, based on the room's geometry and the location of doors and windows, the system can suggest the best placement for a sofa, ensuring it doesn't obstruct passageways or windows.

[0128] Topological persistence operator The calculation method is as follows:

[0129]

[0130] Here, β j (P) is the k-th order Betti number, which reflects the topological characteristics of the point cloud data. Using this method, the system of this invention can effectively extract and represent the geometric and topological features of space, providing important basic information for subsequent design processes.

[0131] The environmental feature fusion unit 12 is communicatively connected to the spatial feature extraction unit 11. Its main task is to receive the spatial feature matrix and environmental sensor data, and fuse these two types of information. In one embodiment of the present invention, the environmental sensor data may include parameters such as light intensity, temperature and humidity, and acoustic characteristics. These parameters are crucial for architectural decoration design because they directly affect the comfort and functionality of the space.

[0132] The environmental feature fusion unit 12 employs a quantum convolutional neural network to achieve data fusion. The advantage of this method lies in its ability to handle high-dimensional data and capture the complex relationships between spatial features and environmental parameters. The fusion process can be represented by the following formula:

[0133]

[0134] Here, E is the fused environmental feature matrix, and Q represents the quantum convolution operation. It is an environmental sensor data matrix. The specific definition of the quantum convolution operation is as follows:

[0135]

[0136] In this formula, U is the quantum gate operation. This represents the tensor product. Through this quantum computing-based approach, the system of this invention can more effectively process and fuse complex spatial and environmental data.

[0137] Suppose we have data on the light intensity of a room and an environmental sensor data matrix. It is an m×1 vector representing the light intensity value at different locations. For example:

[0138]

[0139] Assuming spatial characteristic matrix It has been calculated to be an n×2 matrix (two-dimensional data after random projection):

[0140]

[0141] Quantum convolution operations The calculation can be performed using the following steps:

[0142] 1. Tensor Product

[0143]

[0144] 2. Quantum gate operations Assumption It is a simple rotation matrix:

[0145]

[0146] Where θ = π / 4 (45-degree rotation).

[0147] 3. Calculate the fusion feature matrix E:

[0148]

[0149] When designing an office, the system can optimize the layout of the lighting system based on the distribution of natural light. For example, if a certain area has weak natural light, the system can suggest adding artificial light sources to ensure that the entire office area receives sufficient lighting.

[0150] The AI ​​design engine module 2 of the present invention further includes a user requirement mapping unit 21 and a design scheme generation unit 22. These two units work together to realize the transformation process from user requirements to preliminary design schemes.

[0151] The main function of the user demand mapping unit 21 is to receive the fused feature matrix and establish a nonlinear mapping between user demands and environmental features based on this information. In a preferred embodiment of the invention, this mapping process employs methods from chaos theory and fractal geometry. The advantage of this method is that it can capture the complex nonlinear relationship between user demands and environmental features, thereby generating a more accurate user demand matrix.

[0152] The process of mapping user needs can be represented by the following formula:

[0153]

[0154] Here, N represents the generated user demand matrix. λ is the fractal mapping function, E is the input fusion feature matrix, and λ is the control parameter. The specific definition of the fractal mapping function is as follows:

[0155]

[0156] Through this method, the system of the present invention can more accurately understand and express the user's design requirements, providing important input for the generation of subsequent design solutions.

[0157] Suppose there's a home theater design project where the user wants to install a high-end audio system in the room. The environmental characteristic matrix E contains information such as the room's acoustic characteristics and dimensions. For example:

[0158]

[0159] fractal mapping function The calculation can be performed using the following steps:

[0160] 1. Select control parameter λ:

[0161] Assuming λ = 0.8, it indicates that the user has high requirements for acoustic performance.

[0162] 2. Calculate the user demand matrix N:

[0163]

[0164] In practical applications, a finite number of iterations is usually taken, for example, n=10:

[0165] N = λE(1-E) 10 ,

[0166] In home theater design, the system can recommend a suitable sound system configuration based on the user's preferences (such as the type of action movies they like to watch) and the acoustic characteristics of the room, ensuring that the sound effect is optimal.

[0167] The design scheme generation unit 22 is communicatively connected to the user requirement mapping unit 21. Its main task is to receive the user requirement matrix and generate preliminary design schemes based on this information. In one embodiment of the present invention, the design scheme generation employs a generative adversarial network based on group theory and Lie algebra. The advantage of this method is that it can generate more innovative and diverse design schemes. The generation process can be expressed by the following formula:

[0168]

[0169] Here, D is the generated design scheme matrix, G is the generator function, N is the input user requirement matrix, and Z is the random noise matrix. It is a Lie group element, X iIt is a Lie algebra basis. The specific definition of the generator function G is as follows:

[0170] G(NZ)=σ(W g ·[N,Z]+b g ),

[0171] In this formula, σ is the activation function, and W... g and b g These are weights and biases, respectively. Through this generative method based on advanced mathematical theory, the system of this invention can create more innovative and diverse design solutions to meet the personalized needs of different users.

[0172] Suppose there is a commercial space design project, user needs matrix It includes the client's brand image and market positioning information. The random noise matrix Z is an m×k matrix used to introduce diversity. For example:

[0173]

[0174] The generator function G(N,Z) can be calculated through the following steps:

[0175] 1. Weights and biases W g and b g :

[0176] Assume W g It is a k×n matrix, b g It is an n×1 vector.

[0177]

[0178] 2. Activation function σ:

[0179] Use the ReLU activation function σ(x) = max(0,x).

[0180] 3. Calculate the generator output:

[0181] G(NZ)=σ(W g ·[N,Z]+b g ),

[0182] 4. Lie group elements Assume α i =1,X i It is a set of basis vectors that represent different transformation directions.

[0183] In commercial space design, the system can generate multiple design schemes based on the client's brand image and market positioning, showcasing different decoration styles and layouts. For example, the system can generate a modern minimalist design scheme and a European classical design scheme for the client to choose from.

[0184] The multi-dimensional optimization evaluation module 3 of this invention employs functional analysis and variational methods for optimization evaluation. This method can simultaneously evaluate and optimize multiple dimensions such as functionality, aesthetics, and economy. This module includes an objective function construction unit 31, a constraint setting unit 32, and an optimization solution unit 33.

[0185] The main task of the objective function construction unit 31 is to construct an optimization objective function based on functional, aesthetic, and economic indicators. In a preferred embodiment of the present invention, functional indicators may include space utilization rate, pedestrian flow efficiency, etc.; aesthetic indicators may include color harmony, shape balance, etc.; and economic indicators may include material cost, construction difficulty, etc. These indicators are integrated into a comprehensive objective function to achieve multi-objective optimization.

[0186] The constraint setting unit 32 is responsible for setting constraints such as design space, material selection, and cost limits. These constraints ensure that the generated design scheme is feasible and meets actual design requirements and limitations. For example, in one embodiment of the present invention, constraints may include room size restrictions, budget limits, and requirements for the use of specific materials.

[0187] The optimization solution unit 33 is communicatively connected to the objective function construction unit 31 and the constraint setting unit 32. Its main task is to solve for the optimal design scheme based on the objective function and constraints using the variational method. This process can be represented by the following formula:

[0188]

[0189] here, It is the optimized design scheme. Ω is the Lagrange function, and Ω is the design space. The specific definition of the Lagrange function is as follows:

[0190]

[0191] In this formula, f(D) is the objective function, g1, g2, g3 are constraint functions, and λ1, λ2, λ3 are Lagrange multipliers. Through this method, the system of this invention can comprehensively consider multiple design objectives and constraints, obtaining an optimized design scheme that achieves a balance in terms of function, aesthetics, and economy.

[0192] Suppose there is a design project for a high-end restaurant, and the initial design scheme is as follows: It includes information such as the restaurant's layout and decor. Objective function f(D), constraint function g1(D), Let g3(x) represent cost, aesthetics, and practicality, respectively. For example:

[0193] f(D) = 0.5D1 + 0.3D2,

[0194]

[0195]

[0196] g3(x)=x 2 ,

[0197] Lagrange multipliers λ1, λ2, and λ3 are used to adjust the importance of each objective. For example:

[0198]

[0199] Optimization process:

[0200] By solving Find the optimal design solution

[0201] In upscale restaurant design, the system can optimize the layout and decor of the restaurant while meeting budget constraints, thereby enhancing the customer experience. For example, the system can propose a design scheme that is both aesthetically pleasing and practical, based on the restaurant's budget limitations, while ensuring the restaurant's functionality and comfort.

[0202] The immersive visual interaction module 4 of the present invention further includes a 3D rendering unit 41, a VR interaction unit 42, and an AR preview unit 43. These units work together to provide users with a rich visual experience and interactive methods, making the display and adjustment of design schemes more intuitive and convenient.

[0203] The main function of the 3D rendering unit 41 is to receive the optimized design scheme and generate high-quality 3D rendered images using ray tracing technology. In one embodiment of the present invention, ray tracing technology can simulate the propagation of light in space to generate images with realistic lighting effects, including reflection, refraction, and shadows. This method can produce extremely realistic rendering effects, allowing users to more intuitively experience the visual effects of the design scheme.

[0204] The VR interaction unit 42 is communicatively connected to the 3D rendering unit 41. Its main task is to construct a virtual reality scene based on 3D rendered images and provide immersive spatial roaming and real-time design adjustment functions. In a preferred embodiment of the invention, users can enter a virtual design space through VR devices, observe design details from different angles and distances, and even make real-time design modifications, such as changing wall colors or adjusting furniture positions. This interaction method significantly enhances the user experience, making the design process more intuitive and efficient.

[0205] The AR preview unit 43 is also communicatively connected to the 3D rendering unit 41. Its main function is to overlay 3D rendered images onto the actual environment and provide real-time preview and gesture interaction functions for mobile devices. In one embodiment of the invention, users can overlay virtual design schemes onto the actual space using mobile devices such as smartphones or tablets. This augmented reality technology allows users to intuitively experience the effect of the design scheme in the actual space, greatly improving the visualization of the design scheme and the user's decision-making efficiency.

[0206] Through this innovative visualization and interaction method, the system of this invention not only improves the quality of design presentation but also enhances user participation in the design process, thereby better meeting users' personalized needs and increasing design satisfaction. The system of this invention also includes a knowledge base module 5 and an adaptive learning module 6, which work together to achieve continuous system learning and performance optimization.

[0207] The knowledge base module 5 is communicatively connected to the AI ​​design engine module 2. Its main function is to store historical design cases and design rules, providing knowledge support for the AI ​​design engine module 2. In a preferred embodiment of the invention, the knowledge base module 5 adopts a distributed storage architecture, which can efficiently manage and retrieve large amounts of design data. This data includes, but is not limited to, past design schemes, material libraries, and design specifications.

[0208] A key feature of Knowledge Base Module 5 is its dynamic update capability. As new design examples are continuously added, the knowledge base automatically updates and optimizes its content. For example, when a new design solution is approved and implemented by users, its key features and parameters are extracted and added to the knowledge base. This mechanism ensures that the knowledge base always contains the latest and most relevant design information.

[0209] The adaptive learning module 6 is communicatively connected to the knowledge base module 5. Its main task is to analyze user feedback and design effectiveness, and update the design knowledge and rules in the knowledge base module 5 accordingly. In one embodiment of the invention, the adaptive learning module 6 employs a reinforcement learning algorithm, which can continuously optimize the system's design strategy based on user feedback and actual design results.

[0210] Specifically, the adaptive learning module 6 collects user feedback on the generated design solutions, including satisfaction ratings and specific modification suggestions. Simultaneously, it analyzes the performance of implemented designs in actual use, such as space utilization and user comfort metrics. Based on this information, the adaptive learning module 6 adjusts the design parameters and weights in the AI ​​design engine module 2 to improve the quality of future design solutions.

[0211] Preferably, the adaptive learning module 6 also has cross-project learning capabilities. It can extract common knowledge from different types of design projects and apply this knowledge to new design tasks. For example, certain principles learned from office space design may be applied to home design, thereby achieving knowledge transfer and reuse.

[0212] The system of this invention also includes a BIM integration module 7, which is communicatively connected to the AI ​​design engine module 2 and the multi-dimensional optimization and evaluation module 3. The main function of the BIM integration module 7 is to convert the optimized design scheme into a BIM (Building Information Model) model and realize data interaction with the building information model.

[0213] In a preferred embodiment of the present invention, the BIM integration module 7 employs semantic mapping technology, which can automatically convert the system-generated design scheme into a standard BIM format. This process includes not only the conversion of geometric information, but also the mapping of non-geometric data such as material properties, cost information, and construction procedures. For example, a wall design generated by the system will not only be represented as the correct geometric shape in the BIM model, but will also include detailed information such as its material type, thickness, and sound insulation performance.

[0214] Another important function of BIM integration module 7 is to enable bidirectional data flow. It can not only convert system-generated design schemes into BIM models, but also extract information from existing BIM models for use in the system's design process. This bidirectional interaction greatly improves the compatibility and collaborative capabilities of the system with other architectural design tools.

[0215] Preferably, BIM integration module 7 also supports version control and collaborative design. When multiple designers work on a project simultaneously, BIM integration module 7 can manage different versions of design schemes and coordinate modifications made by each designer to ensure design consistency and integrity.

[0216] The AI ​​design engine module 2 of the present invention also includes a style matching unit 23 and a material recommendation unit 24. These two units work together to further improve the personalization and practicality of the design scheme.

[0217] The main function of the style matching unit 23 is to identify design styles based on deep learning algorithms and select matching design elements from a style element library. In one embodiment of the invention, the style matching unit 23 uses a convolutional neural network (CNN) to analyze and identify different design styles. This neural network has been trained on a large number of design cases and can accurately identify various common design styles, such as modern minimalism, Nordic style, and traditional Chinese style.

[0218] The workflow of style matching unit 23 can be described as follows: First, it analyzes the reference images or text descriptions provided by the user to extract key style features. Then, it searches for matching elements in the system's style element library, which may include specific color combinations, furniture types, decorations, etc. Finally, it integrates these matching elements into the design scheme to ensure overall style consistency.

[0219] Preferably, the style matching unit 23 also has the ability to blend styles. When the user's needs involve a mix of multiple styles, it can intelligently balance different style elements to create a unique and harmonious design effect.

[0220] The material recommendation unit 24 is communicatively connected to the style matching unit 23, and its main task is to recommend suitable decorative materials based on the selected design style and environmental characteristics. In a preferred embodiment of the present invention, the material recommendation unit 24 adopts a multi-criteria decision-making method, comprehensively considering the performance, cost, and environmental protection indicators of materials for optimal selection.

[0221] Specifically, the material recommendation unit 24 first determines the range of materials that match the design style based on the output of the style matching unit 23. Then, it combines environmental characteristics (such as indoor temperature and humidity, lighting conditions, etc.) and user needs (such as budget constraints, environmental requirements, etc.) to select the most suitable material combinations. In this process, the material recommendation unit 24 weighs various factors, such as the durability of the materials, maintenance costs, and environmental friendliness, to ensure that the recommended materials are not only aesthetically pleasing but also practical and sustainable.

[0222] Preferably, the material recommendation unit 24 also has an intelligent substitution function. When a certain ideal material cannot be used due to excessive cost or supply issues, it can automatically recommend alternative materials with similar properties and appearance, thereby ensuring the feasibility and flexibility of the design scheme.

[0223] Through the collaborative work of the style matching unit 23 and the material recommendation unit 24, the system of the present invention can generate design solutions that meet both user aesthetic needs and are practically feasible, greatly improving the personalization and usability of the design. The multi-dimensional optimization evaluation module 3 of the present invention also includes a collaborative design unit 34 and a version management unit 35. These two units work together to greatly enhance the system's collaborative capabilities and design efficiency.

[0224] The main function of the collaborative design unit 34 is to support multi-user remote collaborative design, enabling real-time synchronization of design modifications and evaluation results. In a preferred embodiment of the invention, the collaborative design unit 34 employs distributed real-time collaboration technology, allowing multiple designers to work on the same project simultaneously, regardless of their location.

[0225] Specifically, the collaborative design unit 34 provides a shared virtual design space. When a designer modifies a design, these changes are reflected in real time on the interfaces of all participants. For example, one designer might be adjusting the layout of the living room, while another designer is simultaneously selecting decorative materials for the kitchen. The collaborative design unit 34 ensures that these operations are seamlessly integrated, avoiding conflicts and maintaining consistency in the overall design.

[0226] Preferably, the collaborative design unit 34 also features intelligent conflict resolution. When multiple designers make conflicting modifications to the same element, the system automatically detects and alerts the relevant personnel, while providing conflict resolution suggestions. This significantly reduces potential errors and confusion during the collaboration process.

[0227] The version management unit 35 is communicatively connected to the collaborative design unit 34. Its main task is to record the evolution history of the design scheme and support comparison and rollback between different versions. In one embodiment of the present invention, the version management unit 35 adopts a graph-based version control system, which can efficiently manage complex design modification history.

[0228] The workflow of Version Management Unit 35 can be described as follows: Whenever a major change occurs to the design, the system automatically creates a new version. Each version not only contains complete design data but also records metadata such as the specific content of the change, the designer who implemented the change, and the reason for the change. This allows the design team to easily track the evolution of the design and understand the reasons behind each decision.

[0229] Preferably, the version management unit 35 also provides powerful comparison and rollback capabilities. Designers can intuitively compare the differences between any two versions, including visual comparisons of 3D models. If a change is found to have caused a problem, the designer can easily roll back to a previous version or only undo the specific changes without affecting other completed work.

[0230] Through the collaborative work of the collaborative design unit 34 and the version management unit 35, the system of the present invention greatly improves the efficiency and flexibility of team collaboration, while also providing powerful tracking and management tools for the design process.

[0231] Finally, the present invention also provides an intelligent design method for building decoration, which closely corresponds to the above-mentioned system and realizes intelligentization of the entire process from data acquisition to the generation of the final design scheme.

[0232] The method first collects spatial feature information and environmental parameters through the intelligent sensing and analysis module 1 (step S1). In a preferred embodiment of the present invention, this step may involve using a 3D laser scanner to acquire spatial geometric information, while simultaneously collecting environmental data such as illumination, temperature, and humidity through various sensors.

[0233] Next, the method uses the topological persistence operator and the random projection matrix to generate the spatial feature matrix (step S2).

[0234] Then, the method uses a quantum convolutional neural network to fuse the spatial feature matrix and environmental sensor data to generate a fused feature matrix (step S3).

[0235] Next, based on chaos theory and fractal geometry, the method establishes a nonlinear mapping between user needs and environmental characteristics to generate a user needs matrix (step S4).

[0236] Then, the method utilizes a generative adversarial network based on group theory and Lie algebra to generate a preliminary design scheme based on the user demand matrix (step S5).

[0237] Next, the method uses the multi-dimensional optimization evaluation module 3 to evaluate the functionality, aesthetics, and economy of the preliminary design scheme using functional analysis and variational methods (step S6). Based on the evaluation results, the method generates an optimized design scheme (step S7). Then, the immersive visualization interaction module 4 is used to generate a 3D visualization model of the optimized design scheme (step S8).

[0238] The method also provides an interactive design adjustment interface, allowing users to make real-time modifications (step S9). Finally, the design knowledge base is updated by analyzing user feedback and design effectiveness through adaptive learning module 6 (step S10).

[0239] This invention achieves fully automated processing from data acquisition to final design generation, significantly improving the efficiency and quality of architectural decoration design. The method not only generates high-quality initial design schemes but also continuously optimizes them based on user feedback, realizing intelligent and personalized design processes.

[0240] To verify the superiority of this invention, a typical architectural decoration design scenario was selected for simulation testing. Specifically, a three-bedroom, two-living-room house with an area of ​​120 square meters was chosen as the test subject, and a comprehensive interior decoration design was required for it.

[0241] Example 1 utilizes the intelligent architectural decoration design system and method of this invention. Comparative Example 1 employs a traditional manual design method, with an experienced designer completing the entire design process. Comparative Example 2 utilizes commercially available architectural decoration design software, which possesses basic 3D modeling and rendering capabilities but lacks intelligent optimization and adaptive learning abilities.

[0242] The test conditions are as follows:

[0243] 1. Space Information: A 120-square-meter three-bedroom, two-living-room apartment, including a living room, dining room, master bedroom, second bedroom, study, kitchen, and two bathrooms.

[0244] 2. User requirements: Modern minimalist style, focusing on functionality and comfort, with a budget of less than 300,000 yuan.

[0245] 3. Environmental parameters: Good natural lighting conditions, located in the city center, sound insulation needs to be considered.

[0246] 4. Design cycle: The entire process from preliminary plan to final design must be completed within 2 weeks.

[0247] The following five key metrics were selected to evaluate the quality and efficiency of the design scheme:

[0248] 1. Design completion time: The time required from receiving the design task to generating the final solution.

[0249] 2. Number of iterations: The number of modifications and optimizations required to reach the final solution.

[0250] 3. Space utilization rate: The ratio of effectively used space to total area.

[0251] 4. User satisfaction: A comprehensive score based on a questionnaire survey (out of 100).

[0252] 5. Budget control accuracy: The percentage deviation between actual cost and budget.

[0253] The testing methods for these metrics are as follows:

[0254] 1. Design completion time: Record the actual time from project initiation to final solution determination.

[0255] 2. Number of iterations: The number of modifications and optimizations made during the design process.

[0256] 3. Space utilization rate: The ratio of effective space to total area is calculated using a 3D model.

[0257] 4. User satisfaction: Invite 50 potential users to rate the final solution and calculate the average score.

[0258] 5. Budget control accuracy: Calculate the actual cost based on the final plan and compare it with the budget.

[0259] The test results are shown in the table below:

[0260]

[0261] The test results show that the intelligent architectural decoration design system of the present invention exhibits significant advantages in all indicators:

[0262] 1. Design Completion Time: The design of this invention was completed in just 3 days, which is 75% faster than manual design and 62.5% faster than ordinary software. This fully demonstrates the high efficiency of the system of this invention, whose intelligent algorithm can quickly generate and optimize design schemes.

[0263] 2. Number of iterations: This invention requires only 5 iterations to reach the final solution, which is 3 fewer than manual design and 5 fewer than ordinary software. This demonstrates that the multi-dimensional optimization and evaluation module of this invention can more accurately capture user needs and generate higher quality initial solutions.

[0264] 3. Space Utilization: This invention achieves a space utilization rate of 92%, significantly higher than the other two methods. This indicates that the AI ​​design engine of this invention can more intelligently plan the spatial layout and make full use of every inch of space.

[0265] 4. User Satisfaction: This invention achieved a high score of 92 points, far exceeding the other two methods. This reflects that the system of this invention can better understand and meet user needs, and the generated design scheme is more in line with the user's aesthetic and functional requirements.

[0266] 5. Budget Control Precision: The budget deviation of this invention is only 2.5%, which is very precise. This is thanks to the system's intelligent material recommendation and cost optimization functions, which can strictly control costs while ensuring design quality.

[0267] These test results fully demonstrate the superiority of the present invention. It not only significantly improves design efficiency and shortens the design cycle, but also generates higher-quality design solutions that better meet user needs. In particular, the significant improvements in space utilization and user satisfaction demonstrate the powerful capabilities of the system in understanding user needs and optimizing spatial layout.

[0268] Furthermore, the superior performance of this invention's system is also reflected in its adaptive learning capability. Although this cannot be directly reflected in this single test case, it is foreseeable that as the system processes more projects, its performance will continuously improve, and the quality of the design solutions will become increasingly higher—an advantage that traditional methods and ordinary software do not possess.

[0269] In summary, the intelligent architectural decoration design system and method of the present invention are significantly superior to the prior art in terms of efficiency, quality and user satisfaction, bringing revolutionary progress to the field of architectural decoration design.

[0270] It should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An intelligent design system for architectural ornamentation, characterized in that, The application comprises: An intelligent perception analysis module for: Collecting spatial feature information and environmental parameters; Generating a fusion feature matrix based on the spatial feature information and environmental parameters; An AI design engine module in communication with the intelligent perception analysis module for: Receiving the fusion feature matrix; Generating a preliminary design scheme based on the fusion feature matrix; A multi-dimensional optimization evaluation module in communication with the AI design engine module for: Functionality, aesthetics, and economic evaluation of the preliminary design scheme; Generating an optimized design scheme based on the evaluation results; An immersive visualization interaction module in communication with the multi-dimensional optimization evaluation module for: Receiving the optimized design scheme; Generating a three-dimensional visualization model of the optimized design scheme; Providing an interactive design adjustment interface; The intelligent perception analysis module comprises: A spatial feature extraction unit for: Collecting point cloud data; Generating a spatial feature matrix based on the point cloud data through topological persistence operators and random projection matrices; An environmental feature fusion unit in communication with the spatial feature extraction unit for: Receiving the spatial feature matrix and environmental sensor data; Fusing the spatial feature matrix and environmental sensor data through a quantum convolutional neural network to generate a fusion feature matrix; The AI design engine module comprises: A user demand mapping unit for: Receiving the fusion feature matrix; Establishing a nonlinear mapping between user demand and environmental features based on chaos theory and fractal geometry to generate a user demand matrix; A design scheme generation unit in communication with the user demand mapping unit for: Receiving the user demand matrix; Constructing a generative adversarial network based on group theory and Lie algebra to generate a preliminary design scheme.

2. The architectural finishing intelligent design system of claim 1, wherein, The multi-dimensional optimization evaluation module uses functional analysis and variational methods for optimization evaluation, including: A target function construction unit for: Constructing an optimization target function based on functionality, aesthetics, and economic indicators; A constraint condition setting unit for: Setting design space, material selection, and cost limit constraint conditions; An optimization solving unit in communication with the target function construction unit and constraint condition setting unit for: Solving the optimal design scheme through variational methods based on the target function and constraint conditions.

3. The architectural finishing intelligent design system of claim 1, wherein, The immersive visualization interaction module comprises: A 3D rendering unit for: Receiving the optimized design scheme; Generating high-quality three-dimensional rendering images using ray tracing technology; A VR interaction unit in communication with the 3D rendering unit for: Constructing a virtual reality scene based on the three-dimensional rendering images; Providing immersive space roaming and real-time design adjustment functions; An AR preview unit in communication with the 3D rendering unit for: Superimposing the three-dimensional rendering images onto the actual environment; Providing mobile real-time preview and gesture interaction functions.

4. The architectural finishing intelligent design system of claim 1, wherein, Further comprising: A knowledge base module in communication with the AI design engine module for: Storing historical design cases and design rules; Providing knowledge support for the AI design engine module; An adaptive learning module in communication with the knowledge base module for: Analyzing user feedback and design effectiveness; updating design knowledge and rules in the knowledge base module.

5. The architectural finishing smart design system of claim 1, wherein, Further comprising: a BIM integration module communicatively connected with the AI design engine module and the multi-dimensional optimization evaluation module, configured to: convert the optimized design scheme into a BIM model; achieve data interaction with a building information model.

6. The architectural finishing smart design system of claim 1, wherein, The AI design engine module further comprises: a style matching unit, configured to: identify design styles based on deep learning algorithms; select matching design elements from a style element library; a material recommendation unit communicatively connected with the style matching unit, configured to: recommend suitable decorative materials based on the selected design style and environmental characteristics; optimize selection considering material performance, cost, and environmental indicators.

7. The architectural finishing smart design system of claim 1, wherein, The multi-dimensional optimization evaluation module further comprises: a collaborative design unit, configured to: support multi-user remote collaborative design; real-time synchronization of design modifications and evaluation results; a version management unit communicatively connected with the collaborative design unit, configured to: record the evolution history of the design scheme; support comparison and rollback between different versions.

8. A method of intelligent design of architectural decoration, based on the system according to any one of claims 1-7, characterized in that, The method comprises the following steps: S1, collecting space feature information and environmental parameters through an intelligent perception analysis module; S2, generating a space feature matrix based on the space feature information and environmental parameters through a topological persistence operator and a random projection matrix; S3, using a quantum convolutional neural network to fuse the space feature matrix and environmental sensor data to generate a fused feature matrix; S4, establishing a nonlinear mapping between user demand and environmental characteristics based on chaos theory and fractal geometry to generate a user demand matrix; S5, generating a preliminary design scheme based on the user demand matrix using a generative adversarial network based on group theory and Lie algebra; S6, using a multi-dimensional optimization evaluation module to perform functional, aesthetic, and economic evaluation of the preliminary design scheme using functional analysis and variational methods; S7, generating an optimized design scheme based on the evaluation results; S8, generating a three-dimensional visual model of the optimized design scheme using an immersive visual interaction module; S9, providing an interactive design adjustment interface to support real-time modification by users; S10, updating the design knowledge base through an adaptive learning module to analyze user feedback and design effects.

Images