Intelligent building design method and system based on artificial intelligence generation content

Abstract

The invention discloses an intelligent building design method and system based on artificial intelligence generation content, and relates to the technical field of building information models.By collecting building related data of a city to-be-built area, building a building related data set S, obtaining heat island effect intensity UHI of the city to-be-built area and building a heat island effect intensity prediction model, the building related data set S is obtained; the method comprises the following steps: acquiring the heat island effect intensity UHI (tn) of a to-be-built area of a city at a future time point n, thereby acquiring a heat island effect intensity average change rate # imgabs0 # to evaluate the heat island effect intensity of the to-be-built area of the city, acquiring a plurality of candidate building design schemes according to a GAN (Generative Adversarial Network), screening the building design schemes, outputting a first building design scheme, and outputting a second building design scheme according to the first building design scheme. And a deep reinforcement learning optimization algorithm is used to optimize the building facade, obtain an optimal building design scheme, execute the building engineering, reduce the influence of the urban heat island effect, improve the building energy efficiency and reduce the building energy consumption.

Description

Technical Field

[0001] The present invention relates to the technical field of building information modeling, and in particular to an intelligent building design method and system based on artificial intelligence content generation. Background Art

[0002] In the field of architecture and urban planning, smart buildings have become an important development direction of the global construction industry in recent years. Smart buildings combine emerging technologies such as the Internet of Things, big data analysis and artificial intelligence to optimize the building environment, improve energy efficiency and improve living comfort. In the research and application of smart buildings, artificial intelligence generated content (AIGC) technology has gradually penetrated into the architectural design stage, using computer algorithms to generate architectural plans, optimize spatial layout, and improve the ability of buildings to adapt to complex environments. Among them, microclimate control is an important part of smart buildings, especially in the core areas of cities with dense high-rise buildings. The dense arrangement of buildings will affect air circulation and heat accumulation, resulting in abnormal temperature increases in local areas, namely the urban heat island effect. How to optimize the building shape, window ratio and greening strategy through intelligent design to form a benign interaction between buildings and microclimate has become one of the key issues in current smart building research.

[0003] In the Chinese invention patent with application publication number CN118171369A, an artificial intelligence-based architectural design method is disclosed. The architectural design method specifically includes the following steps: S1. Data collection and integration. Before the design begins, the system automatically collects and integrates relevant geographic information system data, environmental data, community cultural characteristics and real-time regulatory databases; S2. Demand analysis and goal setting. In-depth communication with customers, using artificial intelligence-assisted tools to collect their needs and preferences, and using natural language processing technology to analyze this information. By enabling the architectural design process to better predict and mitigate the impact of buildings on the local environment through community participation and cultural considerations, as well as the use of environmental simulation and assessment tools, the introduction of automatic checkers simplifies the regulatory compliance inspection process, ensuring that the design meets the latest regulatory requirements, ensuring that the design is more local and culturally sensitive, while reducing compliance risks and negative environmental impacts.

[0004] The above architectural design methods can improve the energy efficiency and environmental friendliness of buildings, enhance environmental adaptability, and fully consider and respect local culture, traditions and social values, so as to design buildings with more local characteristics and cultural resonance, and strengthen cultural sensitivity. However, in addition to this, in the existing architectural design methods, architectural design is usually carried out based on the design experience of architects;

[0005] However, this architectural design method has a relatively simple analysis method for microclimate, often relying on empirical judgment or limited climate simulation tools, and it is difficult to provide accurate predictions and optimization solutions for building facade optimization, air flow simulation, etc. Summary of the invention

[0006] In view of the deficiencies in the prior art, the present invention provides an intelligent building design method and system based on artificial intelligence content generation, which solves the problems in the above-mentioned background technology.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: an intelligent building design system based on artificial intelligence generated content, including a data acquisition and preprocessing module, a microclimate prediction and evaluation module, a building design solution generation module, a building facade optimization module and an intelligent feedback module;

[0008] The data acquisition and preprocessing module is used to deploy an intelligent sensor group in the urban area to be built, and in combination with the urban climate monitoring system, obtain the building-related data of the urban area to be built, and preprocess the building-related data of the urban area to be tested to construct a building-related data set S;

[0009] The microclimate prediction and evaluation module is used to perform summary calculations based on the building-related data set S to obtain the urban heat island effect intensity UHI of the area to be built, and use the LSTM algorithm to build a heat island effect intensity prediction model to obtain the average change rate of the heat island effect intensity. Evaluate the intensity change of the urban heat island effect in the area to be built in the future time point n. If the intensity change of the urban heat island effect in the area to be built is abnormal, the building design scheme generation module is carried out;

[0010] The architectural design scheme generation module is used to obtain a number of candidate architectural design schemes based on a generative adversarial network GAN, obtain a comprehensive score PF of the architectural design scheme, screen the architectural design schemes, and output a first architectural design scheme;

[0011] The building facade optimization module is used to optimize the building facade according to the first building design scheme to obtain the optimal building design scheme;

[0012] The intelligent feedback module is used to execute the construction project according to the optimal building design plan, and continuously collect building-related data, output the building-related data to the artificial intelligence generated content AIGC, and optimize the building design plan.

[0013] Preferably, the data acquisition and preprocessing module includes a data acquisition unit and a preprocessing unit;

[0014] The data acquisition unit is used to deploy an intelligent sensor group in the urban area to be built, and collect building-related data in the urban area to be built in combination with the urban climate monitoring system, meteorological station and remote sensing satellite, wherein the building-related data includes climate data, building design data and ground reflection data;

[0015] The climate data refers to the temperature T of the area to be built in the city d and the temperature T outside the urban area to be built urban ;

[0016] The architectural design data refers to building facade material parameters, building form and window types;

[0017] The ground reflection data refers to the solar radiation reflectivity β;

[0018] The preprocessing unit is used to preprocess the collected climate data, building design data and ground reflection data, wherein the preprocessing includes data cleaning, denoising and data standardization, and constructs a building-related data set S based on the preprocessed climate data, building design data and ground reflection data of the urban area to be built.

[0019] Preferably, the microclimate prediction and assessment module includes a heat island effect intensity prediction unit and an assessment unit;

[0020] The heat island effect intensity prediction unit is used to perform summary calculation based on the building-related data set S to obtain the heat island effect intensity UHI of the urban area to be built. The heat island effect intensity UHI is obtained in the following manner:

[0021] UHI=T urban，avg -T d，avg ;

[0022] Among them, T urban，avg represents the average temperature of the area outside the urban area to be built, T d，avg Indicates the average temperature of the area to be built in the city;

[0023] Based on the LSTM algorithm, a heat island effect intensity prediction model is constructed, and relevant data of historical buildings are collected. The relevant data of historical buildings are constructed as a training set, which is input into the heat island effect intensity prediction model for model training. The mean square error method is used to optimize the heat island effect intensity prediction model parameters, and the trained heat island effect intensity prediction model is used to obtain the heat island effect intensity UHI (t n ).

[0024] Preferably, the evaluation unit is used to determine the heat island effect intensity UHI (t n), construct a heat island effect intensity time dataset M, wherein the heat island effect intensity time dataset M is specifically expressed in the form of:

[0025] M=[UHI(t 1 ), UHI(t 2 ), UHI(t 3 ), ..., UHI(t n )];

[0026] In the formula, UHI(t n ) represents the predicted time point t n is the intensity of the urban heat island effect in the area to be built at that time, and n represents the total length of the predicted time period;

[0027] According to the heat island effect intensity time data set M, the average change rate of heat island effect intensity is obtained The average change rate of the intensity of the heat island effect The method of obtaining is:

[0028]

[0029] In the formula, UHI(t i+1 ) represents the predicted time point t i+1 The intensity of the urban heat island effect in the area to be built at that time, i = [1, 2, 3, ..., n];

[0030] Preset heat island effect intensity change threshold The average change rate of the heat island effect is obtained Threshold of heat island effect intensity change Comparative analysis is conducted to evaluate the change in the intensity of the urban heat island effect in the urban area to be built at the future time point n. The specific evaluation contents are as follows:

[0031] If the average change rate of the heat island effect intensity Greater than the heat island effect intensity change threshold Right now It is determined that the intensity change of the urban heat island effect in the area to be built is in an abnormal state, and the building design plan is generated at this time;

[0032] If the average change rate of the heat island effect intensity Less than or equal to the heat island effect intensity change threshold Right now It is determined that the intensity change of the urban heat island effect in the area to be built is in a normal state and no treatment is required.

[0033] Preferably, the architectural design scheme generating module includes a scheme generating unit, a scheme analyzing unit and a scheme screening unit;

[0034] The scheme generating unit is used to determine the building design goal according to the building design data in the building related data set S, wherein the building design data includes building design parameters and building constraints, and the building design goal includes energy consumption and heat island effect intensity optimization;

[0035] Based on the generative adversarial network GAN, combined with the architectural design data and the architectural design goals, several candidate architectural design schemes are generated and formed into a set of candidate architectural design schemes H;

[0036] The candidate building design scheme generation process is as follows:

[0037] The collected architectural design data and architectural design goals are input into the generative adversarial network (GAN) generator. The generative adversarial network (GAN) receives the architectural design data and architectural design goals, and randomly generates a number of candidate architectural design schemes, wherein the specific contents of the candidate architectural design schemes include building type, building area, number of floors and functional zoning.

[0038] Preferably, the scheme analysis unit is used to extract building-related data from each candidate building design scheme in the candidate building design scheme set H, and perform summary calculation to obtain the heat island effect intensity UHI of the urban area to be built after the building is constructed. after , building energy consumption E and building cost CB;

[0039] The intensity of the urban heat island effect UHI in the area to be built after the construction of the building after The method of obtaining is:

[0040]

[0041] Where β represents the solar radiation reflectivity, T urban Indicates the temperature of the area outside the city to be built, T d represents the temperature of the urban area to be built, f building Represents the building form influence coefficient, V building Represents the volume of the building, Indicates the greening coefficient, A rool，green Indicates the green area on the top of the building, A wall，green Indicates the green area on the building wall;

[0042] The building energy consumption E is obtained in the following way:

[0043] E=f material ·A wall +f window ·A window ;

[0044] In the formula, f material represents the thermal conductivity of building materials, fwindow Indicates the thermal conductivity of the window material, A wall Represents the building wall area, A window Indicates the building window area;

[0045] The construction cost CB is obtained in the following way:

[0046]

[0047] Where V j represents the volume of the building in the jth part, C material，j represents the unit material cost of the jth part of the building, m represents the total number of building parts, C labor represents the daily labor cost, T construction represents the construction period, C equipment represents the equipment usage cost, j = [1, 2, 3, …, m].

[0048] Preferably, the scheme screening unit is used to determine the heat island effect intensity UHI of the urban area to be built after the building is constructed. after , building energy consumption E and building cost CB, and perform summary calculation to obtain the comprehensive score PF of the building design scheme. The comprehensive score PF of the building design scheme is obtained as follows:

[0049] PF=ω 1 ·UHI after +ω 2 ·E+ω 3 CB;

[0050] In the formula, ω 1 ,ω 2 and ω 3 They represent the heat island effect intensity UHI of the urban unbuilt area after the construction of the building after , weight coefficients of building energy consumption E and building cost CB;

[0051] According to the candidate architectural design scheme set H, the architectural design scheme comprehensive score PF of each candidate architectural design scheme is obtained, and compared, and the candidate architectural design scheme with the smallest architectural design scheme comprehensive score PF is selected and output as the first architectural design scheme.

[0052] Preferably, the building facade optimization module is used to optimize the building facade design according to the first building design scheme using a deep reinforcement learning optimization algorithm, and the specific optimization process includes initializing the design, defining the state space, defining the action space, setting the reward function and the training process;

[0053] The initialization design refers to extracting specific contents of the first architectural design scheme according to the first architectural design scheme;

[0054] The defining state space refers to obtaining the building facade state space Z according to the first building design scheme, and the building facade state space Z is specifically expressed as:

[0055] Z = {material reflectivity, green area, window ratio};

[0056] The defined action space refers to optimizing and adjusting according to the first building design scheme, including changing the reflectivity of the facade material, increasing or decreasing the window opening ratio, increasing or decreasing the roof greening area, and obtaining the building facade action space D. The building facade action space D is specifically expressed as:

[0057] D = [increase greening, reduce greening, increase window ratio, reduce window ratio, increase reflectivity, reduce reflectivity];

[0058] The setting of the reward function refers to setting the reward function R for the purpose of minimizing the intensity of the heat island effect and maximizing energy efficiency. The specific form of the reward function R is:

[0059]

[0060] Where, UHI red represents the reduction in the intensity of the heat island effect, E imp represents the increase in energy consumption, CB represents the building cost, and Represents the reduction in the intensity of the heat island effect, UHI red , Energy consumption increase E imp and the weight coefficient of the building cost CB;

[0061] The training process refers to obtaining the initial state S of the scheme according to the first architectural design scheme. 0 , where the initial state of the scheme is S 0 Including the reflectivity of the building facade material, the roof greening area and the window ratio, and select any one of the building facade action space D for optimization, and calculate the reward function value R of the optimized building design scheme, and use the Q-learning update formula to update the Q value of the state-action pair of the current building design scheme. The Q-learning update formula is:

[0062]

[0063] In the formula, Q new (s t , d t ) represents the updated Q value, Q(s t , d t ) indicates the current state st Next, select Action d t The expected long-term reward after α represents the learning rate, R t Indicates execution of action d t The immediate reward obtained after γ represents the discount factor. Indicates the next state s t+1 The maximum Q value of all possible actions d′;

[0064] According to the Q value of the state-action pair of the current architectural design scheme, the state-action pair with the largest Q value is selected to update the architectural design scheme, and any item in the building facade action space D is reselected for optimization. The Q value is iterated multiple times until the Q value no longer changes, and the architectural design scheme at this time is output as the optimal architectural design scheme.

[0065] Preferably, the intelligent feedback module is used to execute the construction project according to the optimal building design plan, and monitor the building-related data during the implementation of the construction project in real time, and feed back the data to the artificial intelligence generated content ALGC to dynamically adjust the building design plan.

[0066] Preferably, a smart building design method based on artificial intelligence content generation comprises the following steps:

[0067] Step 1: deploy a smart sensor group in the urban area to be built, and combine it with the urban climate monitoring system to obtain the building-related data of the urban area to be built, and pre-process the building-related data of the urban area to be tested to construct a building-related data set S;

[0068] Step 2: Based on the building-related data set S, a summary calculation is performed to obtain the urban heat island effect intensity UHI of the area to be built in the city, and a heat island effect intensity prediction model is constructed using the LSTM algorithm to obtain the average change rate of the heat island effect intensity Evaluate the change in the intensity of the urban heat island effect in the area to be built in the future time point n. If the intensity of the urban heat island effect in the area to be built is in an abnormal state, the building design scheme generation module is carried out;

[0069] Step 3: Based on the generative adversarial network GAN, several candidate architectural design schemes are obtained, and the comprehensive score PF of the architectural design scheme is obtained, the architectural design scheme is screened, and the first architectural design scheme is output;

[0070] Step 4: Optimize the building facade according to the first building design scheme to obtain the optimal building design scheme;

[0071] Step 5: According to the optimal building design plan, execute the construction project, continuously collect building-related data, output the building-related data to the artificial intelligence generated content AIGC, and optimize the building design plan.

[0072] The present invention provides an intelligent building design method and system based on artificial intelligence content generation, which has the following beneficial effects:

[0073] (1) By combining the microclimate prediction and assessment module, the building design scheme generation module and the building facade optimization module, the intelligent building design system can predict and respond to future changes in the urban heat island effect and optimize the building design scheme based on this change. The system can automatically adjust the building facade design, such as reflectivity, roof greening and window ratio, according to the prediction results of the heat island effect intensity in real time through the generative adversarial network GAN and deep reinforcement learning algorithm, thereby reducing the impact of the building's heat island effect on the environment and improving the building's energy efficiency. This approach will reduce building energy consumption, reduce the demand for air conditioning and heating, and improve the energy efficiency and environmental friendliness of the building.

[0074] (2) By combining the data acquisition and preprocessing module, the microclimate prediction and evaluation module, and the intelligent feedback module, the architectural design can be optimized based on real-time monitoring and evaluation of environmental conditions. In the process of generating and screening candidate design schemes, the system automatically evaluates the comprehensive score PF of each architectural design scheme and further optimizes the building facade through the reinforcement learning algorithm. This process reduces the manual intervention of designers, improves design efficiency and accuracy, and can make adaptive adjustments according to actual environmental changes. This optimization decision-making process can accelerate the implementation of architectural design and improve the scientificity and sustainability of the design.

[0075] (3) By collecting building-related data in real time during the implementation of the construction project and feeding it back to the artificial intelligence generated content AIGC for adjustment, it ensures that the building design can adapt to the ever-changing environmental conditions. As the building is constructed and operated, the system can continuously optimize the building design based on real-time feedback, continuously reduce the impact of the heat island effect, improve building energy efficiency and comfort, and improve the sustainability and adaptability of the building in long-term operation. This continuous adaptive optimization capability enables the building to better cope with future climate change and urban development needs, and improve the long-term benefits of the building. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 This is a block diagram of an intelligent building design system based on artificial intelligence content generation according to the present invention.

[0077] Figure 2 The present invention is a flowchart of an intelligent building design method based on artificial intelligence content generation.

[0078] Figure 3 This is a schematic diagram of the module flow of the building design scheme generation module of the present invention.

[0079] Figure 4 It is a schematic diagram of the flow of the building facade optimization module of the present invention. DETAILED DESCRIPTION

[0080] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0081] Example 1

[0082] See also Figure 1 , Figure 3 and Figure 4 , the present invention provides an intelligent building design system based on artificial intelligence generated content, including a data acquisition and preprocessing module, a microclimate prediction and evaluation module, a building design solution generation module, a building facade optimization module and an intelligent feedback module;

[0083] The data acquisition and preprocessing module is used to deploy an intelligent sensor group in the urban area to be built, and in combination with the urban climate monitoring system, obtain the building-related data of the urban area to be built, and preprocess the building-related data of the urban area to be tested to construct a building-related data set S;

[0084] The microclimate prediction and evaluation module is used to perform summary calculations based on the building-related data set S to obtain the urban heat island effect intensity UHI of the area to be built, and use the LSTM algorithm to build a heat island effect intensity prediction model to obtain the average change rate of the heat island effect intensity. Evaluate the intensity change of the urban heat island effect in the area to be built in the future time point n. If the intensity change of the urban heat island effect in the area to be built is abnormal, the building design scheme generation module is carried out;

[0085] The architectural design scheme generation module is used to obtain a number of candidate architectural design schemes based on a generative adversarial network GAN, obtain a comprehensive score PF of the architectural design scheme, screen the architectural design schemes, and output a first architectural design scheme;

[0086] The building facade optimization module is used to optimize the building facade according to the first building design scheme to obtain the optimal building design scheme;

[0087] The intelligent feedback module is used to execute the construction project according to the optimal building design plan, and continuously collect building-related data, output the building-related data to the artificial intelligence generated content AIGC, and optimize the building design plan.

[0088] In the embodiment, by adopting the data acquisition and preprocessing module, the microclimate prediction and evaluation module, the building design scheme generation module, the building facade optimization module and the intelligent feedback module, the intelligent building design system can achieve efficient building optimization and continuous improvement. First, the data acquisition and preprocessing module collects comprehensive urban climate and building data through intelligent sensors and climate monitoring systems to ensure the accuracy and real-time nature of the data. Then, the microclimate prediction and evaluation module uses these data to predict the intensity of the heat island effect UHI, and uses the LSTM algorithm to analyze the changing trend of the future heat island effect. By comparing with the set threshold, it is determined whether there is an abnormality. If an abnormality is detected, the system will start the building design. The solution generation module uses the generative adversarial network GAN to generate multiple candidate solutions, and selects the first building design solution based on the comprehensive score PF of the building design solution. Then, the building facade optimization module uses deep reinforcement learning to further optimize the building facade, thereby reducing the heat island effect, improving building energy efficiency and reducing costs. Finally, the intelligent feedback module ensures that the building design solution can be adaptively adjusted by collecting data during the construction implementation process in real time. Overall, the system not only improves the environmental protection, energy efficiency and economy of the building design through intelligent data processing, prediction, generation and optimization processes, but also enables the building to dynamically adjust according to environmental changes, thereby improving the long-term sustainability of the building.

[0089] Example 2

[0090] See also Figure 1 , Figure 3 and Figure 4 ,Specifically: the data acquisition and preprocessing module includes a data acquisition unit and a preprocessing unit;

[0091] The data acquisition unit is used to deploy an intelligent sensor group in the urban area to be built, and collect building-related data in the urban area to be built in combination with the urban climate monitoring system, meteorological station and remote sensing satellite, wherein the building-related data includes climate data, building design data and ground reflection data;

[0092] The intelligent sensor group includes a temperature sensor and a reflection photometer;

[0093] The climate data refers to the temperature T of the area to be built in the city d and the temperature T outside the urban area to be built urban , obtained through temperature sensor;

[0094] The said architectural design data refers to the building facade material parameters, building form and window type, which are obtained by collecting customer wishes and material database;

[0095] The ground reflection data refers to the solar radiation reflectivity β, which is obtained by a reflection photometer;

[0096] The preprocessing unit is used to preprocess the collected climate data, building design data and ground reflection data, wherein the preprocessing includes data cleaning, denoising and data standardization, and constructs a building-related data set S based on the preprocessed climate data, building design data and ground reflection data of the urban area to be built.

[0097] In the embodiment, through the precise collection and preprocessing of climate data, building design data and ground reflection data, the data collection and preprocessing module provides a data basis for the system, which not only enhances the scientificity and operability of the building design, but also provides a stable input for subsequent optimization, ensuring the efficiency and sustainability of the building design in terms of energy saving and cooling.

[0098] Example 3

[0099] See also Figure 1 , Figure 3 and Figure 4 ,Specifically: the microclimate prediction and evaluation module includes a heat island effect intensity prediction unit and an evaluation unit;

[0100] The heat island effect intensity prediction unit is used to perform summary calculation based on the building-related data set S to obtain the heat island effect intensity UHI of the urban area to be built. The heat island effect intensity UHI is obtained in the following manner:

[0101] UHI=T urban，avg -T d，avg ;

[0102] Among them, T urban，avg represents the average temperature of the area outside the urban area to be built, T d，avg Indicates the average temperature of the area to be built in the city;

[0103] Based on the LSTM algorithm, a heat island effect intensity prediction model is constructed, and relevant data of historical buildings are collected. The relevant data of historical buildings are constructed as a training set, which is input into the heat island effect intensity prediction model for model training. The mean square error method is used to optimize the heat island effect intensity prediction model parameters, and the trained heat island effect intensity prediction model is used to obtain the heat island effect intensity UHI (t n ).

[0104] The evaluation unit is used to construct a heat island effect intensity time dataset M according to the heat island effect intensity UHI of the urban area to be built in the future time period, wherein the heat island effect intensity time dataset M is specifically expressed in the form of:

[0105] M=[UHI(t 1 ), UHI(t 2 ), UHI(t3 ), ..., UHI(t n )];

[0106] In the formula, UHI(t n ) represents the predicted time point t n is the intensity of the urban heat island effect in the area to be built at that time, and n represents the total length of the predicted time period;

[0107] According to the heat island effect intensity time data set M, the average change rate of heat island effect intensity is obtained The average change rate of the intensity of the heat island effect The method of obtaining is:

[0108]

[0109] In the formula, UHI(t i+1 ) represents the predicted time point t i+1 The intensity of the urban heat island effect in the area to be built at that time, i = [1, 2, 3, ..., n];

[0110] Preset heat island effect intensity change threshold The average change rate of the heat island effect is obtained Threshold of heat island effect intensity change Comparative analysis is conducted to evaluate the change in the intensity of the urban heat island effect in the urban area to be built at the future time point n. The specific evaluation contents are as follows:

[0111] If the average change rate of the heat island effect intensity Greater than the heat island effect intensity change threshold Right now It is determined that the intensity change of the urban heat island effect in the area to be built is in an abnormal state, and the building design plan is generated at this time;

[0112] If the average change rate of the heat island effect intensity Less than or equal to the heat island effect intensity change threshold Right now It is determined that the intensity change of the urban heat island effect in the area to be built is in a normal state and no treatment is required;

[0113] The temperature data of the urban area to be built in the past month were collected, as shown in Table 1 below:

[0114]

[0115] Table 1

[0116] According to the heat island effect intensity prediction model, the heat island effect intensity UHI in the next 5 days is predicted, as shown in Table 2 below:

[0117] date <![CDATA[Predicted T urban > <![CDATA[Predicted T d > Predicted UHI(℃) October 18, 2024 30.0 29.0 4.0 October 19, 2024 33.2 29.2 4.0 October 20, 2024 33.4 29.5 3.9 October 21, 2024 33.5 29.7 3.8 October 22, 2024 33.6 30.0 3.6

[0118] Table 2

[0119] According to the above table, the heat island effect intensity time dataset M is obtained. The heat island effect intensity time dataset M is specifically expressed as:

[0120] M=[4.0, 4.0, 3.9, 3.8, 3.6];

[0121] Calculate the average rate of change of heat island effect intensity

[0122]

[0123] The average change rate of the intensity of the heat island effect Threshold of heat island effect intensity change For comparison, Setting the threshold for changes in the intensity of the heat island effect is -0.2, -0.1>-0.2, it is determined that the city's unbuilt area is in an abnormal state, and the building design plan generation module is required.

[0124] In the embodiment, by constructing a microclimate prediction and evaluation module, especially a heat island effect intensity prediction unit and an evaluation unit, the heat island effect change trend of the urban area to be built can be evaluated in real time and accurately. First, the heat island effect intensity prediction unit uses the LSTM algorithm to construct a heat island effect intensity prediction model, trains it, and predicts the heat island effect intensity of the urban area to be built in the future time period, and optimizes the heat island effect intensity prediction model parameters by the mean square error method, and optimizes the heat island effect intensity prediction model. Secondly, the evaluation module calculates the heat island effect intensity UHI (t n ), construct the heat island effect intensity time dataset M, and calculate the average change rate of the heat island effect intensity By comparing with the preset heat island effect intensity change threshold By conducting comparative analysis, the system can promptly identify abnormal changes in the heat island effect, and then guide the generation of subsequent building design plans to alleviate the regional heat island effect and improve the environmental adaptability and energy efficiency of buildings. This process not only improves the environmental quality of the city, but also provides a scientific basis for architectural design, avoids over-reliance on empirical design methods, and ensures that design plans can effectively respond to future climate change.

[0125] Example 4

[0126] See also Figure 1 , Figure 3 and Figure 4 ,Specifically: the architectural design scheme generation module includes a scheme generation unit, a scheme analysis unit and a scheme screening unit;

[0127] The scheme generating unit is used to determine the building design goal according to the building design data in the building related data set S, wherein the building design data includes building design parameters and building constraints, and the building design goal includes energy consumption and heat island effect intensity optimization;

[0128] Based on the generative adversarial network GAN, combined with the architectural design data and the architectural design goals, several candidate architectural design schemes are generated and formed into a set of candidate architectural design schemes H;

[0129] The generative adversarial network GAN includes a generator and a discriminator, wherein the task of the generator is to generate a building design scheme based on input data, including building facades, structural layout and greening design, and its output is a specific building design scheme. The task of the discriminator is to evaluate whether the generated design scheme meets the requirements and evaluate the impact of the building design scheme on the heat island effect;

[0130] The candidate building design scheme generation process is as follows:

[0131] The collected architectural design data and architectural design goals are input into a generative adversarial network (GAN) generator. The generative adversarial network (GAN) receives the architectural design data and architectural design goals, and randomly generates a number of candidate architectural design schemes, wherein the candidate architectural schemes include building type, building area, number of floors and functional zoning.

[0132] The scheme analysis unit is used to extract building-related data from each candidate building design scheme in the candidate building design scheme set H, perform summary calculations, and obtain the heat island effect intensity UHI of the urban area to be built after the building is built. after , building energy consumption E and building cost CB;

[0133] The intensity of the urban heat island effect UHI in the area to be built after the construction of the building after The method of obtaining is:

[0134]

[0135] Where β represents the solar radiation reflectivity, T urban Indicates the temperature of the area outside the city to be built, T d represents the temperature of the urban area to be built, f building Represents the building form influence coefficient, V building Represents the volume of the building, Indicates the greening coefficient, A rool，green Indicates the green area on the top of the building, A wall，green Indicates the green area on the building wall;

[0136] The building form influence coefficient f building The building form is experimentally obtained by modeling it using a fluid dynamics model;

[0137] The greening factor Obtained through building design codes and green building standards databases;

[0138] The volume V of the building building 、Green area on the top of the building A rool，green and the green area A on the building wall wall，green Obtained through candidate architectural design proposals;

[0139] The building energy consumption E is obtained in the following way:

[0140] E=f material ·A wall +f window ·A window ;

[0141] In the formula, f material represents the thermal conductivity of building materials, f window Indicates the thermal conductivity of the window material, A wall Represents the building wall area, A window Indicates the building window area;

[0142] The thermal conductivity of the building material f material and the thermal conductivity f of the window material window Obtained through the building materials database;

[0143] The building wall area A wall and building window area A window Obtained through candidate architectural design proposals;

[0144] The construction cost CB is obtained in the following way:

[0145]

[0146] Where V j represents the volume of the building in the jth part, C material，j represents the unit material cost of the jth part of the building, m represents the total number of building parts, C labor represents the daily labor cost, T construction represents the construction period, C equipment represents the equipment usage cost, j = [1, 2, 3, …, m].

[0147] The daily labor cost C labor and construction period T construction Obtained through the database of construction contractors;

[0148] The equipment usage cost C equipment Obtained from the equipment management department.

[0149] The scheme screening unit is used to screen the heat island effect intensity UHI of the urban area to be built after the building is built. after , building energy consumption E and building cost CB, and perform summary calculation to obtain the comprehensive score PF of the building design scheme. The comprehensive score PF of the building design scheme is obtained as follows:

[0150] PF=ω 1 ·UHI after +ω 2 ·E+ω 3 CB;

[0151] In the formula, ω 1 ,ω 2 and ω 3 They represent the heat island effect intensity UHI of the urban unbuilt area after the construction of the building after , the weight coefficient of building energy consumption E and building cost CB, where the specific value of the weight coefficient is set by the customer according to the actual situation, 0<ω 1 <1,0<ω 2 <1,0<ω 3 <1, and ω 1 +ω 2 +ω 3 =1;

[0152] According to the candidate architectural design scheme set H, the architectural design scheme comprehensive score PF of each candidate architectural design scheme is obtained, and compared, and the candidate architectural design scheme with the smallest architectural design scheme comprehensive score PF is selected and output as the first architectural design scheme.

[0153] In the embodiment, by combining the generative adversarial network (GAN) technology, the system can automatically generate a variety of candidate design schemes according to the building design goals, such as heat island effect optimization and energy consumption reduction. The generator creates a plurality of candidate building design schemes based on the input design data, covering different building types, floor layouts, functional zoning and other diversified designs. At the same time, the discriminator evaluates the candidate building design schemes in real time. This process eliminates the manual design bias in the traditional design method and improves the diversity and innovation of the building design. In the scheme evaluation stage, the heat island effect intensity UHI of the urban area to be built after the construction of each candidate building scheme is evaluated. after , building energy consumption E and building cost CB, the system can score each candidate building design scheme, and based on quantitative data, eliminate subjective factors to ensure the optimization of the design scheme. In addition, the scheme screening unit comprehensively considers the intensity of the heat island effect UHIafter , building energy consumption E and building cost CB, and select the plan with the lowest comprehensive score PF among the candidate building design plans to reduce energy consumption in building design, reduce building operation costs, and effectively control the negative impact of urban heat island effect on the urban environment, realizing the intelligent optimization of building design, which not only improves the design efficiency and reduces the risk of human intervention in the design process, but also ensures the optimization of building design plans in multiple dimensions such as function, environmental impact and economic benefits, providing strong support for the sustainable development of future buildings.

[0154] Example 5

[0155] See also Figure 1 , Figure 3 and Figure 4 Specifically: the building facade optimization module is used to optimize the building facade design according to the first building design scheme using a deep reinforcement learning optimization algorithm, and the specific optimization process includes initializing the design, defining the state space, defining the action space, setting the reward function and the training process;

[0156] The initialization design refers to extracting specific contents of the first architectural design scheme according to the first architectural design scheme;

[0157] The defining state space refers to obtaining the building facade state space Z according to the first building design scheme, and the building facade state space Z is specifically expressed as:

[0158] Z = {material reflectivity, green area, window ratio};

[0159] The defined action space refers to optimizing and adjusting according to the first building design scheme, including changing the reflectivity of the facade material, increasing or decreasing the window opening ratio, increasing or decreasing the roof greening area, and obtaining the building facade action space D. The building facade action space D is specifically expressed as:

[0160] D = [increase greening, reduce greening, increase window ratio, reduce window ratio, increase reflectivity, reduce reflectivity];

[0161] The setting of the reward function refers to setting the reward function R for the purpose of minimizing the intensity of the heat island effect and maximizing energy efficiency. The specific form of the reward function R is:

[0162]

[0163] Where, UHI red represents the reduction in the intensity of the heat island effect, E imp represents the increase in energy consumption, CB represents the building cost, and Represents the reduction in the intensity of the heat island effect, UHI red , Energy consumption increase E imp and the weight coefficient of the construction cost CB, where the specific value of the weight coefficient is set by the customer based on actual conditions. and

[0164] The training process refers to obtaining the initial state S of the scheme according to the first architectural design scheme. 0 , where the initial state of the scheme is S 0 Including the reflectivity of the building facade material, the roof greening area and the window ratio, and select any one of the building facade action space D for optimization, and calculate the reward function value R of the updated building design scheme, and use the Q-learning update formula to update the Q value of the state-action pair of the current building design scheme. The Q-learning update formula is:

[0165]

[0166] In the formula, Q new (s t , d t ) represents the updated Q value, Q(s t , d t ) indicates the current state s t Next, select Action d t The expected long-term reward after α represents the learning rate, R t Indicates execution of action d t The immediate reward obtained after γ represents the discount factor. Indicates the next state s t+1 The maximum Q value of all possible actions d';

[0167] The Q value represents the reward of a state-action pair. In the Q-learning algorithm, the Q value is used to evaluate the comprehensive reward expected when taking a certain action in a certain state. The larger the Q value, the better the choice of the state-action pair. Among them, the Q-learning algorithm is a reinforcement learning algorithm used to train intelligent agents to learn how to take the best action strategy through interaction with the environment to maximize long-term returns. The long-term reward of each state-action pair is evaluated by learning the Q value. The Q-learning algorithm guides the intelligent agent to learn the optimal strategy by continuously updating the Q value.

[0168] According to the Q value of the state-action pair of the current architectural design scheme, the state-action pair with the largest Q value is selected to update the architectural design scheme, and any item in the building facade action space D is reselected for optimization. The Q value is iterated multiple times until the Q value no longer changes, and the architectural design scheme at this time is output as the optimal architectural design scheme.

[0169] As follows, the initial state of the scheme is S 0 =[0.3,0.3,0.5], that is, the reflectivity of the facade material is 0.3, the roof greening area is 0.3, and the window ratio is 0.5. In the current state S 0 =[0.3,0.3,0.5], possible actions d t for:

[0170] 1. Increase the reflectivity of the facade;

[0171] 2. Increase the green roof area;

[0172] 3. Increase the window opening ratio;

[0173] Select Actiond t , get the Q value as follows:

[0174] Q(s t , d 1 )=5,Q(s t , d 2 )=6,Q(s t , d 3 )=4;

[0175] Among them, d 1 Indicates increasing the facade reflectivity, corresponding to a Q value of 5, d 2 Indicates increasing the green roof area, corresponding to a Q value of 6, d 3 Indicates increasing the window opening ratio, corresponding to a Q value of 4;

[0176] Select the action d with the largest Q value 2 , that is, increase the roof greening area, when executing action d 2 After that, get the new building design plan and enter the new state s t+1 :

[0177] s t+1 =[0.3,0.6,0.5];

[0178] According to the new state s t+1 , get the new state s t+1The reduction in heat island effect, improvement in energy efficiency, and construction cost of the building design scheme are used to obtain the reward function R. Based on the reward function R, the Q-learning update formula is used to update the Q value of the current state-action pair, and the training process is repeated to continuously update the Q value. The maximum Q value is selected in each state until the Q value no longer changes. The building design scheme at this time is output as the optimal building design scheme.

[0179] In the embodiment, by using a deep reinforcement learning optimization algorithm, the building facade optimization module can automatically adjust the facade parameters in the building design plan, such as reflectivity, roof greening and window ratio, to minimize the intensity of the heat island effect and maximize the energy efficiency of the building. This process can optimize the plan according to real-time building design needs through iterative training and adaptive feedback, and can also ensure the sustainability and efficiency of the design plan. Specifically, the building facade optimization module continuously optimizes the design of the building facade by calculating the reward function R of the building design plan based on the feedback mechanism of the Q-learning algorithm. This optimization process enables the building to reduce the environmental impact while improving It improves energy efficiency, reduces building energy consumption, and effectively controls construction costs. With the continuous adjustment and updating of the design plan, the system can adaptively adjust the facade of the building to cope with possible changes in the urban environment in the future, further improving the comfort and sustainability of the building. In addition, the application of deep reinforcement learning makes this optimization process no longer rely on traditional experience and manual adjustments, but automatically selects the optimal building design plan from a large number of candidate designs through intelligent algorithms, which improves design efficiency, reduces human errors, and improves the overall quality and environmental protection effect of building design. Ultimately, it realizes intelligent and sustainable building design and promotes the construction industry to develop in a more environmentally friendly and energy-saving direction.

[0180] Example 6

[0181] See also Figure 1 , Figure 3 and Figure 4 Specifically: the intelligent feedback module is used to execute the construction project according to the optimal building design plan, and monitor the building-related data in real time during the implementation of the construction project, and feed it back to the artificial intelligence generated content ALGC to dynamically adjust the building design plan.

[0182] In the embodiment, by continuously monitoring the building-related data during the implementation of the construction project and feeding it back to the AIGC system for artificial intelligence content generation, dynamic adjustment of the building design plan is achieved. Specifically, the module continuously evaluates the actual effect of the current building design by collecting and analyzing the environmental data, energy efficiency data and heat island effect changes obtained in real time during the construction process. Through this continuous feedback mechanism, the system can promptly discover potential problems in the design implementation and make adjustments, respond to sudden environmental changes during the construction process, and adjust the building plan according to the actual data of the use phase. As the building is used, the system continuously optimizes the design to ensure long-term comfort and energy-saving effects, improve the building's adaptability and long-term sustainability, and provide the construction industry with an intelligent and dynamically responsive design optimization model.

[0183] Example 7

[0184] Please refer to Figure 2 ,Specifically: A smart building design method based on artificial intelligence ,generated content, comprising the following steps,

[0185] Step 1: deploy a smart sensor group in the urban area to be built, and combine it with the urban climate monitoring system to obtain the building-related data of the urban area to be built, and pre-process the building-related data of the urban area to be tested to construct a building-related data set S;

[0186] Step 2: Based on the building-related data set S, a summary calculation is performed to obtain the urban heat island effect intensity UHI of the area to be built in the city, and a heat island effect intensity prediction model is constructed using the LSTM algorithm to obtain the average change rate of the heat island effect intensity Evaluate the intensity change of the urban heat island effect in the area to be built in the future time point n. If the intensity change of the urban heat island effect in the area to be built is abnormal, the building design scheme generation module is carried out;

[0187] Step 3: Based on the generative adversarial network GAN, several candidate architectural design schemes are obtained, and the comprehensive score PF of the architectural design scheme is obtained, the architectural design scheme is screened, and the first architectural design scheme is output;

[0188] Step 4: Optimize the building facade according to the first building design scheme to obtain the optimal building design scheme;

[0189] Step 5: According to the optimal building design plan, execute the construction project, continuously collect building-related data, output the building-related data to the artificial intelligence generated content AIGC, and optimize the building design plan.

[0190] In the embodiment, by deploying an intelligent sensor group and an urban climate monitoring system, building-related data is collected and preprocessed in real time to provide accurate environmental information and design parameters for subsequent decision-making, and the intensity of the urban heat island effect in the area to be built is predicted through a microclimate prediction module and an LSTM algorithm, and the abnormal state of the heat island effect is evaluated through a rate of change analysis, so that the building design scheme can cope with future environmental changes. After that, a variety of candidate building design schemes are generated using the generative adversarial network GAN technology, and the building design scheme with the smallest comprehensive score PF is screened out through the building design scheme comprehensive score PF, and output as the first building design scheme. This process can automatically generate design schemes that adapt to environmental changes and reduce human design deviations. Subsequently, the building facade is optimized through deep reinforcement learning to further improve the energy efficiency and environmental adaptability of the building design, reduce the building's heat island effect and energy consumption, and optimize energy efficiency and comfort. Finally, through the real-time data collection and adjustment of the intelligent feedback module during the building construction process, the design of the building in the early stage of construction meets expectations, and can also improve the long-term adaptability and sustainability of the building through continuous optimization, improve the building design efficiency, reduce energy consumption, reduce environmental impact, and achieve more green, intelligent and efficient building design and construction.

[0191] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent building design system based on artificial intelligence content generation, characterized by: It includes data acquisition and preprocessing module, microclimate prediction and evaluation module, building design scheme generation module, building facade optimization module and intelligent feedback module; The data acquisition and preprocessing module is used to deploy an intelligent sensor group in the urban area to be built, and in combination with the urban climate monitoring system, obtain the building-related data of the urban area to be built, and preprocess the building-related data of the urban area to be tested to construct a building-related data set S; The microclimate prediction and evaluation module is used to perform summary calculations based on the building-related data set S to obtain the urban heat island effect intensity UHI of the area to be built, and use the LSTM algorithm to build a heat island effect intensity prediction model to obtain the average change rate of the heat island effect intensity. Evaluate the intensity change of the urban heat island effect in the area to be built in the future time point n. If the intensity change of the urban heat island effect in the area to be built is abnormal, the building design scheme generation module is carried out; The architectural design scheme generation module is used to obtain a number of candidate architectural design schemes based on a generative adversarial network GAN, obtain a comprehensive score PF of the architectural design scheme, screen the architectural design schemes, and output a first architectural design scheme; The building facade optimization module is used to optimize the building facade according to the first building design scheme to obtain the optimal building design scheme; The intelligent feedback module is used to execute the construction project according to the optimal building design plan, and continuously collect building-related data, output the building-related data to the artificial intelligence generated content AIGC, and optimize the building design plan.

2. The intelligent building design system based on artificial intelligence content generation according to claim 1, characterized in that: The data acquisition and preprocessing module includes a data acquisition unit and a preprocessing unit; The data acquisition unit is used to deploy an intelligent sensor group in the urban area to be built, and collect building-related data in the urban area to be built in combination with the urban climate monitoring system, meteorological station and remote sensing satellite, wherein the building-related data includes climate data, building design data and ground reflection data; The climate data refers to the temperature T of the area to be built in the city d and the temperature T outside the urban area to be built urban ; The architectural design data refers to building facade material parameters, building form and window types; The ground reflection data refers to the solar radiation reflectivity β; The preprocessing unit is used to preprocess the collected climate data, building design data and ground reflection data, wherein the preprocessing includes data cleaning, denoising and data standardization, and constructs a building-related data set S based on the preprocessed climate data, building design data and ground reflection data of the urban area to be built.

3. The intelligent building design system based on artificial intelligence content generation according to claim 2, characterized in that: The microclimate prediction and assessment module includes a heat island effect intensity prediction unit and an assessment unit; The heat island effect intensity prediction unit is used to perform summary calculation based on the building-related data set S to obtain the heat island effect intensity UHI of the urban area to be built. The heat island effect intensity UHI is obtained in the following manner: UHI=T urban，avg -T d，avg ; Among them, T urban，avg represents the average temperature of the area outside the urban area to be built, T d，avg Indicates the average temperature of the area to be built in the city; Based on the LSTM algorithm, a heat island effect intensity prediction model is constructed, and relevant data of historical buildings are collected. The relevant data of historical buildings are constructed as a training set, which is input into the heat island effect intensity prediction model for model training. The mean square error method is used to optimize the heat island effect intensity prediction model parameters, and the trained heat island effect intensity prediction model is used to obtain the heat island effect intensity UHI (t n ).

4. The intelligent building design system based on artificial intelligence content generation according to claim 3, characterized in that: The evaluation unit is used to evaluate the heat island effect intensity UHI (t n ), construct a heat island effect intensity time dataset M, wherein the heat island effect intensity time dataset M is specifically expressed in the form of: M=[COVER(t1),COVER(t2),COVER(t3),...,COVER(t n )]; In the formula, UHI(t n ) represents the predicted time point t n is the intensity of the urban heat island effect in the area to be built at that time, and n represents the total length of the predicted time period; According to the heat island effect intensity time data set M, the average change rate of heat island effect intensity is obtained The average change rate of the intensity of the heat island effect The method of obtaining is: In the formula, UHI(t i+1 ) represents the predicted time point t i+1 The intensity of the urban heat island effect in the area to be built at that time, i = [1, 2, 3, ..., n]; Preset heat island effect intensity change threshold The average change rate of the heat island effect is obtained Threshold of heat island effect intensity change Comparative analysis is conducted to evaluate the change in the intensity of the urban heat island effect in the urban area to be built at the future time point n. The specific evaluation contents are as follows: If the average change rate of the heat island effect intensity Greater than the heat island effect intensity change threshold Right now It is determined that the intensity change of the urban heat island effect in the area to be built is in an abnormal state, and the building design plan is generated at this time; If the average change rate of the heat island effect intensity Less than or equal to the heat island effect intensity change threshold Right now It is determined that the intensity change of the urban heat island effect in the area to be built is in a normal state and no treatment is required.

5. The intelligent building design system based on artificial intelligence content generation according to claim 4, characterized in that: The architectural design scheme generation module includes a scheme generation unit, a scheme analysis unit and a scheme screening unit; The scheme generating unit is used to determine the building design goal according to the building design data in the building related data set S, wherein the building design data includes building design parameters and building constraints, and the building design goal includes energy consumption and heat island effect intensity optimization; Based on the generative adversarial network GAN, combined with the architectural design data and the architectural design goals, several candidate architectural design schemes are generated and formed into a set of candidate architectural design schemes H; The candidate building design scheme generation process is as follows: The collected architectural design data and architectural design goals are input into the generative adversarial network (GAN) generator. The generative adversarial network (GAN) receives the architectural design data and architectural design goals, and randomly generates a number of candidate architectural design schemes, wherein the specific contents of the candidate architectural design schemes include building type, building area, number of floors and functional zoning.

6. The intelligent building design system based on artificial intelligence content generation according to claim 5, characterized in that: The scheme analysis unit is used to extract building-related data from each candidate building design scheme in the candidate building design scheme set H, perform summary calculations, and obtain the heat island effect intensity UHI of the urban area to be built after the building is built. after , building energy consumption E and building cost CB; The intensity of the urban heat island effect UHI in the area to be built after the construction of the building after The method of obtaining is: In the formula, β represents the solar radiation reflectivity, T urban Indicates the temperature of the area outside the city to be built, T d represents the temperature of the urban area to be built, f building Represents the building form influence coefficient, V building Represents the volume of the building. Indicates the greening coefficient, A rool，green Indicates the green area on the top of the building, A wall，green Indicates the green area on the building wall; The building energy consumption E is obtained in the following way: E=f material A wall +f window A window ; In the formula, f material represents the thermal conductivity of building materials, f window Indicates the thermal conductivity of the window material, A wall Represents the building wall area, A window Indicates the building window area; The construction cost CB is obtained in the following way: Where V j represents the volume of the building in the jth part, C material，j represents the unit material cost of the jth part of the building, m represents the total number of building parts, C labor represents the daily labor cost, T construction represents the construction period, C equipment represents the equipment usage cost, j = [1, 2, 3, …, m].

7. The intelligent building design system based on artificial intelligence content generation according to claim 6, characterized in that: The scheme screening unit is used to screen the heat island effect intensity UHI of the urban area to be built after the building is built. after , building energy consumption E and building cost CB, and perform summary calculation to obtain the comprehensive score PF of the building design scheme. The comprehensive score PF of the building design scheme is obtained as follows: PF=ω1·UHI after +ω2·E+ω3·CB; Where ω1, ω2 and ω3 represent the urban heat island effect intensity UHI of the unbuilt area after the construction of the building after , weight coefficients of building energy consumption E and building cost CB; According to the candidate architectural design scheme set H, the architectural design scheme comprehensive score PF of each candidate architectural design scheme is obtained, and compared, and the candidate architectural design scheme with the smallest architectural design scheme comprehensive score PF is selected and output as the first architectural design scheme.

8. The intelligent building design system based on artificial intelligence content generation according to claim 7, characterized in that: The building facade optimization module is used to optimize the building facade design according to the first building design scheme using a deep reinforcement learning optimization algorithm, wherein the specific optimization process includes initializing the design, defining the state space, defining the action space, setting the reward function and the training process; The initialization design refers to extracting specific contents of the first architectural design scheme according to the first architectural design scheme; The defining state space refers to obtaining the building facade state space Z according to the first building design scheme, and the building facade state space Z is specifically expressed as: Z = {material reflectivity, green area, window ratio}; The defining action space refers to optimizing and adjusting according to the first building design scheme, including changing the reflectivity of the facade material, increasing or decreasing the window opening ratio, increasing or decreasing the roof greening area, and obtaining the building facade action space D. The building facade action space D is specifically expressed as follows: D = [increase greening, reduce greening, increase window ratio, reduce window ratio, increase reflectivity, reduce reflectivity]; The setting of the reward function refers to setting the reward function R for the purpose of minimizing the intensity of the heat island effect and maximizing energy efficiency. The specific form of the reward function R is: Where, UHI red represents the reduction in the intensity of the heat island effect, E imp represents the increase in energy consumption, CB represents the building cost, and Represents the reduction in the intensity of the heat island effect, UHI red , Energy consumption increase E imp and the weight coefficient of the building cost CB; The training process refers to obtaining the initial state S0 of the scheme according to the first building design scheme, wherein the initial state S0 of the scheme includes the reflectivity of the building facade material, the roof greening area and the window ratio, and selecting any one of the building facade action space D for optimization, calculating the reward function value R of the optimized building design scheme, and using the Q-learning update formula to update the Q value of the state-action pair of the current building design scheme, wherein the Q-learning update formula is: In the formula, Q new (s t , d t ) represents the updated Q value, Q(s t , d t ) indicates the current state s t Next, select Action d t The expected long-term reward after α represents the learning rate, R t Indicates execution of action d t The immediate reward obtained after γ represents the discount factor. Indicates the next state s t+1 The maximum Q value of all possible actions d′; According to the Q value of the state-action pair of the current architectural design scheme, the state-action pair with the largest Q value is selected to update the architectural design scheme, and any item in the building facade action space D is reselected for optimization. The Q value is iterated multiple times until the Q value no longer changes, and the architectural design scheme at this time is output as the optimal architectural design scheme.

9. The intelligent building design system based on artificial intelligence content generation according to claim 8, characterized in that: The intelligent feedback module is used to execute the construction project according to the optimal building design plan, and monitor the building-related data in real time during the implementation of the construction project, and feed back to the artificial intelligence generated content ALGC to dynamically adjust the building design plan.

10. An intelligent building design method based on artificial intelligence generated content, used to implement an intelligent building design system based on artificial intelligence generated content as described in any one of claims 1 to 9 above, characterized in that: The following steps are included: Step 1: deploy a smart sensor group in the urban area to be built, and combine it with the urban climate monitoring system to obtain the building-related data of the urban area to be built, and pre-process the building-related data of the urban area to be tested to construct a building-related data set S; Step 2: Based on the building-related data set S, a summary calculation is performed to obtain the urban heat island effect intensity UHI of the area to be built in the city, and a heat island effect intensity prediction model is constructed using the LSTM algorithm to obtain the average change rate of the heat island effect intensity Evaluate the intensity change of the urban heat island effect in the area to be built in the future time point n. If the intensity change of the urban heat island effect in the area to be built is abnormal, the building design scheme generation module is carried out; Step 3: Based on the generative adversarial network GAN, several candidate architectural design schemes are obtained, and the comprehensive score PF of the architectural design scheme is obtained, the architectural design scheme is screened, and the first architectural design scheme is output; Step 4: Optimize the building facade according to the first building design scheme to obtain the optimal building design scheme; Step 5: According to the optimal building design plan, execute the construction project, continuously collect building-related data, output the building-related data to the artificial intelligence generated content AIGC, and optimize the building design plan.

Images