Building curtain wall design optimization method based on BIM technology

Abstract

The invention relates to the technical field of building engineering, and discloses a building curtain wall design optimization method based on a BIM technology, and the method comprises the following steps: S1, obtaining and feedback: collecting climate data through a sensor, and carrying out the linkage with a BIM model; s2, building a BIM model: building the model on a BIM platform; s3, modeling: defining parameters based on the BIM model; s4, intelligent algorithm optimization: integrating energy efficiency and structure analysis tools, and optimizing energy efficiency and structure performance in real time; s5, design scheme optimization comparison: generating and evaluating a plurality of design schemes; s6, simulation and optimization: performing 4D construction simulation by using a BIM platform; and S7, real-time maintenance: carrying out real-time monitoring and maintenance. By integrating dynamic climate data, a BIM model and an intelligent optimization algorithm, multi-objective optimization of a building curtain wall system is realized, and the method can adaptively adjust design parameters, dynamically optimize energy efficiency according to a real-time environment, and enhance structural performances such as wind pressure resistance and seismic resistance.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of building engineering, in particular to a building curtain wall design optimization method based on BIM technology. BACKGROUND

[0002] In the modern construction field, building curtain walls are not only an important part of building appearance design, but also directly affect the energy efficiency, structural performance and sustainability of buildings. With the increasing demand for environmental protection and energy saving, traditional curtain wall design and management methods have exposed some key problems and challenges. Curtain wall design not only needs to consider visual aesthetics, but also must optimize energy efficiency, structural safety and life cycle cost. However, existing technologies mostly rely on static design parameters, which cannot dynamically adapt to complex and changing climate conditions and building environment requirements.

[0003] Firstly, the current curtain wall design method mainly relies on the experience of designers and static calculation models, which makes it difficult for the design scheme to adapt to the dynamic climate change in the region where the building is located. In particular, in terms of energy efficiency optimization, traditional design usually cannot flexibly adjust the parameters of the curtain wall system, such as glass light transmittance, sunshade device angle, etc., resulting in poor energy efficiency performance of the building under different seasons or weather conditions. At the same time, the existing curtain wall design is relatively lagging behind in the optimization of structural performance such as wind pressure resistance and earthquake resistance, which cannot effectively cope with extreme climate conditions, increasing the risk of structural failure.

[0004] Secondly, in the construction stage, existing building projects usually rely on manual experience and traditional construction management tools for task planning and resource scheduling. Due to the complexity of the construction environment and the involvement of multiple parties, traditional methods are prone to resource conflicts, construction delays and low resource utilization efficiency. The fixed construction sequence and scheduling method lack flexibility and are difficult to dynamically adjust according to actual conditions, which not only prolongs the construction period, but also increases the construction cost. In addition, unpredictable factors in the construction process, such as weather changes and site restrictions, are also difficult to effectively cope with through traditional construction plans.

[0005] Finally, the operation and maintenance management of building curtain wall systems have an important impact on the overall energy efficiency and service life of buildings. However, traditional curtain wall maintenance management methods usually rely on regular inspections and manual maintenance plans. This method not only consumes time and effort, but also easily leads to a decline in the operating efficiency of the curtain wall system. The fixed maintenance period often fails to match the actual operating state of the system, resulting in frequent maintenance needs and unnecessary operating costs. At the same time, the inability to predict potential failures of the curtain wall system increases the risk of sudden problems and damage, affecting the long-term stability of the system. SUMMARY

[0006] In view of the defects of the prior art, the building curtain wall design optimization method based on BIM technology is provided, which solves the problem that the original design is difficult to adapt to the dynamic climate change of the region where the building is located and cannot effectively cope with extreme climate conditions.

[0007] To achieve the above object, the following technical scheme is adopted: a building curtain wall design optimization method based on BIM technology, comprising the following steps:

[0008] S1, acquisition and real-time feedback of dynamic climate data: through the integration of various sensor devices, real-time climate data around the building is acquired, and the data is linked with the BIM model to generate dynamic thermal and optical performance feedback of the curtain wall system;

[0009] S2, BIM model construction and multi-dimensional dynamic data integration: a three-dimensional geometric model and material attribute model of the building curtain wall are constructed on the BIM platform;

[0010] S3, multi-objective parameterized modeling and adaptive optimization: based on the BIM model, a parameterized modeling tool is used to define the dynamic parameters of the curtain wall system, and the parameters are automatically adjusted according to the dynamic climate data;

[0011] S4, energy efficiency and structural performance optimization driven by intelligent optimization algorithm: the BIM model is integrated with energy efficiency analysis tools and structural analysis tools, the energy efficiency, thermal performance, wind pressure resistance and seismic performance of the curtain wall system are optimized in real time through intelligent algorithms, and the design parameters are automatically adjusted according to the external dynamic climate data;

[0012] S5, multi-objective optimization and multi-scheme design comparative analysis: based on climate data and energy efficiency requirements, multiple curtain wall design schemes are generated, multi-objective optimization algorithms are used to comprehensively analyze the energy efficiency, structural performance and life cycle cost of the schemes, and the performance of different schemes is compared and evaluated;

[0013] S6, construction simulation and construction optimization based on construction scene simulation: 4D construction simulation is carried out based on the optimal design scheme in the BIM platform, resource scheduling, construction sequence and construction feasibility in the construction scene are analyzed through virtual construction technology, and construction plan is generated by combining AI optimization algorithm;

[0014] S7, adaptive regulation and maintenance in operation stage based on big data and intelligent algorithm: an intelligent sensor system is integrated, the running state of the curtain wall system is monitored in real time through big data analysis and machine learning algorithm, climate, energy consumption and system performance data are analyzed, the curtain wall system is adaptively adjusted, and predictive maintenance plan is provided.

[0015] Preferably, in the S1 step, the climate data includes solar radiation intensity, wind speed, wind direction, air humidity and temperature data, and the plurality of sensor devices are arranged in an array.

[0016] Preferably, in the S2 step, the BIM model contains the geometric shape and material information of the curtain wall, and further integrates the climate data, energy efficiency analysis data and structural performance data obtained from the sensors in real time to form a parameterized environment based on dynamic data.

[0017] Preferably, in the S3 step, the parameterized modeling tool is based on a hybrid intelligent optimization algorithm of genetic algorithm and artificial neural network, and the formula is:

[0018] x optimal =argmax(f ANN (GA(x i )))

[0019]

[0020] x optimal =argmax(f ANN (x i ))

[0021] wherein x i is an individual design parameter, including multi-dimensional design parameters of the curtain wall system, a is a crossover rate in the genetic algorithm, determining the mixing ratio of parent gene combination, β is a mutation rate in the genetic algorithm, δ is a random disturbance term added in the mutation process, f ANN (x i ) is an individual fitness predicted by the artificial neural network, representing the performance of the design parameters, σ(.) is an activation function, commonly ReLU or Sigmoid, w k is the weight of the neural network, determining the influence of each input parameter on the output result, and b is the bias term of the neural network, adjusting the network output.

[0022] The parameters include the total solar transmittance of the glass material, the adjustment angle of the shading device, and the heat transfer coefficient of the curtain wall material.

[0023] Preferably, in the S4 step, the intelligent algorithm adopts a particle swarm algorithm, and the formula is:

[0024]

[0025] x i (t+1)=x i (t)+v i (t+1)

[0026] wherein v iLet x be the velocity of the i-th particle in the t-th iteration. i (t) represents the position of the i-th particle in the t-th iteration. For the historical best position of the i-th particle, g best The global optimal position is represented by ω, the inertia weight controls the search speed, c1 and c2 are acceleration factors, and r1 and r2 are random numbers used to introduce randomness.

[0027] Preferably, in step S5, the design scheme generation includes the following steps:

[0028] (1) Determining input variables: First, determine the input variables, which will affect the performance of the curtain wall design;

[0029] The input variables include: the light transmittance of the glass material, the angle of the shading device, the heat transfer coefficient of the curtain wall material, structural design parameters, and climate data;

[0030] (2) Combining climate data with energy efficiency requirements: Obtain the annual average solar radiation, wind speed, temperature and humidity of the area where the building project is located from sensors or meteorological databases, and then determine the building's target energy consumption limits, lighting requirements and thermal comfort requirements.

[0031] (3) Determination of multi-objective design variables: For different input parameters of each design scheme, the light transmittance of the glass can be changed from low to high, the angle of the shading device can be adjusted, and the heat transfer coefficient of the material can be selected from different material combinations;

[0032] (4) Generate multiple design schemes: Use a genetic algorithm or other optimization algorithm to randomly initialize a set of design parameters:

[0033] P(0)={x1(0),x2(0),…,x n (0)}

[0034] Wherein, P(0) is the initial design population, and each x1(0) represents a design scheme;

[0035] (5) Iterative generation of new schemes: Through crossover and mutation operations in the genetic algorithm, new design schemes are continuously generated.

[0036] Preferably, in step S5, the multi-objective algorithm formula is:

[0037] f total (x i )=w1·f energy (x i )+w2·f structure (x i )+w3·f cost (x i )

[0038] f energy (x i )=E cooling (x i )+E heating (x i )+E lighting (x i )

[0039]

[0040] f cost (x i )=C initial (x i )+C maintenance (x i )+C operational (x i )

[0041] wherein f total (x i ) is the total fitness of the i-th design scheme; w1, w2, w3 are the weights of different optimization objectives; E cooling (x i ) is the refrigeration energy consumption, E heating (x i ) is the heating energy consumption, E lighting (x i ) is the lighting energy consumption, F fail (x i ) is the maximum bearing capacity of the curtain wall design scheme, F applied (x i ) is the actual applied load, C initial (x i ) is the initial material and construction cost of the design scheme, C maintenance (x i ) is the maintenance cost of the design scheme, C operational (x i ) is the operation energy consumption cost of the design scheme.

[0042] Preferably, in the S6 step, the 4D function is used for the installation sequence, progress control and construction period arrangement of the curtain wall system.

[0043] Preferably, in the S6 step, the construction plan is generated in combination with the AI optimization algorithm, specifically comprising the following steps:

[0044] a. Input construction-related data:

[0045] Construction resource data: including the quantity, available time and cost of workers, materials, equipment and other resources;

[0046] Building BIM model: providing geometric information, component information, and physical properties of construction objects;

[0047] Construction sequence constraints: including legal construction steps, technical specifications, physical limitations of construction site, etc.;

[0048] Time and duration requirements: total duration requirements of the project, and its phased milestone targets;

[0049] Weather and environmental conditions: especially the weather conditions that affect the construction progress;

[0050] b. Select AI optimization algorithm: using particle swarm optimization algorithm to find the balance between time and resource cost by simulating the movement of particles in the search space, which is suitable for construction task sequencing optimization;

[0051] c. Model input and settings:

[0052] Minimize total duration: calculate the optimal sequencing of construction tasks through AI algorithm to achieve the shortest construction period;

[0053] Minimize construction cost: optimize the scheduling of construction resources (such as workers, equipment, materials) to reduce construction cost under the premise of meeting construction progress;

[0054] Avoid resource conflicts: avoid the demand for the same resources by multiple trades or construction steps at the same time through reasonable resource allocation and time table setting;

[0055] Minimize risk: adjust the arrangement of construction steps to avoid weather risks or construction conflicts according to weather forecast data and construction site risk prediction;

[0056] d. After the above steps are revised, the final scheme is produced after comprehensive evaluation.

[0057] Preferably, in the S7 step, the big data analysis obtains data sources through a sensor array, and transmits them to a big data platform for cleaning and analysis through the Internet of Things, and the machine learning algorithm adopts a supervised learning algorithm, specifically including:

[0058] Linear regression: it is used to predict the trend of building energy consumption with climate change, and a simple linear relationship can be used to model the response of the internal temperature of the building to the change of external temperature;

[0059] Decision tree: it is used for performance prediction under the interaction of various climate factors, and the decision tree can generate a rule set with high precision to predict the behavior of the curtain wall system under specific climate conditions;

[0060] Support Vector Machine: It is used for energy efficiency prediction and optimization of multi-dimensional features, such as predicting building energy efficiency performance by monitoring multiple factors such as light, wind speed, temperature and humidity, etc.

[0061] The application provides a building curtain wall design optimization method based on BIM technology. The method has the following advantages:

[0062] 1. The application realizes multi-objective optimization of building curtain wall system by integrating dynamic climate data, BIM model and intelligent optimization algorithm. The method can adaptively adjust curtain wall design parameters, dynamically optimize energy efficiency performance according to real-time environmental data, and enhance the structural performance of wind pressure resistance, earthquake resistance and other systems, to ensure that the curtain wall system has the best energy efficiency and safety under different climate conditions, so as to improve energy efficiency and significantly optimize structural performance.

[0063] 2. The application combines BIM technology and AI optimization algorithm to automatically generate and optimize construction plans during the construction phase, effectively shortening the construction period, reducing resource conflicts, and improving resource utilization. Through 4D construction simulation and particle swarm optimization algorithm, the construction task scheduling and resource scheduling are optimized, the total construction period is shortened, and the construction cost is also reduced, to ensure the smooth progress of the construction process and efficient allocation of resources.

[0064] 3. The application combines big data analysis and machine learning algorithm to manage the whole life cycle of the curtain wall system. By monitoring the running state of the system in real time, using machine learning model to predict the maintenance requirements and energy efficiency changes of the system, the predictive maintenance plan can be generated in advance, the unnecessary maintenance frequency and sudden failure can be reduced, and the system operation is more stable and reliable, so that the maintenance requirements are reduced, the life cycle cost is significantly reduced, and the long-term operation cost of the system is optimized. BRIEF DESCRIPTION OF DRAWINGS

[0065] Figure 1 The method flowchart of the application. DETAILED DESCRIPTION

[0066] The technical solutions in the embodiments of the application will be described in detail below with reference to the drawings of the application specification. Obviously, the described embodiments are only a part of the embodiments of the application, not all embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the application.

[0067] Example 1: Intelligent climate response curtain wall system design based on BIM model

[0068] Implementation steps:

[0069] Dynamic Climate Data Acquisition and Feedback: Utilize various sensors around the building to acquire real-time climate data, including solar radiation, wind speed, temperature, humidity, etc. The data is transmitted to the BIM model through the Internet of Things, forming a dynamic feedback loop;

[0070] BIM Model Construction and Data Integration: Construct a three-dimensional geometric model of the curtain wall on the BIM platform, integrate real-time climate data and energy efficiency analysis data, and generate a dynamic parameterized environment for the curtain wall system;

[0071] Multi-objective Parameterized Modeling and Adjustment: Use an optimization algorithm combining genetic algorithms and artificial neural networks to generate parameterized models and adaptively adjust glass light transmittance, shading angle, material heat transfer coefficient, etc. according to climate data;

[0072] Energy Efficiency and Structural Performance Optimization: Optimize the energy efficiency and thermal performance of the curtain wall system through particle swarm optimization, automatically adjust design parameters based on dynamic climate data, and achieve optimized design with high energy efficiency and structural safety;

[0073] Adaptive Adjustment and Feedback: Real-time monitoring of curtain wall operation status based on big data analysis, automatic adjustment of curtain wall design through machine learning algorithms (such as support vector machines, linear regression), and improvement of energy efficiency.

[0074] Summary of the embodiment:

[0075] This embodiment combines BIM technology with real-time dynamic climate data, based on hybrid optimization of genetic algorithms and artificial neural networks, to achieve adaptive design and adjustment of curtain wall systems, maintaining optimal energy efficiency and structural performance in changing climate environments.

[0076] Example 2: Building Curtain Wall Energy Efficiency Design Scheme Generation Based on Multi-objective Optimization

[0077] Implementation steps:

[0078] Climate Data and Energy Efficiency Demand Analysis: Obtain annual climate data from sensors or meteorological databases, and combine with building energy efficiency requirements to define input variables such as glass light transmittance, shading device angle, etc.

[0079] BIM Model Construction: Construct a geometric model on the BIM platform and generate an energy efficiency demand analysis model for the curtain wall based on sensor data;

[0080] Multi-objective Design Variable Setting: Set multiple design schemes through genetic algorithms, combining different design parameters to generate multiple curtain wall design schemes;

[0081] Multi-objective Optimization and Comparative Analysis: Use MOGA (Multi-objective Genetic Algorithm) to comprehensively analyze the energy efficiency, structural performance, and life cycle cost of each design scheme, and generate multiple balanced schemes;

[0082] Best design option selection: Based on optimization weights and design objectives, select the best curtain wall design option.

[0083] Summary of the embodiment:

[0084] This embodiment balances multi-dimensional data such as energy efficiency, cost, and structural safety through a multi-objective optimization algorithm. It generates multiple design options through intelligent algorithms and compares them to ultimately determine the best solution, achieving multi-objective optimization of curtain wall design.

[0085] Example 3: AI algorithm-based curtain wall system construction plan generation

[0086] Implementation steps:

[0087] Construction data input and analysis: Collect construction-related data, including workers, equipment, materials, and weather information, and build geometric models of construction objects through BIM models;

[0088] AI algorithm selection and optimization: Through the particle swarm algorithm, simulate the sequencing and resource allocation of construction tasks, and optimize the scheduling of construction tasks to reduce the construction period and cost;

[0089] 4D construction simulation: Based on the 4D function of the BIM platform, simulate the installation sequence and progress control of the curtain wall system, and optimize resource scheduling through virtual construction technology;

[0090] Construction plan generation and adjustment: According to the simulation results and real-time data, dynamically adjust the construction plan in combination with AI algorithms to avoid resource conflicts and construction delays;

[0091] Construction process monitoring and feedback: Real-time monitoring of the construction process through big data, and AI algorithms adjust the construction progress according to feedback data.

[0092] Summary of the embodiment:

[0093] This embodiment uses AI algorithms to generate construction plans, combined with BIM technology and 4D simulation, to achieve dynamic optimization of construction tasks and efficient scheduling of resources, avoiding resource waste and construction delays in traditional construction processes.

[0094] Example 4: Reinforcement learning-based dynamic adjustment and maintenance of curtain wall systems

[0095] Implementation steps:

[0096] Real-time sensor data collection: Real-time monitoring of temperature, humidity, solar radiation, and other data of the curtain wall system through sensor devices and transmission to the big data platform;

[0097] Reinforcement Learning Algorithm Training: Using the Q-learning algorithm, learn the optimal adjustment strategy under different climate conditions through continuous trial and error, optimize the transmittance, shading angle, and other parameters of the curtain wall;

[0098] Adaptive Adjustment and Feedback: Through real-time data, the reinforcement learning model automatically adjusts the design parameters of the curtain wall to achieve adaptive optimization of energy efficiency and comfort;

[0099] Performance Monitoring and Maintenance Prediction: Based on historical data, machine learning models such as LSTM predict the performance degradation of the curtain wall system and provide maintenance recommendations in advance;

[0100] Maintenance Plan Generation: Based on reinforcement learning feedback, generate an automatic maintenance plan and optimize the maintenance cycle through big data analysis.

[0101] Summary of the Embodiment:

[0102] This embodiment realizes adaptive adjustment and maintenance of the curtain wall system through reinforcement learning, ensuring system stability in a changing environment while automatically generating maintenance plans to improve long-term reliability.

[0103] Embodiment 5: Implementation Steps for Curtain Wall System Full Life Cycle Management Based on Big Data and Machine Learning:

[0104] Data Collection and Cleaning: Collect sensor data, construction data, and historical climate data through a big data platform, and perform data cleaning and preprocessing;

[0105] Machine Learning Model Construction: Based on supervised learning models such as linear regression and support vector machines, analyze the energy efficiency and structural performance of the curtain wall system, and make performance predictions based on these data;

[0106] Life Cycle Cost Analysis: Based on energy consumption, maintenance, and update data of the curtain wall system, conduct full life cycle cost analysis and optimization;

[0107] Maintenance and Adjustment Plan Generation: Based on the prediction results of machine learning, automatically generate maintenance plans and adjust system parameters based on energy efficiency prediction models;

[0108] Feedback and Optimization: Through real-time monitoring feedback, continuously optimize machine learning models to achieve intelligent management of the curtain wall system throughout its life cycle.

[0109] Summary of the Embodiment:

[0110] This embodiment uses big data analysis and machine learning technology to manage the full life cycle of building curtain walls, automatically generating energy efficiency adjustment and maintenance plans to ensure efficient operation and economic optimization of the system throughout its life cycle.

[0111] Summary of Examples:

[0112] These five examples demonstrate how BIM technology, big data analysis, AI optimization algorithms, and machine learning can be used to improve the performance, energy efficiency, and economic efficiency of building curtain wall systems at different stages of the design, optimization, construction, adjustment, and full life cycle management. In the design stage, genetic algorithms and multi-objective optimization algorithms make the curtain wall design more intelligent and flexible. In the construction stage, AI algorithms help to achieve efficient scheduling of resources. In the operation stage, reinforcement learning and big data analysis ensure the adaptive adjustment and maintenance of the system. Finally, these technical means converge into a full life cycle management system to ensure that the curtain wall system is always in the best state during its life cycle.

[0113] These examples not only demonstrate the innovations of the invention, but also further optimize the original scheme by combining various technical means to ensure higher feasibility and superiority in different scenarios.

[0114] Test Experiment:

[0115] Experiment 1: Intelligent Climate-Responsive Curtain Wall System Design Based on BIM Model

[0116] Test Procedure:

[0117] Experiment Goal: Test the adaptive adjustment capability of the curtain wall system under different climate conditions, mainly test the energy efficiency performance (such as cooling and heating energy consumption), indoor temperature change and curtain wall material performance change.

[0118] Experiment Steps:

[0119] Arrange sensors in the experimental building to monitor external solar radiation, temperature, wind speed and humidity in real time;

[0120] Use BIM model to establish curtain wall design and adjust glass light transmittance, sunshade device angle, etc. through AI algorithm;

[0121] Run the system under different climate conditions (sunny, cloudy, high wind speed, etc.), record the adaptive adjustment times and energy efficiency performance of the curtain wall system;

[0122] Compare the experimental results with the traditional curtain wall system with fixed sunshade angle and light transmittance.

[0123] Experiment Data:

[0124]

[0125] Test Results:

[0126] Energy Efficiency Improvement: The BIM-based intelligent climate-responsive curtain wall system can improve energy efficiency by 10-15% compared to traditional solutions. Through adaptive regulation, indoor temperature remains more stable, reducing HVAC energy consumption.

[0127] Experiment 2: Building Curtain Wall Energy Efficiency Design Scheme Generation Based on Multi-objective Optimization

[0128] Test Process:

[0129] Experiment Goal: Generate multiple curtain wall design schemes through multi-objective optimization algorithms and test their performance in energy efficiency, structural performance, and life cycle cost.

[0130] Experiment Steps:

[0131] Generate multiple target design variable combinations based on climate data, energy efficiency requirements, and material performance.

[0132] Use genetic algorithms to generate multiple design schemes and use MOGA algorithms to evaluate these schemes.

[0133] Measure the energy efficiency, structural performance (wind pressure resistance, seismic performance), and life cycle cost of each design scheme.

[0134] Compare the fitness of multiple design schemes and select the optimal scheme.

[0135] Experiment Data:

[0136]

[0137] Test Results:

[0138] Improved Comprehensive Fitness: Through multi-objective optimization algorithms, the comprehensive fitness of the optimal design scheme reaches 89, significantly better than other design schemes, and ensures the balance between energy efficiency and structural performance.

[0139] Experiment 3: Curtain Wall System Construction Plan Generation Based on AI Algorithms

[0140] Test Process:

[0141] Experiment Goal: Generate construction plans through AI algorithms and test their performance in resource scheduling, construction conflict avoidance, and project duration reduction.

[0142] Experiment Steps:

[0143] Input construction resource data, construction sequence constraints, weather and environmental conditions, and other information.

[0144] Use particle swarm optimization algorithms to generate multiple construction plans and compare the project duration and resource allocation of different plans.

[0145] Test the conflict avoidance and resource scheduling during construction.

[0146] Compare the duration difference between the optimized construction plan and the traditional manual construction plan.

[0147] Experimental data:

[0148]

[0149] Test results:

[0150] Duration reduction and cost decrease: The construction plan generated by AI shortened the duration by about 14% and reduced the construction cost by 20%, while avoiding resource conflicts. Resource utilization also significantly improved.

[0151] Experiment 4: Dynamic adjustment and maintenance of curtain wall system based on reinforcement learning

[0152] Test process:

[0153] Experimental goal: Test the effect of reinforcement learning algorithm on real-time adjustment and maintenance of curtain wall system, especially the stability of system performance under uncertain climate conditions.

[0154] Experimental steps:

[0155] Use Q-learning algorithm to train the adjustment strategy of curtain wall system, continuously adjust parameters (such as shading angle, light transmittance) under different climate conditions.

[0156] Through real-time sensor monitoring system state (energy consumption, structural stress, etc.), test the adaptive ability of reinforcement learning algorithm.

[0157] Measure the energy efficiency performance and maintenance requirements of the system, and compare with traditional adjustment methods.

[0158] Experimental data:

[0159]

[0160] Test results:

[0161] System stability and energy efficiency improvement: Reinforcement learning algorithm significantly improves the energy efficiency performance of the system, while greatly reducing the maintenance frequency and failure rate, making the system run more efficiently and reliably.

[0162] Experiment 5: Whole life cycle management of curtain wall system based on big data and machine learning

[0163] Test process:

[0164] Experimental goal: Test the effect of whole life cycle management through big data and machine learning algorithm, especially the life cycle cost optimization and system operation reliability.

[0165] Experimental Steps:

[0166] Collect sensor data, energy consumption data, and maintenance records of building curtain wall systems, preprocess and analyze the data.

[0167] Predict the energy efficiency performance and maintenance needs of the system through machine learning models, and automatically generate maintenance plans.

[0168] Test the changes in the life cycle cost of the system and compare it with the traditional maintenance method without prediction.

[0169] Experimental Data:

[0170]

[0171] Test Results:

[0172] Life Cycle Cost Optimization: Through the whole life cycle management of big data and machine learning, the total cost of the system is reduced by 28%, the energy efficiency is improved by 15%, the maintenance cost is reduced, and the failure rate is greatly reduced.

[0173] Summary

[0174] Through these 5 experiments, it can be seen that various embodiments of the invention have shown significant superiority in different application scenarios, especially in terms of energy efficiency improvement, construction efficiency optimization, system regulation stability and life cycle cost optimization:

[0175] Experiment 1 verifies that through BIM and adaptive regulation, the energy efficiency of the curtain wall system can be effectively improved, and the energy consumption can be reduced;

[0176] Experiment 2 generates design schemes through multi-objective optimization algorithm, which can balance energy efficiency, structural performance and cost, and select the best design scheme;

[0177] Experiment 3 shows the optimization effect of AI algorithm on construction plan, which is superior to traditional method in terms of construction period and resource scheduling;

[0178] Experiment 4 realizes intelligent regulation and efficient maintenance of the system through reinforcement learning, enhances the stability of system operation;

[0179] Experiment 5 realizes the whole life cycle management of big data and machine learning, significantly reduces the life cycle cost of the system, and improves the reliability of the system.

[0180] Comparison Experiment:

[0181] Comparison Experiment 1: Curtain wall design scheme generation based on BIM and AI optimization vs. traditional manual design scheme generation

[0182] Experimental Objectives:

[0183] Compare the differences in energy efficiency, structural performance, and life cycle cost between the curtain wall design generated by BIM and AI optimization algorithm and the traditional manual design, to verify the advantages of AI optimization algorithm and BIM technology.

[0184] Prior art solutions:

[0185] Designers generate curtain wall design based on experience and traditional calculation tools such as Excel and AutoCAD;

[0186] Curtain wall system design is based on static climate data and does not dynamically adjust design parameters, mainly considering the worst-case scenario during design;

[0187] Design evaluation relies on simple calculations of building energy consumption and materials, lacking detailed analysis of life cycle cost optimization and structural performance.

[0188] Invention solutions:

[0189] Generate design based on BIM and AI optimization:

[0190] Use BIM technology to build a three-dimensional model of the curtain wall system, integrate dynamic climate data such as wind speed, solar radiation, temperature, etc., and automatically generate multi-objective optimization solutions based on actual project requirements;

[0191] Use AI optimization algorithms such as multi-objective genetic algorithm (MOGA) to automatically optimize design, analyze energy efficiency, structural performance, and life cycle cost, etc.

[0192] Dynamically adjust parameters such as glass transmittance, shading device angle, material heat transfer coefficient, etc., to ensure the adaptability of the solution under different climate conditions.

[0193] Comparison experiment process:

[0194] Input conditions:

[0195] Climate conditions: Use the same building project's climate data, including annual average solar radiation, wind speed, temperature, humidity.

[0196] Design goals: Minimize energy consumption, optimize structural performance, and reduce life cycle cost.

[0197] Experimental steps:

[0198] Traditional technology team generates curtain wall design based on static climate data;

[0199] Use the invention to generate multiple design solutions based on BIM technology and AI optimization algorithm, and automatically optimize them using MOGA;

[0200] Compare the performance of each design scheme in energy efficiency, structural performance, material cost, construction cost, life cycle cost, etc.

[0201] Experimental data:

[0202]

[0203]

[0204] Comparative analysis:

[0205] Energy efficiency improvement: BIM and AI optimization scheme saves 20% in annual energy consumption, effectively reduces building energy consumption through adaptive adjustment of design parameters;

[0206] Structural performance: The invention scheme is superior to the traditional scheme in wind pressure resistance and seismic performance, ensuring the stability of the building curtain wall;

[0207] Design efficiency: AI optimization algorithm automatically generates multiple design schemes, reducing design time by about 67%, and providing multiple design options;

[0208] Life cycle cost: The design scheme based on BIM and AI optimization can reduce the life cycle cost of the building by about 17%;

[0209] Comparison experiment 2: BIM-based dynamic construction plan generation vs. traditional manual construction plan

[0210] Experimental objectives:

[0211] Compare the differences between the dynamic construction plan generated based on BIM and AI algorithm and the traditional manual construction plan in terms of construction period, resource scheduling efficiency, construction cost, etc., to verify the advantages of BIM and AI optimization in construction planning.

[0212] Prior art scheme:

[0213] The construction plan is manually arranged by the project manager according to the building drawings and resources;

[0214] Resource scheduling is adjusted manually, lacking real-time optimization mechanism, which is easy to cause resource conflict or construction delay;

[0215] The arrangement of construction sequence is relatively fixed and difficult to adjust in real time according to the actual situation.

[0216] Invention scheme:

[0217] Construction plan based on BIM and AI optimization:

[0218] Integrate building geometric information, construction sequence constraints, resource data and weather information using BIM model, and generate multiple construction plans through particle swarm optimization algorithm;

[0219] AI algorithms can automatically optimize the construction period, resource allocation, and construction sequence, reducing resource conflicts and construction delays.

[0220] Adjust construction plans based on real-time data, dynamically adjust task scheduling and resource allocation, and optimize construction efficiency.

[0221] Comparison experiment process:

[0222] Input conditions:

[0223] Construction resources: same number of workers, materials, and equipment.

[0224] Construction constraints: including weather impact, construction site constraints, time period requirements, etc.

[0225] Experimental steps:

[0226] Traditional construction management team manually arranges construction plans, allocates resources and construction period based on building drawings;

[0227] Use BIM platform combined with AI optimization algorithm to automatically generate construction plan, and optimize construction period and resource scheduling;

[0228] Compare the performance of two construction plans in construction period, resource conflict, resource utilization rate, construction cost and other dimensions.

[0229] Experimental data:

[0230]

[0231] Comparison and analysis:

[0232] Shorten the construction period: BIM and AI optimized construction plan reduces the construction period by about 17%, and significantly reduces the construction delay;

[0233] Resource scheduling efficiency: through dynamic optimization of resource allocation, AI algorithm generated construction plan makes resource utilization rate increase by 20%, avoiding resource conflict;

[0234] Construction cost reduction: due to more efficient resource scheduling and shorter construction period, construction cost is reduced by 10%;

[0235] Efficiency of completing construction tasks: AI optimization makes the efficiency of completing construction tasks increase by 10%, and the more flexible construction adjustment mechanism ensures the smooth execution of construction plan.

[0236] While embodiments of the application have been shown and described, it is to be understood that the embodiments described are merely exemplary of the principles and application of the present application. Numerous modifications and adaptions can be effected without departing from the spirit and scope of the present application, which is not limited to the exact construction and arrangement described. It is intended, therefore, to cover all modifications and adaptions that fall within the scope of the claims and their equivalents.

Claims

1. A method for designing and optimizing a building curtain wall based on BIM technology, characterized in that, The method comprises the following steps: S1, acquisition and real-time feedback of dynamic climate data: by integrating various sensor devices, real-time climate data around the building is acquired, and the data is linked with the BIM model to generate dynamic thermal and optical performance feedback of the curtain wall system; S2, BIM model construction and multi-dimensional dynamic data integration: a three-dimensional geometric model and material attribute model of the building curtain wall are constructed on the BIM platform; S3, multi-objective parameterized modeling and adaptive optimization: based on the BIM model, a parameterized modeling tool is used to define the dynamic parameters of the curtain wall system, and the parameters are automatically adjusted according to the dynamic climate data; S4, energy efficiency and structural performance optimization driven by intelligent optimization algorithm: the BIM model is integrated with energy efficiency analysis tools and structural analysis tools, and the energy efficiency, thermal performance, wind pressure resistance and seismic performance of the curtain wall system are optimized in real time through intelligent algorithms, and the design parameters are automatically adjusted according to the external dynamic climate data; S5, multi-objective optimization and multi-scheme design comparative analysis: based on climate data and energy efficiency requirements, multiple curtain wall design schemes are generated, and multi-objective optimization algorithms are used to comprehensively analyze the energy efficiency, structural performance and life cycle cost of the schemes, and the performance of different schemes is compared and evaluated; S6, construction simulation and construction optimization based on construction scene simulation: 4D construction simulation is carried out based on the optimal design scheme in the BIM platform, resource scheduling, construction sequence and construction feasibility in the construction scene are analyzed through virtual construction technology, and construction plan is generated by combining AI optimization algorithm; S7, adaptive regulation and maintenance in operation stage based on big data and intelligent algorithm: an intelligent sensor system is integrated, the running state of the curtain wall system is monitored in real time through big data analysis and machine learning algorithm, climate, energy consumption and system performance data are analyzed, the curtain wall system is adaptively adjusted, and predictive maintenance plan is provided.

2. The building curtain wall design optimization method based on BIM technology according to claim 1, characterized in that, In the S1 step, the climate data includes solar radiation intensity, wind speed, wind direction, air humidity and temperature data, and the various sensor devices are arranged in an array.

3. The method of claim 1, wherein the method further comprises: In the S2 step, the BIM model includes the geometric shape and material information of the curtain wall, and also integrates real-time climate data, energy efficiency analysis data and structural performance data obtained from sensors to form a parameterized environment based on dynamic data.

4. The building curtain wall design optimization method based on BIM technology according to claim 1, characterized in that, In the S3 step, the parameterized modeling tool is based on a hybrid intelligent optimization algorithm of genetic algorithm and artificial neural network, and the formula is: x optimal = argmax(f ANN (GA(x i ))) x optimal = argmax(fANN(x i )) where x i is the individual design parameter, including the multi-dimensional design parameter of the curtain wall system, a is the crossover rate in the genetic algorithm, which determines the mixing ratio of the parent gene combination, β is the mutation rate in the genetic algorithm, δ is the random disturbance term added in the mutation process, f ANN (x i ) is the individual fitness predicted by the artificial neural network, which represents the performance of the design parameter, σ(.) is the activation function, commonly used are ReLU or Sigmoid, w k is the weight of the neural network, which determines the influence of each input parameter on the output result, and b is the bias term of the neural network, which adjusts the network output. The parameters include the total solar transmittance of glass material, the adjustment angle of shading device, and the heat transfer coefficient of curtain wall material.

5. The method of claim 1, wherein the method further comprises: In the S4 step, the intelligent algorithm uses a particle swarm algorithm, and the formula is: x i (t+1) = x i (t) + v i (t+1) where v i is the velocity of the i-th particle at the t-th iteration, x i (t) is the position of the i-th particle at the t-th iteration, is the historical best position of the i-th particle, g best is the global best position, ω is the inertia weight, controls the search speed, c1 and c2 are acceleration factors, and r1 and r2 are random numbers used to introduce randomness.

6. The method of claim 1, wherein the method further comprises: In the S5 step, the design scheme generation comprises the following steps: (1) input variable determination: first, determine the input variables, which will affect the performance of the curtain wall design; The input variables include the light transmittance of glass material, the angle of shading device, the heat transfer coefficient of curtain wall material, structural design parameters and climate data; (2) Combining climate data with energy efficiency requirements: Obtain the annual average solar radiation, wind speed, temperature, humidity of the region where the building project is located from sensors or meteorological databases, and determine the target energy consumption limit, lighting requirements, and thermal comfort requirements of the building; (3) Determining multi-objective design variables: For different input parameters of each design scheme, the light transmittance of the glass can vary from low to high, the angle of the shading device can be adjusted, and the heat transfer coefficient of the material can be selected from different material combinations; (4) Generating multiple design schemes: Using genetic algorithms or other optimization algorithms, a set of design parameters is randomly initialized: P(0) = {x1(0), x2(0),..., x n (0)} Where P(0) is the initial design population, and each x1(0) represents a design scheme; (5) Iteratively generating new schemes: New design schemes are continuously generated through crossover and mutation operations in genetic algorithms.

7. The method of claim 1, wherein the method further comprises: In the S5 step, the multi-objective algorithm formula is: f total (x i )=w1·f energy (x i )+w2·f structure (x i )+w3·f cost (x i ) f energy (x i ) = E cooling (x i ) + E heating (x i ) + E lighting (x i ) f cost (x i )=C initial (x i )+C maintenance (x i )+C operational (x i ) wherein f total (x i ) is the total fitness of the i-th design scheme; w1, w2, w3 are the weights of different optimization objectives; E cooling (x i ) is the refrigeration energy consumption, E heating (x i ) is the heating energy consumption, E lighting (x i ) is the lighting energy consumption, F fail (x i ) is the maximum bearing capacity of the curtain wall design scheme, F applied (x i ) is the actual applied load, C initial (x i ) is the initial material and construction cost of the design scheme, C maintenance (x i ) is the maintenance cost of the design scheme, C operational (x i ) is the operation energy consumption cost of the design scheme.

8. The building curtain wall design optimization method based on BIM technology according to claim 1, characterized in that, In the S6 step, the 4D function is used for installation sequence, progress control, and schedule arrangement of the curtain wall system.

9. The method of claim 1, wherein the method further comprises: In the S6 step, the construction plan generated by combining the AI optimization algorithm specifically includes the following steps: a. Input construction-related data: Construction resource data: including the quantity, available time, and cost of workers, materials, and equipment resources; Building BIM model: providing geometric information, component information, and physical properties of construction objects; Construction sequence constraints: including legally prescribed construction steps, technical specifications, physical limitations of the construction site, etc.; Time and duration requirements: total duration requirements of the project and its phased milestone targets; Weather and environmental conditions: especially the data input of weather conditions that affect construction progress; b. Select AI optimization algorithm: adopt particle swarm optimization algorithm to simulate the movement of particles in the search space, which can find a balance between time and resource cost, and is suitable for task sequencing optimization; c. Model input and settings: Minimize total duration: calculate the optimal sequencing of construction tasks through AI algorithms to achieve the shortest construction period; Minimize construction cost: optimize the scheduling of construction resources (such as workers, equipment, and materials) to reduce construction costs while meeting construction progress; Avoid resource conflicts: avoid the demand for the same resources by multiple trades or construction steps at the same time through reasonable resource allocation and time table setting; Minimize risk: adjust the arrangement of construction steps to avoid weather risks or construction conflicts based on weather forecast data and construction site risk prediction; d. After the above steps, the final scheme is produced after comprehensive evaluation.

10. The method of claim 1, wherein the method further comprises: In the S7 step, big data analysis obtains data sources through a sensor array, and transmits them to a big data platform for cleaning and analysis. The machine learning algorithm uses a supervised learning algorithm, which specifically includes: Linear regression: used to predict the trend of building energy consumption with climate change, a simple linear relationship can be used to model the response of the internal temperature of the building to external temperature changes; Decision tree: used for performance prediction under the interaction of multiple climate factors, a decision tree can generate a rule set with high precision to predict the behavior of the curtain wall system under specific climate conditions; Support Vector Machine: It is used for energy efficiency prediction and optimization of multi-dimensional features, such as predicting building energy efficiency performance by monitoring light, wind speed, temperature and humidity, etc.

Images