Method and system for optimizing material consumption of curtain wall

Abstract

The invention relates to the technical field of constructional engineering and intelligent optimization, in particular to a curtain wall material consumption optimization method and system.A self-adaptive multi-layer feedback modeling system is adopted to generate accurate three-dimensional model data, a multi-layer feedback analysis system is combined to dynamically adjust real-time construction data, and a feedback optimization instruction is generated; an optimal material distribution scheme is generated through a collaborative optimization algorithm, and the balance between material supply and actual demands is ensured; and finally, generating a material consumption list based on the optimized data, and applying the material consumption list to construction purchase and execution. According to the method, through dynamic adjustment and optimization strategies, material waste is greatly reduced, the material utilization rate is increased, and the construction efficiency is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of construction engineering and intelligent optimization technology, and in particular to a method and system for optimizing the amount of curtain wall materials used. Background Art

[0002] In curtain wall construction, material costs usually account for a considerable proportion of the total cost. Therefore, accurate calculation of material usage can not only effectively control costs, but also avoid material waste and construction delays. The prior art (Chinese invention patent, publication number: CN118133390A, name: A method and system for optimizing the amount of curtain wall materials) mainly relies on building information modeling (BIM) and computer-aided planning software to predict and plan material usage. Although these technologies have improved the accuracy and efficiency of planning to a certain extent, they still have significant deficiencies. BIM technology relies on a large amount of accurate and complete design data. If the design data is inaccurate or incomplete, the material planning results will be seriously affected. In addition, computer-aided planning software often requires high professional knowledge and skills, and is prone to computational redundancy when processing large amounts of data, resulting in increased computational time. These all make the prior art inefficient and inaccurate in practical applications. Summary of the invention

[0003] In view of the many problems existing in the above-mentioned prior art, the present invention provides a method and system for optimizing the usage of curtain wall materials. The present invention uses an adaptive multi-layer feedback modeling system and a collaborative optimization algorithm to accurately predict the demand and optimize the distribution of curtain wall materials. The system uses a multi-level feedback analysis system to dynamically adjust the material distribution strategy through the real-time collected construction progress data and material consumption data to ensure the balance between material supply and actual demand. At the same time, the collaborative optimization algorithm is used to generate the optimal material distribution plan, which effectively reduces material costs and improves construction efficiency. The effect of the present invention is to significantly reduce material waste, optimize material utilization, and ensure efficient and smooth construction.

[0004] A method for optimizing the amount of curtain wall materials, comprising the following steps:

[0005] The geometric shape data and the curtain wall material data of the building curtain wall are collected by using a variety of sensors deployed at different positions of the building curtain wall, and the data are cleaned and noise filtered to generate pre-processed curtain wall point cloud data and pre-processed curtain wall material data;

[0006] Inputting the pre-processed curtain wall point cloud data and the pre-processed curtain wall material data into an adaptive multi-layer feedback modeling system for layered processing, optimizing the model through an inter-layer feedback mechanism, and generating final three-dimensional model data containing complete geometric structure and material information;

[0007] The final three-dimensional model data is combined with environmental data, historical construction data and supply chain data, multi-level prediction is performed through a hierarchical neural network, and optimized material distribution scheme data is generated through a collaborative optimization algorithm, wherein the material distribution scheme data describes the type, quantity, specification and distribution location of the material;

[0008] During the construction process, the construction progress data and actual material usage data are collected in real time, and the data are compared with the optimized material distribution plan data. Feedback optimization instructions are generated through a multi-level feedback analysis system, and the material prediction model and material distribution plan are adaptively adjusted based on the feedback results to generate the final optimized material distribution data;

[0009] A curtain wall material usage list is generated based on the final optimized material distribution data, and material procurement and construction execution are carried out according to the list, and construction data records and project archiving data are generated for subsequent analysis and optimization.

[0010] Preferably, the multiple sensors include laser radar, ultrasonic sensor and high-resolution photogrammetry equipment, the laser radar is used to obtain three-dimensional geometric shape data of the curtain wall, the ultrasonic sensor is used to detect physical property data of the curtain wall material, and the high-resolution photogrammetry equipment is used to capture image data of the curtain wall surface, and the geometric shape data, material data and image data are respectively fused to generate pre-processed curtain wall point cloud data and pre-processed curtain wall material data.

[0011] Preferably, the hierarchical processing in the adaptive multi-layer feedback modeling system includes: dividing the pre-processed curtain wall point cloud data into multiple geometric hierarchical data, each hierarchical data representing geometric details of different precisions, dividing the pre-processed curtain wall material data into multiple material property hierarchical data, each hierarchical data representing different material properties, optimizing the geometric models and material models of each hierarchical level through an inter-layer feedback mechanism, and based on the layer-by-layer optimization and integration of the data of each hierarchical level, finally generating the final three-dimensional model data containing complete geometric structure and material information.

[0012] Preferably, the hierarchical neural network includes a neural network structure of multiple levels. After the final three-dimensional model data is combined with the environmental data, historical construction data and supply chain data, it is first input into the first layer of neural network for preliminary material demand forecasting. The forecast result is passed as input data to the second layer of neural network for supply chain fluctuation forecasting, and is further adjusted in the third layer of neural network in combination with the construction environment data to finally generate material demand forecast data. The material demand forecast data is generated based on the following calculation expression:

[0013]

[0014] Among them, D trepresents the material demand in time period t; M i represents the initial demand for the i-th material; F i represents the adjustment factor based on environmental data, historical construction data and supply chain data; H t Indicates the revised value of material demand based on historical data.

[0015] Preferably, the material demand forecast data is obtained through the output results of a multi-layer neural network and is used to optimize the material supply and use strategy.

[0016] Preferably, the collaborative optimization algorithm uses the material demand forecast data as the initial input, uses the genetic algorithm for preliminary optimization, generates multiple material distribution schemes, and inputs these schemes into the particle swarm optimization algorithm for further optimization to select the scheme with the highest efficiency and the minimum resource consumption, and finally globally optimizes the selected scheme through the simulated annealing algorithm to generate optimized material distribution scheme data, and the material distribution scheme data is described by the following calculation expression:

[0017]

[0018] Among them, S opt represents the optimized material distribution scheme; C j represents the cost of the jth material; P j represents the distribution location priority of the jth material; R k represents the resource utilization rate in the kth time period; T k Represents the resource utilization cost in the kth time period.

[0019] Preferably, the multi-level feedback analysis system monitors and collects construction progress data and actual material consumption data in real time during construction, dynamically compares and analyzes the obtained real-time monitoring data with the pre-formulated optimized material distribution plan, and generates corresponding feedback optimization instructions. The feedback optimization instructions are used to dynamically adjust the material demand prediction model and the material distribution plan according to the actual needs of the construction site to ensure the balance between material supply and actual construction needs. The feedback optimization instructions are generated by the following calculation expression:

[0020] F o =P c -A u

[0021] Among them, F o Indicates feedback optimization instruction; P c Indicates the planned material usage; A u Indicates actual material usage.

[0022] Preferably, the feedback optimization instructions are used to adjust the material distribution strategy in real time during the actual construction process to reduce material waste and improve material utilization, and the adjustment process is achieved through continuous monitoring by a multi-level feedback analysis system.

[0023] Preferably, a curtain wall material usage list is generated based on the final optimized material distribution data, the list including material types, specifications, quantities and specific usage locations. Material procurement and construction execution are carried out according to the list. After the construction is completed, construction data records and project archiving data are generated and used for analysis and optimization of subsequent projects, providing a reference for future material demand forecasting and distribution optimization.

[0024] A system for implementing the method for optimizing the amount of curtain wall materials, the system comprising:

[0025] A multi-sensor module, which is used to be deployed at different positions of the building curtain wall to collect geometric shape data and material data of the curtain wall, and the multi-sensor module includes a laser radar, an ultrasonic sensor and a high-resolution photogrammetry device;

[0026] A data processing module is used to perform data cleaning and noise filtering on the geometric shape data and material data received from the multi-sensor module to generate pre-processed curtain wall point cloud data and pre-processed curtain wall material data;

[0027] An adaptive multi-layer feedback modeling module is used to receive the pre-processed curtain wall point cloud data and the pre-processed curtain wall material data, perform layered processing on them, optimize the model through an inter-layer feedback mechanism, and generate final three-dimensional model data containing complete geometric structure and material information;

[0028] A hierarchical neural network prediction module is used to combine the final three-dimensional model data with environmental data, historical construction data and supply chain data, generate material demand prediction data through multi-level prediction, and generate optimized material distribution plan data through a collaborative optimization algorithm, wherein the material distribution plan data describes the type, quantity, specification and distribution location of the material;

[0029] A multi-level feedback analysis module is used to collect construction progress data and actual material usage data in real time during the construction process, compare and analyze the data with the optimized material distribution plan data, generate feedback optimization instructions, and adaptively adjust the material prediction model and material distribution plan based on the feedback results to generate the final optimized material distribution data;

[0030] The material management module is used to generate a curtain wall material usage list based on the final optimized material distribution data, and to purchase materials and execute construction according to the list, and to generate construction data records and project archiving data for subsequent analysis and optimization.

[0031] Compared with the prior art, the advantages and beneficial effects of the present invention are:

[0032] The present invention realizes the dynamic adjustment of material usage and the generation of optimal distribution strategy by introducing a multi-level feedback analysis system and a collaborative optimization algorithm;

[0033] The present invention collects construction progress and material consumption data in real time through continuous monitoring by a multi-level feedback analysis system, and generates feedback optimization instructions, thereby effectively reducing material waste and improving material utilization;

[0034] The present invention uses a collaborative optimization algorithm to perform multi-stage optimization of material distribution to ensure efficient use of materials at each construction stage;

[0035] The adaptive multi-layer feedback modeling system adopted by the present invention realizes accurate modeling of geometric structure and material information, further improving the accuracy of material usage prediction;

[0036] The above advantages not only make up for the shortcomings of the existing technology, but also further optimize the material management process and improve construction efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a schematic diagram of the process of the present invention;

[0038] Figure 2 It is a flow chart of the adaptive multi-layer feedback modeling system in the present invention;

[0039] Figure 3 This is a flow chart of the collaborative optimization algorithm in the present invention;

[0040] Figure 4 It is a flow chart of the multi-level feedback analysis system in the present invention;

[0041] Figure 5 It is a structural block diagram of the system of the present invention. DETAILED DESCRIPTION

[0042] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present disclosure. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present disclosure. However, it is obvious that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present disclosure.

[0043] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. The terms "comprise", "include", etc. used herein indicate the existence of the features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0044] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification, and should not be interpreted in an idealized or overly rigid manner.

[0045] like Figure 1 As shown, a method for optimizing the amount of curtain wall materials includes the following steps:

[0046] The geometric shape data and the curtain wall material data of the building curtain wall are collected by using a variety of sensors deployed at different positions of the building curtain wall, and the data are cleaned and noise filtered to generate pre-processed curtain wall point cloud data and pre-processed curtain wall material data;

[0047] Preferably, the multiple sensors include laser radar, ultrasonic sensor and high-resolution photogrammetry equipment, the laser radar is used to obtain three-dimensional geometric shape data of the curtain wall, the ultrasonic sensor is used to detect physical property data of the curtain wall material, and the high-resolution photogrammetry equipment is used to capture image data of the curtain wall surface, and the geometric shape data, material data and image data are respectively fused to generate pre-processed curtain wall point cloud data and pre-processed curtain wall material data.

[0048] In the present invention, it is first necessary to accurately obtain the geometric shape and material information of the building curtain wall, which is the basis for all subsequent calculations and optimizations. To this end, a variety of sensors deployed at different locations on the building curtain wall can comprehensively collect the required data. Specifically, these sensors include lidar, ultrasonic sensors, and high-resolution photogrammetry equipment.

[0049] Laser radar is mainly used to obtain the three-dimensional geometric shape data of the curtain wall. It calculates the three-dimensional coordinate information of the curtain wall surface by emitting lasers and receiving reflected signals. This method can generate high-precision point cloud data, which is a set of three-dimensional coordinates of each reflection point, reflecting the overall shape and detailed features of the curtain wall. In the present invention, the point cloud data obtained by the laser radar will serve as the basis for preprocessing the curtain wall point cloud data to ensure the accuracy of the geometric model.

[0050] Ultrasonic sensors are used to detect the physical property data of curtain wall materials. By emitting ultrasonic waves and measuring their propagation speed and attenuation in different materials, the properties of different materials, such as density, hardness, and thickness, can be identified. These physical property data are crucial for accurately calculating the amount of materials used, because different materials have different requirements and installation methods during construction. By obtaining this data, it is possible to better match the actual needs of materials and avoid excessive or insufficient material preparation.

[0051] High-resolution photogrammetry equipment is used to capture image data of the curtain wall surface. These image data not only include visual information such as the color and texture of the curtain wall, but also can extract the surface characteristics of the material through special algorithms. For example, by analyzing the color changes and texture details of the image, the coverage area and installation conditions of different materials can be identified. This is critical for the subsequent optimization of material distribution and management of color consistency.

[0052] After data collection is completed, the geometry data, material data and image data are fused separately. The purpose of data fusion is to integrate data from different sensors into unified pre-processed data, which not only contains the three-dimensional geometric structure of the curtain wall, but also includes the physical properties and surface characteristics of the material. During the data fusion process, the system will eliminate possible errors between sensors through algorithms and ensure the consistency and integrity of the data. The final generated pre-processed curtain wall point cloud data and pre-processed curtain wall material data will be used as input for subsequent modeling and optimization processes, providing a solid foundation for accurate calculation of material usage.

[0053] This multi-sensor fusion data collection and processing method can significantly improve the accuracy and reliability of the data, ensuring that every detail is accurately considered in the material usage calculation process. This method can not only reduce material waste and improve construction efficiency, but also ensure the consistency of the quality and aesthetics of the curtain wall throughout the construction process through precise control of material data. Through this technical means, the present invention can better match material requirements in actual construction, avoid construction problems caused by inaccurate data, and achieve precise management of curtain wall materials.

[0054] like Figure 2 As shown, the pre-processed curtain wall point cloud data and the pre-processed curtain wall material data are input into the adaptive multi-layer feedback modeling system for layered processing, and the model is optimized through the inter-layer feedback mechanism to generate the final three-dimensional model data containing complete geometric structure and material information;

[0055] Preferably, the hierarchical processing in the adaptive multi-layer feedback modeling system includes: dividing the pre-processed curtain wall point cloud data into multiple geometric hierarchical data, each hierarchical data representing geometric details of different precisions, dividing the pre-processed curtain wall material data into multiple material property hierarchical data, each hierarchical data representing different material properties, optimizing the geometric models and material models of each hierarchical level through an inter-layer feedback mechanism, and based on the layer-by-layer optimization and integration of the data of each hierarchical level, finally generating the final three-dimensional model data containing complete geometric structure and material information.

[0056] In the calculation of the amount of curtain wall materials, in order to ensure the accuracy of the calculation results and adapt to the requirements of complex construction environments, the pre-processed curtain wall point cloud data and pre-processed curtain wall material data need to be further processed and optimized. Specifically, these data will be input into the adaptive multi-layer feedback modeling system, which generates the final three-dimensional model data containing complete geometric structure and material information through layered processing and inter-layer feedback mechanism. This process is crucial in the present invention because it determines the basis for subsequent material demand prediction and distribution optimization.

[0057] The hierarchical processing in the adaptive multi-layer feedback modeling system is a process of refining data layer by layer. In this process, the pre-processed curtain wall point cloud data is decomposed into multiple geometric hierarchical data, each of which represents geometric details of different precisions. The purpose of this hierarchical process is to enable the model to gradually approach the actual curtain wall geometry, from macroscopic structure to microscopic details. For example, the initial layer may only contain a rough geometric outline, while higher-level data will contain finer edges, surfaces, and other complex geometric features. This hierarchical process can effectively reduce the consumption of computing resources and ensure the accuracy of the final model through layer-by-layer optimization.

[0058] Similar to the geometric hierarchical data, the pre-processed curtain wall material data is also divided into multiple material property hierarchical data, each of which reflects different material properties, such as material density, glossiness, color variation, and surface roughness. Through this layering, the modeling system can grasp the basic properties of the material at a low level and capture more complex material behaviors at a high level, such as reflective properties under different lighting conditions or material aging over time. These material property hierarchical data are crucial for accurately estimating material requirements because they directly affect the physical properties of the material and its actual performance in construction.

[0059] At each stage of the layering process, the system will apply the inter-layer feedback mechanism to optimize the geometric models and material models of each layer. Specifically, the inter-layer feedback mechanism means that during the generation process of each layer of the model, it will be dynamically adjusted according to the data of the adjacent layers to correct possible deviations or errors. For example, if the geometric model of a certain layer is inconsistent with the expected shape of the upper layer, the system will readjust the data of that layer based on the feedback information until the geometric structure and material properties of the model are consistent and coordinated between the layers. This optimization process ensures that the final generated 3D model not only has a high degree of geometric accuracy, but also truly reflects the material properties of the curtain wall.

[0060] Through layer-by-layer optimization and data integration, the final 3D model data generated by the adaptive multi-layer feedback modeling system will contain complete geometric structure and material information. This final model data has extremely high accuracy and consistency, and can provide a solid foundation for subsequent material demand forecasting and material distribution optimization. Specifically, the model will accurately reflect the actual geometric form and material properties of the curtain wall, so that the material calculation process can fully consider various complex factors in construction, thereby improving material utilization, reducing waste, and optimizing the construction process.

[0061] Through this sophisticated modeling and optimization process, the present invention can not only effectively improve the calculation accuracy of curtain wall material usage, but also dynamically adapt to actual demand changes during construction, providing a more flexible and practical material management solution. This approach is particularly suitable for complex and large-scale construction projects, which can significantly improve construction efficiency, ensure that the project is completed within budget, and meet high-quality construction standards.

[0062] like Figure 3 As shown, the final three-dimensional model data is combined with environmental data, historical construction data and supply chain data, multi-level prediction is performed through a hierarchical neural network, and optimized material distribution plan data is generated through a collaborative optimization algorithm, wherein the material distribution plan data describes the type, quantity, specification and distribution location of the material;

[0063] In order to ensure the accurate demand and optimized distribution of materials, the present invention combines the final 3D model data with environmental data, historical construction data and supply chain data, performs multi-level predictions through hierarchical neural networks, and generates optimized material distribution scheme data through collaborative optimization algorithms. This process is the core step to ensure the optimization of material calculation and distribution, which directly affects the efficiency of material utilization and the economy of construction.

[0064] The final 3D model data is combined with environmental data, historical construction data and supply chain data to comprehensively consider the geometric structure, material properties and actual conditions of the building curtain wall and the construction environment. The final 3D model data provides accurate geometric information and material properties of the building curtain wall, which is the basis for material demand forecasting. Environmental data (such as climate conditions, temperature changes, wind speed, etc.) provides a dynamic background of the construction site, which can help predict changes in material demand under different environmental conditions. Historical construction data includes material consumption and construction progress in previous similar projects, which can provide reference for material demand in the current project. Supply chain data reflects the supply status, price fluctuations and delivery cycle of materials, which is crucial for formulating reasonable material procurement and distribution plans.

[0065] By inputting these multi-dimensional data into a hierarchical neural network, the system can make multi-level predictions of material demand. A hierarchical neural network is a deep learning model that can handle complex data relationships. It can analyze data at different levels to produce more accurate predictions. Specifically, the primary level of the neural network is responsible for processing building geometry and material information to generate basic material demand forecasts; the intermediate level of the neural network combines environmental data and historical construction data to dynamically adjust material demand; the advanced level of the neural network combines supply chain data to optimize the procurement and supply cycle of materials, thereby predicting the demand for materials at different time points. Through this layer-by-layer precision prediction method, the hierarchical neural network can generate accurate and dynamically adaptable material demand data.

[0066] It is not enough to have accurate demand forecasts. In order to maximize material utilization and minimize waste, the system further optimizes the material distribution plan through collaborative optimization algorithms. Collaborative optimization algorithms include genetic algorithms, particle swarm optimization algorithms, and simulated annealing algorithms, which can select the best solution from multiple alternatives. Genetic algorithms generate multiple material distribution plans by simulating the process of biological evolution, and find the optimal solution through selection and mutation operations. The particle swarm optimization algorithm further optimizes the results of the genetic algorithm by simulating group behavior to ensure that the distribution of materials on the construction site can minimize transportation and storage costs. Finally, the simulated annealing algorithm performs a global search for the optimized solution to further eliminate the influence of the local optimal solution and ensure that the selected solution is optimal overall.

[0067] The resulting material distribution plan data describes the types, quantities, specifications, and specific distribution locations of the materials. These data provide detailed guidance for the construction process, so that the distribution, storage, and use of materials can be as efficient as possible. For example, in the early stages of construction, only some basic support materials may be required, while a large amount of glass or metal sheets may be required when installing curtain wall panels. By accurately predicting and optimizing the distribution of these needs, the construction party can prepare the corresponding materials in advance to avoid construction delays or waste caused by material shortages or excessive stockpiling.

[0068] Through this multi-level prediction and collaborative optimization, the present invention not only improves the accuracy of material demand prediction, but also optimizes the distribution and use of materials, reduces construction costs, and shortens the construction period. This method is particularly suitable for large and complex construction projects, and can significantly improve the economy and efficiency of construction while ensuring quality.

[0069] Preferably, the hierarchical neural network includes a neural network structure of multiple levels. After the final three-dimensional model data is combined with the environmental data, historical construction data and supply chain data, it is first input into the first layer of neural network for preliminary material demand forecasting. The forecast result is passed as input data to the second layer of neural network for supply chain fluctuation forecasting, and is further adjusted in the third layer of neural network in combination with the construction environment data to finally generate material demand forecast data. The material demand forecast data is generated based on the following calculation expression:

[0070]

[0071] Among them, D t represents the material demand in time period t; M i represents the initial demand for the i-th material; F i represents the adjustment factor based on environmental data, historical construction data and supply chain data; H t Indicates the revised value of material demand based on historical data. i Used to correct the initial demand to make it more in line with the actual situation; H t The forecast results are then fine-tuned taking into account actual material consumption from historical projects.

[0072] In the present invention, the hierarchical neural network plays a key role. It processes and predicts various input data through a multi-level neural network structure to generate accurate material requirements. Specifically, the final three-dimensional model data is combined with environmental data, historical construction data and supply chain data, and is first input into the first layer of the neural network for preliminary material demand prediction. This layer of neural network mainly focuses on the geometric shape and material information of the curtain wall, and combines the environmental data to calculate the initial material demand under specific construction conditions. This preliminary prediction takes into account the actual structure and material distribution of the building curtain wall, but does not fully consider the impact of supply chain fluctuations and changes in the construction environment.

[0073] The preliminary material demand forecasts generated by the first layer of neural networks are passed to the second layer of neural networks. The second layer of neural networks mainly processes supply chain data and analyzes the supply capacity and market volatility of materials. The main goal of supply chain volatility forecasting is to evaluate the stability of material supply and possible price fluctuations, and then adjust the preliminary forecast of material demand. For example, if a key material is in short supply in the market, the second layer of neural networks may increase the demand for the material based on this information, or recommend advance procurement within a specific period of time. The forecast results of this layer are closer to the actual material demand because it begins to take into account the difficulty of obtaining materials and cost fluctuations.

[0074] The output of the second layer of neural network is passed as input data to the third layer of neural network, which is further adjusted in combination with the construction environment data. The construction environment data includes factors such as climate conditions, temperature, humidity, wind speed, etc. during the construction period. These environmental factors will directly affect the use of materials. For example, in a high temperature environment, some materials may require more protection measures, or in high humidity conditions, the consumption of materials may increase. Therefore, the third layer of neural network further adjusts the material demand by taking these environmental factors into consideration, making the prediction results closer to the actual situation on the construction site.

[0075] After these three levels of gradual prediction and adjustment, the hierarchical neural network generates the final material demand forecast data. This forecast data not only includes the type and quantity of materials, but also clearly plans the demand time point for each material to ensure that the supply of materials can perfectly match the demand during the construction process, avoiding shortages or excessive reserves of materials.

[0076] Through this calculation expression, the system can integrate the prediction results at each level to generate accurate material demand. This prediction process takes into account a variety of complex factors, and through the multi-level processing of neural networks, the accuracy and real-time nature of material demand prediction are ensured. In practical applications, the present invention can effectively reduce material waste, optimize resource allocation, and improve construction efficiency. This hierarchical prediction model is particularly suitable for large and complex construction projects, and can maintain a high degree of coordination between material supply and demand in a changing construction environment, ensuring the smooth progress of the project.

[0077] Preferably, the material demand forecast data is obtained through the output results of a multi-layer neural network and is used to optimize the material supply and use strategy.

[0078] In the present invention, material demand forecast data plays a key role, and its generation process depends on the output results of multi-layer neural networks. Through the multi-layer neural network, the system can obtain accurate material demand forecast data by processing and predicting various input data, and use it to optimize the supply and use strategy of materials to ensure that the material supply in the construction process is highly matched with the actual demand, avoiding resource waste and supply shortage.

[0079] The multi-layer neural network generates material demand forecast data by processing the final 3D model data, environmental data, historical construction data and supply chain data layer by layer. Each layer of the neural network processes data of different dimensions respectively. The initial layer focuses on the building geometry and material information to generate basic material demand forecasts; the intermediate layer dynamically adjusts the supply chain data to consider the market supply status and price fluctuations of the materials; the advanced layer further adjusts the material demand in combination with the construction environment data to ensure that the forecast results are consistent with the actual conditions of the construction site. This layer-by-layer processing and forecasting method allows the material demand forecast data to not only take into account the actual needs of the building structure, but also dynamically adapt to changes in the external environment and market conditions.

[0080] The generated material demand forecast data is highly accurate and real-time, and can be directly used to optimize the material supply and use strategy. Specifically, the optimization of the material supply strategy is based on the accurate forecast of the material demand. By analyzing the forecast data, the system can determine the best purchase time and quantity for each material, avoiding storage pressure and excessive stockpiling caused by premature purchases, and avoiding material shortages caused by insufficient purchases. The system can also adjust the procurement plan in real time based on the supply chain data, reserve key materials in advance when the market supply is tight, or make flexible purchases when prices fluctuate greatly to reduce costs.

[0081] In terms of optimizing material usage strategies, material demand forecast data provides clear guidance. The system can reasonably arrange the distribution and use of materials based on the forecast data to ensure that sufficient materials are available at each stage of construction, while avoiding waste caused by excess materials. For example, in the early stages of curtain wall installation, less support materials may be required, but as construction progresses, the demand for glass or metal sheets will increase rapidly. Through accurate prediction of material demand and phased usage planning, the system can optimize the distribution and use of materials to ensure that each material can be reasonably used at the best time and place.

[0082] A typical implementation example is the curtain wall installation project in a large construction project. The project needs to take into account the construction progress, weather changes, and market supply conditions at the same time. Through the material demand forecast data generated by the multi-layer neural network, project managers can accurately plan the demand curve of various materials in the next few months, and then formulate accurate procurement and use strategies. For example, based on the forecast results, the project can decide to reserve a certain key material in advance within a certain period of time to cope with the upcoming construction peak, without delaying the construction period due to insufficient material supply during the peak period. In addition, in the optimization of material use strategies, the system can also adjust the construction plan in real time according to the forecast data to ensure the rational use of materials and dynamically replenish them according to actual consumption.

[0083] Preferably, the collaborative optimization algorithm uses the material demand forecast data as the initial input, uses the genetic algorithm for preliminary optimization, generates multiple material distribution schemes, and inputs these schemes into the particle swarm optimization algorithm for further optimization to select the scheme with the highest efficiency and the minimum resource consumption, and finally globally optimizes the selected scheme through the simulated annealing algorithm to generate optimized material distribution scheme data, and the material distribution scheme data is described by the following calculation expression:

[0084]

[0085] Among them, S opt represents the optimized material distribution scheme, which is the optimal scheme finally selected by the system, taking into account multiple influencing factors; C j represents the cost of the jth material, including procurement cost, transportation cost, and storage cost, etc., to ensure that the material distribution is cost-effective; P j R represents the distribution location priority of the jth material, which reflects the importance of different locations of materials in the construction site and ensures that key materials are distributed preferentially in places where they are more needed; k represents the resource utilization rate in the kth time period, which takes into account the effective use of materials in different time periods to ensure efficient use of resources; T kIt represents the resource utilization cost in the kth time period, including the storage, transportation and use costs of materials in a specific time period.

[0086] In the calculation of curtain wall material usage, the collaborative optimization algorithm plays a key role. It performs multi-stage optimization processing on the material demand forecast data and finally generates a material distribution plan that strikes the best balance between cost, resource utilization and material distribution location priority. Specifically, the collaborative optimization algorithm gradually optimizes the material distribution plan through a combination of genetic algorithm, particle swarm optimization algorithm and simulated annealing algorithm to ensure the efficiency and economy of material use during the construction process.

[0087] The collaborative optimization algorithm uses the material demand forecast data as the initial input and uses a genetic algorithm for preliminary optimization. The genetic algorithm is an optimization method based on natural selection and genetic mechanisms. It generates multiple material distribution schemes by simulating the selection, crossover and mutation operations in the biological evolution process. In the present invention, the preliminary optimization stage of the genetic algorithm generates several feasible material distribution schemes, which provide a variety of possible material distribution strategies after considering the material demand, supply chain conditions and basic cost constraints. The advantage of the genetic algorithm is that it can explore a wide range of solution spaces and quickly find several near-optimal solutions, providing a good foundation for subsequent optimization.

[0088] These material distribution schemes generated by the genetic algorithm are input into the particle swarm optimization algorithm for further optimization. The particle swarm optimization algorithm is an optimization method based on swarm intelligence, which simulates the process of individuals in a group jointly finding the optimal solution through mutual cooperation and learning. In the present invention, the particle swarm optimization algorithm further refines the material distribution scheme, optimizes the distribution of materials in different locations and the timing of material use. Specifically, the particle swarm optimization algorithm finds the best balance between maximum efficiency and minimum resource consumption by simulating the actual distribution of materials at the construction site. The optimization at this stage pays special attention to the practical operability of the construction site, such as minimizing the material handling and storage costs while ensuring the construction progress. Through the particle swarm optimization algorithm, the system can select one or more material distribution schemes with higher efficiency and lower cost.

[0089] On the basis of particle swarm optimization, the system inputs these optimized material distribution schemes into the simulated annealing algorithm for global optimization. The simulated annealing algorithm is a random search algorithm that simulates the annealing principle in the physical process and accepts poor solutions with a certain probability, thereby avoiding falling into the local optimal solution. In the present invention, the global optimization process of the simulated annealing algorithm can effectively eliminate the local unreasonableness in the material distribution scheme and ensure that the selected scheme is optimal overall. For example, when considering the overall cost of material distribution, the simulated annealing algorithm may choose a scheme that slightly increases the cost in some time periods but significantly saves resources in other time periods, thereby achieving the global optimum.

[0090] Through this optimization process, the collaborative optimization algorithm can find a material distribution plan that meets construction requirements and minimizes resource consumption. In practical applications, this optimization plan can significantly reduce material waste, reduce construction costs, and improve overall construction efficiency. For example, in a large-scale construction project, this optimization algorithm can ensure that key materials are given priority at critical moments in construction, while avoiding waste and storage problems caused by excessive accumulation of materials, thereby achieving optimal resource allocation and maximum cost savings. This collaborative optimization method is particularly suitable for complex construction projects that require refined management. It can help project managers make optimal decisions in a changing construction environment and ensure the smooth progress and ultimate success of the project.

[0091] like Figure 4 As shown, during the construction process, the construction progress data and actual material usage data are collected in real time, the data are compared with the optimized material distribution plan data, feedback optimization instructions are generated through a multi-level feedback analysis system, and the material prediction model and material distribution plan are adaptively adjusted based on the feedback results to generate the final optimized material distribution data;

[0092] In the process of calculating and optimizing the amount of curtain wall materials, the real-time data collection and feedback adjustment mechanism at the construction site is an important link to ensure the rationality of material allocation and the smooth progress of construction. Specifically, during the construction process, through the sensors and monitoring systems deployed on site, the construction progress data and actual material usage data are collected in real time, and these real-time data are compared with the previously generated optimized material distribution plan data. Feedback optimization instructions are generated through a multi-level feedback analysis system, so as to adaptively adjust the material prediction model and material distribution plan, and finally generate more accurate optimized material distribution data.

[0093] The real-time collection of construction progress data and actual material usage data is the basis of this feedback mechanism. Construction progress data includes information such as the specific stage of the current construction, the amount of work completed, and the estimated remaining construction period; actual material usage data records the specific usage of various materials in each construction stage. These data are automatically collected through sensors and intelligent monitoring systems and uploaded to the central control system in real time. This real-time data collection method enables the system to continuously monitor the progress of construction and capture any changes that may affect material demand, such as accelerated or delayed construction speed, abnormal fluctuations in material consumption, etc.

[0094] The system compares these real-time collected data with the pre-set optimized material distribution plan data. The optimized material distribution plan data is generated based on the pre-construction material demand forecast and supply chain analysis. It contains detailed information such as the type, quantity and distribution location of materials to be used in each construction stage. By comparing the actual construction data with this plan, the system can identify deviations between the actual situation and the expected plan. For example, if the actual material consumption at a certain stage exceeds expectations, it may mean that an unexpected situation has occurred during the construction process and the material supply plan needs to be adjusted immediately.

[0095] The multi-level feedback analysis system plays a core role. The system not only compares the differences between actual data and planned data, but also generates feedback optimization instructions based on these differences. These instructions specifically include adjustments to the increase or decrease of material supply, optimization and re-planning of material distribution locations, etc. The power of the multi-level feedback analysis system lies in its ability to process complex multi-dimensional data and dynamically adjust according to real-time changes. For example, if the system finds that a certain material is consumed too quickly during the construction process, it may immediately generate instructions to recommend purchasing in advance or deploying the material from other construction areas that are not in urgent need to ensure that the entire construction progress is not affected.

[0096] After the feedback optimization instructions are generated, the system will adaptively adjust the material forecast model and material distribution plan based on these instructions. Adaptive adjustment means that the system will not only correct the material distribution of the current construction stage, but also re-evaluate the subsequent stages of the entire construction project and update the material demand forecast model. This dynamic adjustment mechanism ensures that the material supply chain of the entire construction process can flexibly respond to the actual situation on site, avoiding construction stagnation due to insufficient materials or waste due to excess materials.

[0097] The system generates the final optimized material distribution data based on the adjusted material prediction model and distribution plan. This data set not only reflects the material demand in the current construction phase, but also provides updated material allocation guidance for subsequent construction. This continuous iterative optimization approach can maximize the efficiency of material utilization, ensure the stability of the construction progress and the controllability of project costs.

[0098] A specific implementation example could be a curtain wall installation project for a high-rise building. Assume that in the mid-term stage of the project, the system detects that due to changes in weather conditions, the construction progress is ahead of schedule, resulting in the actual use of materials exceeding the initial expectations. After the system collects this data in real time, it generates feedback optimization instructions through a multi-level feedback analysis system, suggesting that some of the materials originally planned to be used in the next stage be deployed to the current construction area in advance, and at the same time, it is recommended to speed up the subsequent procurement process of the materials to avoid material shortages in subsequent construction. Through this adaptive adjustment mechanism, project managers can respond to unexpected situations in construction in a timely manner, keep the construction going smoothly, and ensure the optimal allocation of material resources.

[0099] Preferably, the multi-level feedback analysis system monitors and collects construction progress data and actual material consumption data in real time during construction, dynamically compares and analyzes the obtained real-time monitoring data with the pre-formulated optimized material distribution plan, and generates corresponding feedback optimization instructions. The feedback optimization instructions are used to dynamically adjust the material demand prediction model and the material distribution plan according to the actual needs of the construction site to ensure the balance between material supply and actual construction needs. The feedback optimization instructions are generated by the following calculation expression:

[0100] F o =P c -A u

[0101] Among them, F o represents feedback optimization instructions, which are specific operation instructions generated by the system for adjusting material supply and distribution; c Indicates the planned material usage, that is, the amount of material expected to be used in a certain stage according to the optimized material distribution plan; A u It represents the actual material usage, that is, the material consumption recorded in real time during the construction process.

[0102] In the process of calculating and managing the amount of curtain wall materials, the multi-level feedback analysis system is a key tool to ensure the balance between material supply and actual construction needs. The system monitors and collects construction progress data and actual material consumption data in real time during construction, and dynamically compares and analyzes these real-time monitoring data with the pre-established optimized material distribution plan to generate corresponding feedback optimization instructions. These instructions are used to dynamically adjust the material demand prediction model and material distribution plan according to the actual needs of the construction site, ensuring that the supply of materials during the construction process can respond to changes in actual demand in a timely manner and avoid material shortages or surpluses.

[0103] The core of the multi-level feedback analysis system lies in the monitoring and collection of real-time data. Various information on the construction site, including the current construction progress, the amount of work completed, and the actual consumption of various materials, will be captured in real time and transmitted to the central control system through sensors and monitoring systems deployed on site. The construction progress data includes the completion status of each stage of the project, while the actual material consumption data reflects the real-time use of each material. Through these data, the system can accurately grasp the material consumption of the current construction stage as a basis for judging whether the material supply matches the demand.

[0104] Once the system collects this real-time data, it will dynamically compare and analyze it with the previously developed optimized material distribution plan. The optimized material distribution plan is generated before construction begins based on the expected material demand forecast and supply chain conditions, aiming to ensure that there is sufficient material supply at each construction stage. However, various unexpected situations often occur during the construction process, such as accelerated progress, material waste, or additional material requirements. By comparing actual data with the pre-planned plan, the system is able to identify these deviations and generate corresponding feedback optimization instructions through a multi-level feedback analysis system.

[0105] Through this calculation, the system can clearly determine the degree of deviation in material usage. u Less than the planned usage P c , indicating that the current material supply is in excess, the system may suggest reducing the subsequent supply of materials or allocating materials to other construction areas; on the contrary, if A u Greater than P c , it means that there is a shortage of materials at present. The system may suggest increasing the material supply or speeding up the purchase and distribution of materials to avoid affecting the construction progress.

[0106] The application effect of the multi-level feedback analysis system is remarkable. Through continuous real-time monitoring and feedback, the system can dynamically adjust the material supply strategy to ensure that the material supply at the construction site closely matches the actual demand. This real-time adjustment mechanism can not only reduce material waste and reduce project costs, but also prevent construction stagnation caused by material shortages, thereby improving overall construction efficiency.

[0107] For example, in a curtain wall construction project of a large commercial building, if the consumption of a certain curtain wall panel exceeds expectations due to accelerated construction, the system will generate feedback optimization instructions based on the actual material usage, suggesting that project managers allocate more panels from suppliers in advance, or allocate materials from other less urgent construction areas to ensure the continuation of the project. Conversely, if it is found that the usage of a certain supporting material is lower than expected, the system may suggest postponing or reducing the subsequent purchase of the material to avoid material accumulation and capital occupation.

[0108] In general, the multi-level feedback analysis system effectively improves the flexibility and response speed of material management through real-time data monitoring, dynamic comparative analysis and the generation of feedback optimization instructions. It can adjust the material distribution and supply strategy according to the real-time situation of the construction site to ensure the balance between material supply and demand, thereby optimizing the construction process, reducing costs, and improving the management level and success rate of the entire project.

[0109] Preferably, the feedback optimization instructions are used to adjust the material distribution strategy in real time during the actual construction process to reduce material waste and improve material utilization, and the adjustment process is achieved through continuous monitoring by a multi-level feedback analysis system.

[0110] In the process of calculating and managing the amount of curtain wall materials, feedback optimization instructions are a key tool to ensure real-time adjustment of material distribution strategies. This instruction is achieved through continuous monitoring by a multi-level feedback analysis system, with the aim of reducing material waste and improving material utilization during the actual construction process. By analyzing and feeding back real-time data from the construction site, the system can flexibly adjust the material distribution and usage strategy to ensure the most efficient use of material resources.

[0111] The core of the multi-level feedback analysis system lies in its real-time monitoring capability. The system continuously monitors the construction progress and the actual consumption of materials at every stage of construction. Through sensors and data acquisition devices deployed at the construction site, the system can obtain the usage status, inventory level and current construction needs of each material in real time. This data is immediately transmitted to the central control system for dynamic analysis and decision-making.

[0112] When the system identifies problems in material distribution and use, such as material consumption being too fast or too slow, it generates corresponding feedback optimization instructions. These instructions directly affect the subsequent material distribution strategy. For example, if the system finds that a certain key material is consumed faster than expected, it may recommend allocating more materials to the current construction area in advance, or allocating resources from other less urgent areas to ensure that the construction progress is not affected. Conversely, if the actual consumption of a certain material is lower than expected, the system may recommend reducing the distribution of the material, postponing or canceling subsequent purchases to prevent waste caused by excess materials.

[0113] The generation and execution of this feedback optimization instruction is a dynamic cycle. The system continuously adjusts the material distribution strategy through a multi-level feedback analysis mechanism to adapt to the actual situation on the construction site. For example, in the curtain wall construction of a high-rise building, a large amount of structural support materials may be required in the early stage, but as the construction progresses, the demand for these materials gradually decreases. At this time, the system will generate feedback instructions based on real-time data, adjust the distribution of support materials, and shift resources to other more urgent material needs, such as glass panels or insulation materials.

[0114] This adjustment process is made possible through continuous monitoring by a multi-level feedback analysis system. The system not only responds passively to problems in material use, but also proactively predicts possible fluctuations in material demand through continuous analysis of various data, and adjusts the distribution strategy in advance. This foresight enables material resources to be optimally configured and utilized throughout the construction cycle. For example, in the event of a sudden change in weather, the system may predict that the demand for certain materials will change, and make adjustments in advance to prevent construction delays or material shortages.

[0115] The effect of this real-time adjustment is significant. On the one hand, it can significantly reduce material waste. By accurately allocating material supply, it avoids excessive accumulation of materials and the resulting waste. On the other hand, it improves the utilization rate of materials, ensuring that each batch of materials can be used where and when they are most needed, maximizing the economic value of the materials.

[0116] A specific implementation example may be in a curtain wall installation project for a large commercial complex. When the system detects that the demand for curtain wall panels of a certain material exceeds expectations during the construction process, and the supply chain cannot respond quickly in the short term, the system will generate feedback optimization instructions based on the actual data on site, suggesting that the material be allocated to key construction areas first and its use in non-key areas be reduced. At the same time, the system will prompt managers to urgently deploy other alternative materials to ensure the continuity of the construction progress. This dynamic adjustment strategy not only avoids construction interruptions, but also ensures the best use of resources.

[0117] In summary, feedback optimization instructions play a vital role in the calculation and management of curtain wall material usage. Through continuous monitoring and real-time adjustment of the multi-level feedback analysis system, it can ensure the dynamic balance between material supply and demand, reduce waste, improve utilization, and ultimately provide a solid guarantee for the smooth completion of the project. This intelligent material management method is particularly suitable for complex large-scale construction projects, and can effectively improve the efficiency of construction management and the success rate of the project.

[0118] A curtain wall material usage list is generated based on the final optimized material distribution data, and material procurement and construction execution are carried out according to the list, and construction data records and project archiving data are generated for subsequent analysis and optimization.

[0119] Preferably, a curtain wall material usage list is generated based on the final optimized material distribution data, the list including material types, specifications, quantities and specific usage locations. Material procurement and construction execution are carried out according to the list. After the construction is completed, construction data records and project archiving data are generated and used for analysis and optimization of subsequent projects, providing a reference for future material demand forecasting and distribution optimization.

[0120] In the process of calculating and managing the amount of curtain wall materials, generating an accurate material usage list is a key step to ensure smooth construction and effective use of material resources. This list is generated based on the final optimized material distribution data, and contains detailed material types, specifications, quantities, and specific usage locations of each material. The procurement and construction execution of materials are strictly based on this list to ensure seamless connection between the supply and use of materials during the construction process. After the construction is completed, the system will generate construction data records and project archive data, which will not only be used for the summary and analysis of the current project, but also serve as an important reference for material demand forecasting and distribution optimization for future projects.

[0121] The final optimized material distribution data is generated through a multi-level prediction and optimization process, reflecting the optimal configuration of materials throughout the construction cycle. These data comprehensively consider the geometric structure of the building curtain wall, material properties, environmental impact, historical construction data, supply chain conditions and other factors to ensure that materials can be reasonably allocated and used at each construction stage. Based on these optimized data, the generated material usage list is highly accurate and practical, and can directly guide material procurement and construction execution.

[0122] The material bill of quantities details the type, specification, quantity and specific location of each material required during the construction process. The type and specification section ensures that the purchased materials fully meet the requirements of the construction design, while the quantity and location section ensures that the supply and distribution of materials can meet the actual construction needs. For example, for curtain wall glass panels, the bill will not only detail the required glass type and size, but also specify the installation location of each piece of glass. This detailed description allows material management personnel to accurately match suppliers' products when purchasing, avoiding construction delays caused by inconsistent specifications or insufficient quantities.

[0123] During the construction process, all materials are purchased and used strictly according to the material usage list. When purchasing materials, managers can directly place orders based on the list to ensure that the purchased materials can meet the construction needs in a timely and accurate manner. During construction, the construction team follows the guidance of the list to reasonably allocate and use materials to ensure that each piece of material can be effectively used at the right time and place. This precise execution method can not only greatly reduce material waste, but also reduce construction costs and improve the economy and efficiency of the overall project.

[0124] After the construction is completed, the system will generate detailed construction data records and project archive data. The construction data records include information such as the actual amount of materials used, construction progress, and quality inspection results, while the project archive data is a summary and summary of the entire construction process. These data are not only helpful for the summary and audit of the current project, but also provide valuable reference for subsequent projects. For example, managers can use these data to analyze the actual use of materials, evaluate the effectiveness of optimization strategies, and improve future material demand forecasting models and distribution optimization plans.

[0125] A specific implementation example could be a curtain wall installation project for a high-rise building. At the beginning of construction, the system generates a detailed material usage list based on the optimized material distribution data, listing the specifications, quantity, and installation location of each required curtain wall material. Based on the list, the procurement department accurately places purchase orders to ensure that the materials are in place on time at all stages of construction. During the construction process, the construction team strictly distributes and uses materials according to the guidelines in the list, reducing material waste and improving construction efficiency. After the construction is completed, the system automatically generates construction data records and project archive data. By analyzing these data, project managers can summarize their experience and provide more accurate material management solutions for similar projects in the future.

[0126] Through this series of precise calculations and management processes, the present invention can ensure the efficient use of curtain wall materials, reduce waste, lower construction costs, and provide continuously optimized reference data for future projects. This refined and intelligent material management method is particularly suitable for large and complex construction projects, and can significantly improve the overall efficiency of construction management and the success rate of projects.

[0127] like Figure 5 As shown, a system for implementing the method for optimizing the amount of curtain wall materials, the system comprising:

[0128] A multi-sensor module is used to be deployed at different positions of the building curtain wall to collect geometric shape data and material data of the curtain wall, and the multi-sensor module includes a laser radar, an ultrasonic sensor and a high-resolution photogrammetry device; the multi-sensor module is deployed at different positions of the building curtain wall to collect geometric shape data and material data of the curtain wall. The module includes a laser radar, an ultrasonic sensor and a high-resolution photogrammetry device. The laser radar can accurately capture the three-dimensional geometric shape data of the curtain wall, and construct a detailed three-dimensional point cloud of the curtain wall by emitting laser pulses and measuring their return time. The ultrasonic sensor is used to detect the physical property data of the curtain wall material, especially to capture the internal structure information of the material that cannot be obtained by the visual sensor. The high-resolution photogrammetry device generates high-precision two-dimensional image data by capturing images of the curtain wall surface. These data are initially fused in the multi-sensor module to ensure the accuracy of subsequent processing.

[0129] The data processing module is used to perform data cleaning and noise filtering on the geometric shape data and material data received from the multi-sensor module to generate pre-processed curtain wall point cloud data and pre-processed curtain wall material data; after receiving the geometric shape data and material data transmitted from the multi-sensor module, the data processing module first performs data cleaning and noise filtering. The purpose of data cleaning is to remove erroneous data or abnormal data caused by factors such as the environment and equipment during the acquisition process to ensure the accuracy of the data. Noise filtering is to remove high-frequency noise components in the measurement data through algorithms, making the generated point cloud data and material data clearer and more accurate. Finally, the data processing module generates pre-processed curtain wall point cloud data and pre-processed curtain wall material data, providing a high-quality data foundation for subsequent modeling and optimization.

[0130] The adaptive multi-layer feedback modeling module is used to receive the pre-processed curtain wall point cloud data and the pre-processed curtain wall material data, perform hierarchical processing on them, optimize the model through the inter-layer feedback mechanism, and generate the final three-dimensional model data containing complete geometric structure and material information; the adaptive multi-layer feedback modeling module receives the pre-processed data output by the data processing module and performs hierarchical processing on them. The core of this module is to optimize the geometric model and material model through the hierarchical feedback mechanism. Specifically, the pre-processed curtain wall point cloud data is divided into multiple geometric hierarchical data, each hierarchical data represents geometric details of different precisions; at the same time, the pre-processed curtain wall material data is also divided into multiple material characteristic hierarchical data, each hierarchical data represents different material characteristics. Through the inter-layer feedback mechanism, each hierarchical model is continuously adaptively optimized, and finally generates three-dimensional model data containing complete geometric structure and material information. This processing method not only improves the accuracy of the model, but also can better adapt to building structures of different complexities.

[0131] The hierarchical neural network prediction module is used to combine the final three-dimensional model data with environmental data, historical construction data and supply chain data, generate material demand prediction data through multi-level prediction, and generate optimized material distribution scheme data through collaborative optimization algorithm, wherein the material distribution scheme data describes the type, quantity, specification and distribution location of the material; the hierarchical neural network prediction module combines the final three-dimensional model data generated by the adaptive multi-layer feedback modeling module with environmental data, historical construction data and supply chain data to perform multi-level prediction. The module first performs a preliminary prediction of material demand through the multi-layer structure of the neural network, then adjusts it through the prediction of supply chain fluctuations, and finally further optimizes it in combination with the construction environment data. Through this multi-level prediction, the system can generate material demand prediction data and generate optimized material distribution scheme data through collaborative optimization algorithm. The scheme data describes the type, quantity, specification and distribution location of the material in detail to ensure the efficient use of the material during the construction process.

[0132] The multi-level feedback analysis module is used to collect construction progress data and actual material usage data in real time during the construction process, compare and analyze the data with the optimized material distribution plan data, generate feedback optimization instructions, and adaptively adjust the material prediction model and material distribution plan based on the feedback results to generate the final optimized material distribution data; the multi-level feedback analysis module collects construction progress data and actual material usage data in real time during the construction process, and compares and analyzes these data with the optimized material distribution plan data. Through this dynamic comparative analysis, the system can generate feedback optimization instructions for adaptively adjusting the material demand prediction model and material distribution plan according to the actual needs of the construction site. Ultimately, the optimized material distribution data generated by the system can better adapt to the actual construction situation and ensure the balance of material supply and demand and the continuity of construction.

[0133] The material management module is used to generate a curtain wall material usage list based on the final optimized material distribution data, and to procure materials and execute construction according to the list, and to generate construction data records and project archiving data for subsequent analysis and optimization. The material management module generates a curtain wall material usage list based on the final optimized material distribution data. The list details the required material types, specifications, quantities and specific locations of use. The construction team procures materials and executes construction according to the list to ensure that the construction progress is not affected by material supply. After the construction is completed, the system generates construction data records and project archiving data for subsequent analysis and optimization. These data not only help to summarize the lessons learned from the current project, but also provide an important reference for material demand forecasting and distribution optimization for future projects.

[0134] Through this comprehensive and sophisticated calculation and management process, the system of the present invention can significantly reduce material waste, reduce construction costs, and improve overall construction efficiency. The present invention is particularly suitable for complex construction projects and can effectively optimize resource allocation to ensure efficient implementation and smooth completion of the project.

[0135] In practical applications, a preliminary three-dimensional curtain wall processing model is generated by extracting the design information of the curtain wall, classifying and summarizing the extracted information. On the cloud platform, multiple users can collaborate on the same project, generate a wireframe model of the curtain wall through an adaptive multi-layer feedback modeling module, and then convert the wireframe model into three-dimensional model data containing complete geometric structure and material information. Next, the three-dimensional model is given facade information, and the curtain wall unit panel is generated and stored in the database. The unit panel is given specific building address information, and the related aluminum profile mold data is passed to the database. Finally, through the collaborative optimization algorithm, a parametric processing model is completed to guide the factory to carry out efficient and accurate processing and production.

[0136] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware.

[0137] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A method for optimizing the amount of curtain wall materials, characterized in that: The following steps are involved: The geometric shape data and the curtain wall material data of the building curtain wall are collected by using a variety of sensors deployed at different positions of the building curtain wall, and the data are cleaned and noise filtered to generate pre-processed curtain wall point cloud data and pre-processed curtain wall material data; Inputting the pre-processed curtain wall point cloud data and the pre-processed curtain wall material data into an adaptive multi-layer feedback modeling system for layered processing, optimizing the model through an inter-layer feedback mechanism, and generating final three-dimensional model data containing complete geometric structure and material information; The final three-dimensional model data is combined with environmental data, historical construction data and supply chain data, multi-level prediction is performed through a hierarchical neural network, and optimized material distribution scheme data is generated through a collaborative optimization algorithm, wherein the material distribution scheme data describes the type, quantity, specification and distribution location of the material; During the construction process, the construction progress data and actual material usage data are collected in real time, and the data are compared with the optimized material distribution plan data. Feedback optimization instructions are generated through a multi-level feedback analysis system, and the material prediction model and material distribution plan are adaptively adjusted based on the feedback results to generate the final optimized material distribution data; A curtain wall material usage list is generated based on the final optimized material distribution data, and material procurement and construction execution are carried out according to the list, and construction data records and project archiving data are generated for subsequent analysis and optimization.

2. The method for optimizing the amount of curtain wall materials according to claim 1, characterized in that: The multiple sensors include laser radar, ultrasonic sensor and high-resolution photogrammetry equipment. The laser radar is used to obtain three-dimensional geometric shape data of the curtain wall, the ultrasonic sensor is used to detect physical property data of the curtain wall material, and the high-resolution photogrammetry equipment is used to capture image data of the curtain wall surface, and the geometric shape data, material data and image data are respectively fused to generate pre-processed curtain wall point cloud data and pre-processed curtain wall material data.

3. The method for optimizing the amount of curtain wall materials according to claim 2, characterized in that: The hierarchical processing in the adaptive multi-layer feedback modeling system includes: dividing the pre-processed curtain wall point cloud data into multiple geometric hierarchical data, each hierarchical data represents geometric details of different precisions, dividing the pre-processed curtain wall material data into multiple material property hierarchical data, each hierarchical data represents different material properties, optimizing the geometric models and material models of each hierarchical level through the inter-layer feedback mechanism, and finally generating the final three-dimensional model data containing complete geometric structure and material information based on the layer-by-layer optimization and integration of the data of each hierarchical level.

4. The method for optimizing the amount of curtain wall materials according to claim 1, characterized in that: The hierarchical neural network includes a neural network structure of multiple levels. After the final three-dimensional model data is combined with environmental data, historical construction data and supply chain data, it is first input into the first layer of neural network for preliminary material demand forecasting. The forecast result is passed as input data to the second layer of neural network for supply chain fluctuation forecasting, and is further adjusted in the third layer of neural network in combination with construction environment data to finally generate material demand forecast data. The material demand forecast data is generated based on the following calculation expression: Among them, D t represents the material demand in time period t; M i represents the initial demand for the i-th material; F i represents the adjustment factor based on environmental data, historical construction data and supply chain data; H t Indicates the revised value of material demand based on historical data.

5. The method for optimizing the amount of curtain wall material according to claim 4, characterized in that: The material demand forecast data is obtained through the output results of a multi-layer neural network and is used to optimize the material supply and use strategy.

6. The method for optimizing the amount of curtain wall material according to claim 4, characterized in that: The collaborative optimization algorithm uses the material demand forecast data as the initial input, uses the genetic algorithm for preliminary optimization, generates multiple material distribution schemes, and inputs these schemes into the particle swarm optimization algorithm for further optimization to select the scheme with the highest efficiency and the minimum resource consumption. Finally, the selected scheme is globally optimized by the simulated annealing algorithm to generate optimized material distribution scheme data, which is described by the following calculation expression: Among them, S opt represents the optimized material distribution scheme; C j represents the cost of the jth material; P j represents the distribution position priority of the jth material; R k represents the resource utilization rate in the kth time period; T k Represents the resource utilization cost in the kth time period.

7. The method for optimizing the amount of curtain wall materials according to claim 1, characterized in that: The multi-level feedback analysis system monitors and collects construction progress data and actual material consumption data in real time during construction, dynamically compares and analyzes the obtained real-time monitoring data with the pre-established optimized material distribution plan, and generates corresponding feedback optimization instructions. The feedback optimization instructions are used to dynamically adjust the material demand prediction model and the material distribution plan according to the actual needs of the construction site to ensure the balance between material supply and actual construction needs. The feedback optimization instructions are generated by the following calculation expression: F o =P c -A u Among them, F o Indicates feedback optimization instruction; P c Indicates the planned material usage; A u Indicates actual material usage.

8. The method for optimizing the amount of curtain wall material according to claim 7, characterized in that: The feedback optimization instructions are used to adjust the material distribution strategy in real time during the actual construction process to reduce material waste and improve material utilization. The adjustment process is achieved through continuous monitoring by a multi-level feedback analysis system.

9. The method for optimizing the amount of curtain wall material according to claim 1, characterized in that: A curtain wall material usage list is generated based on the final optimized material distribution data. The list includes material types, specifications, quantities and specific usage locations. Material procurement and construction execution are carried out according to the list. After the construction is completed, construction data records and project archiving data are generated and used for analysis and optimization of subsequent projects, providing a reference for future material demand forecasting and distribution optimization.

10. A system for implementing the method for optimizing the amount of curtain wall materials according to any one of claims 1 to 9, characterized in that: The system comprises: A multi-sensor module, which is used to be deployed at different positions of the building curtain wall to collect geometric shape data and material data of the curtain wall, and the multi-sensor module includes a laser radar, an ultrasonic sensor and a high-resolution photogrammetry device; A data processing module is used to perform data cleaning and noise filtering on the geometric shape data and material data received from the multi-sensor module to generate pre-processed curtain wall point cloud data and pre-processed curtain wall material data; An adaptive multi-layer feedback modeling module is used to receive the pre-processed curtain wall point cloud data and the pre-processed curtain wall material data, perform layered processing on them, optimize the model through an inter-layer feedback mechanism, and generate final three-dimensional model data containing complete geometric structure and material information; A hierarchical neural network prediction module is used to combine the final three-dimensional model data with environmental data, historical construction data and supply chain data, generate material demand prediction data through multi-level prediction, and generate optimized material distribution plan data through a collaborative optimization algorithm, wherein the material distribution plan data describes the type, quantity, specification and distribution location of the material; A multi-level feedback analysis module is used to collect construction progress data and actual material usage data in real time during the construction process, compare and analyze the data with the optimized material distribution plan data, generate feedback optimization instructions, and adaptively adjust the material prediction model and material distribution plan based on the feedback results to generate the final optimized material distribution data; The material management module is used to generate a curtain wall material usage list based on the final optimized material distribution data, and to purchase materials and execute construction according to the list, and to generate construction data records and project archiving data for subsequent analysis and optimization.

Images