A Construction Scene Modeling Method Based on Gaussian Process Regression

Abstract

The present invention discloses a method for constructing a construction scenario model based on Gaussian process regression, which relates to the technical field of construction management. The method includes: extracting the first-stage tasks from the target construction task, constructing the first task model, and modeling the construction scenario through Gaussian process regression. Dynamically predicting risks by gradually activating multiple Gaussian models according to the construction progress to generate a risk situation. Using the task model and risk situation of the first stage as a benchmark, generating an incremental model for the second stage through progressive modeling, and continuously activating the Gaussian model for risk prediction during the execution of the second stage. Gradually updating the construction risk situation of multiple stages through the construction progress, and outputting a dynamic construction situation cloud map. It solves the technical problem that the existing construction scenario modeling method cannot accurately predict the risk changes during the construction process in a complex construction environment, and achieves the technical effect of accurately predicting the risk changes during the construction process in a complex construction environment.

Description

Technical Field

[0001] This application relates to the technical field of construction management, and particularly to a construction scenario modeling method based on Gaussian process regression. Background Art

[0002] During the construction process, various factors such as construction progress, materials, and climate will affect the execution of construction tasks, and the construction site usually faces a complex and dynamically changing environment. Traditional construction risk prediction methods often rely on static models, lack dynamic adaptation to real-time data, and cannot effectively cope with the impact brought by on-site uncertainties. With the continuous improvement of the demand for intelligent management of building construction, the existing risk prediction models cannot meet the requirements of high precision and high real-time performance. Therefore, a new type of construction scenario modeling method is needed, which can accurately predict risks during the construction process in real time, so as to optimize the construction process and reduce potential losses caused by risks.

[0003] In the current related technologies, there is a technical problem that the construction scenario modeling method cannot accurately predict the risk changes during the construction process in a complex construction environment. Summary of the Invention

[0004] This application solves the technical problem that the existing construction scenario modeling method cannot accurately predict the risk changes during the construction process in a complex construction environment by providing a construction scenario modeling method based on Gaussian process regression.

[0005] This application provides a construction scenario modeling method based on Gaussian process regression, including:

[0006] Extract the first-stage tasks from the target construction task, and perform Gaussian process regression modeling on the construction scenarios of the first-stage tasks to generate a first-task model. Among them, the first-task model includes multiple first-stage Gaussian models; during the execution of the first-stage tasks, gradually activate the multiple first-stage Gaussian models in the first-task model according to the construction progress, and perform dynamic risk parallel prediction on the first-stage tasks to generate a first-stage risk situation; use the first-task model and the first-stage risk situation as the stage modeling benchmark, and perform construction scenario modeling on the second-stage tasks through stage-by-stage progressive modeling to generate a second incremental model; during the execution of the second-stage tasks, use the first-task model and the first-stage risk situation as the prediction starting point, and gradually activate multiple second-stage Gaussian models in the second incremental model according to the construction progress for dynamic risk prediction, and output the second-stage risk situation; and so on, perform dynamic prediction of the risk situation of the multi-stage construction scenario of the target construction task according to the construction progress, and output a construction situation cloud map.

[0007] A construction scenario modeling method based on Gaussian process regression proposed in this application first extracts the first-stage tasks from the target construction task, constructs the first task model, and models the construction scenario through Gaussian process regression. During the execution of the first-stage tasks, multiple Gaussian models are gradually activated according to the construction progress for dynamic risk prediction to generate a risk situation. Using the task model and risk situation of the first stage as a benchmark, a second-stage incremental model is generated through progressive modeling, and Gaussian models are continuously activated for risk prediction during the execution of the second stage. The construction risk situation of multiple stages is gradually updated through the construction progress, and a dynamic construction situation cloud map is output. By combining the Gaussian process regression model and selecting suitable kernel functions and mean functions for parameter adjustment, the technical effect of accurately predicting the risk changes during the construction process in a complex construction environment is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] To more clearly illustrate the technical solutions of the embodiments of the present disclosure, the accompanying drawings of the embodiments of the present disclosure will be briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the operations described above or below do not necessarily need to be executed precisely in sequence. On the contrary, according to the need, various steps can be executed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or several steps can be removed from these processes.

[0009] Figure 1 Schematic flowchart of a construction scenario modeling method based on Gaussian process regression provided by an embodiment of the present application;

[0010] Figure 2 Schematic flowchart of the output process of the risk situation in the first stage of a construction scenario modeling method based on Gaussian process regression provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0011] The above description is only an overview of the technical solutions of this application. In order to be able to understand the technical means of this application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of this application more obvious and understandable, the following specifically gives the detailed description of this application.

[0012] In order to make the purpose, technical solutions and advantages of this application clearer, the present application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations of this application. All other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of this application.

[0013] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments. However, it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict. The terms "first" and "second" involved are only used to distinguish similar objects and do not represent a specific order for the objects. The terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or server comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application.

[0014] The embodiments of this application provide a method for constructing a scenario model using Gaussian process regression, as Figure 1 shown. The method includes:

[0015] Step S100, extract the first-phase tasks from the target construction task and perform Gaussian process regression modeling on the first-phase tasks to generate a first-task model. Among them, the first-task model includes multiple first-phase Gaussian models. Specifically, first, extract the first-phase tasks from the target construction task to ensure that the tasks are divided according to the specific time arrangements and requirements in the construction plan. For example, a certain construction project involves multiple phases, such as earthwork, reinforced concrete pouring, and equipment installation, and each phase has different tasks and requirements. Model the first-phase tasks using the Gaussian process regression method. The modeling process is trained based on historical data and construction environment factors (such as weather, construction progress, resource usage, etc.) and can capture and quantify the potential risks and uncertainties in the construction process. For example, during the reinforced concrete pouring phase, it is affected by factors such as weather changes, personnel flow, or equipment failures. Then, based on Gaussian process regression, generate a first-phase task model, which includes multiple Gaussian models, and each model represents a different subtask in the first phase, such as foundation treatment, concrete pouring, formwork support, etc. The Gaussian model of each subtask makes predictions based on factors such as the progress and resource consumption of the actual task. For example, predict the impact of the delay in earthwork operations on subsequent construction through the Gaussian regression model.

[0016] In a possible implementation, the first-phase tasks are extracted from the target construction task, and Gaussian process regression modeling of the construction scenario of the first-phase tasks is performed to generate a first task model, where the first task model includes multiple first-phase Gaussian models. Step S100 further includes step S110 of optimizing the grid scale according to the construction characteristics of the first-phase tasks to obtain a first grid scale. Specifically, when optimizing the grid scale, it is first necessary to conduct a detailed analysis based on the construction characteristics of the first-phase tasks, including geographical and topographical conditions of the construction site, construction activity complexity, construction material distribution, and environmental factors. For example, deep foundation pit areas or high-altitude operation areas usually require more detailed grid division because the risks in the above areas are relatively high and the environmental changes are relatively large. When determining the grid scale, the key factors include spatial position differences, construction density, and environmental changes. For example, some areas may have intensive construction and require a higher grid resolution, while other areas can use a larger grid scale. In addition, environmental changes on the construction site, such as climate or soil changes, will also affect the selection of the grid scale. Through numerical optimization methods, the most suitable grid scale can be selected according to the risk prediction accuracy and calculation efficiency under different grid scales. For example, a smaller grid unit may be required in the foundation construction stage to accurately simulate the bearing capacity of the foundation, while a larger grid can be used in the exterior wall construction stage to improve efficiency. After selecting the preliminary grid scale, it is also necessary to verify it through actual construction data, compare the deviation between the model prediction and the actual situation. If the accuracy does not meet the requirements, adjust the grid density to ensure that the final grid scale can effectively improve the risk prediction accuracy.

[0017] Step S120, the first task model is processed into a plurality of first task pixels by using the first grid scale. Specifically, according to the selected first grid scale, the construction task area is first processed into a grid, and each grid unit represents a small area in the construction scene. For example, if the first grid scale is 1 square meter, the entire construction area will be divided into multiple 1-square-meter small grid units, and each unit becomes a task pixel. Each task pixel carries the construction characteristic data of the area, such as the construction progress, building materials used, climate conditions, status of construction equipment, etc. The area represented by the task pixel may include areas such as concrete pouring areas, steel bar binding areas, and foundation treatment areas, and the characteristics and construction status of each area are independently recorded. To ensure the accuracy of the data for each pixel, the data collection for each pixel includes not only sensor data at the construction site but also feedback from operators and external factors such as weather changes. The data is associated with information such as spatial location through a Gaussian regression model, so as to generate a dynamic risk prediction for each pixel. For example, a certain area may cause a delay in construction progress due to weather effects, or safety risks due to terrain complexity. The model can predict potential problems and provide early warnings to management personnel. Through gridification and data association, each area during the construction process can be monitored in real time, so as to achieve more efficient construction management and risk control. The dynamic data of the task pixels is updated as the construction progress advances, and through the iterative adjustment of the Gaussian process regression model, the prediction of construction risks always maintains accuracy.

[0018] Step S130, Gaussian process regression modeling is performed on the plurality of first task pixels to generate a plurality of first-stage Gaussian models. Specifically, during the implementation of Gaussian process regression modeling, first, the construction characteristic data of each first task pixel is prepared, including variables such as construction progress, weather conditions, and equipment status. Each pixel also contains specific spatial location data (such as area number, coordinates, etc.). After the data is standardized, it is passed as input to the Gaussian process regression model. During the modeling process, a suitable kernel function (such as a radial basis function or a Matérn kernel function) is selected to establish the correlation between different task pixels. For example, the area where a certain pixel is located may be greatly affected by climate change during the construction process, and the kernel function can capture the influence and perform modeling. Next, the model uses the training data to calculate the risk change trend of each pixel through regression analysis and outputs the Gaussian model of each task pixel. For example, the model may predict that the risk in a certain area will increase in a certain period in the future, and the reason may be that the equipment on which the area depends fails or external conditions change (such as rainfall). The Gaussian model will reflect the comprehensive influence of various factors such as construction progress, equipment condition, and environmental conditions, so as to provide dynamic risk prediction for construction management.

[0019] In a possible implementation, grid scale optimization is performed according to the construction characteristics of the first-phase task to obtain the first grid scale. Step S110 further includes step S111 of gridifying the first task model into multiple standard grid units based on the standardized grid scale. Specifically, in the process of gridifying the first task model based on the standardized grid scale, it is first clear that the goal of gridification is to decompose the complex construction scenario into multiple standard grid units, and each unit reflects the local characteristics of the construction area. For example, in the urban bridge construction scenario, the scale of the grid unit can be determined according to the pier spacing or the type of foundation to ensure that the fineness of the grid division can capture the local characteristics without adding too much computational burden. In actual operation, the space of the construction scenario is divided into uniform units according to the preset grid scale, and the units are given specific attributes, such as the spatial position of the construction area, topographic features, material strength, etc. Taking underground construction as an example, a grid unit may contain data such as foundation bearing capacity, concrete strength grade, and groundwater level change. In the construction of multi-story buildings, different grid units can respectively describe the material characteristics of the foundation area, the load distribution of the above-ground floors, and the support strength of the underground structure. When the grid unit is initialized, initial attribute values are given according to the known construction data, such as the initial bearing capacity of the foundation area and the foundation pit depth, and are continuously updated dynamically during the construction progress. The gridification process enables the entire construction model to be presented in a more intuitive and detailed manner, laying a solid foundation for subsequent Gaussian regression modeling and dynamic risk prediction, and at the same time improving the computational efficiency and applicability of the model.

[0020] Step S112: Collect the first construction characteristic data from the first-phase task, and calculate the characteristic gradient changes for the multiple standard grid units based on the first construction characteristic data to obtain multiple characteristic gradient units. Specifically, during the construction process, first collect relevant construction characteristic data from the first-phase task. The characteristic data includes terrain height, soil type, strength grade of construction materials, environmental temperature and humidity, foundation settlement amount, etc. For example, in bridge construction, the bearing capacity of the foundation, soil type, and settlement conditions are key characteristics, while in high-rise building construction, the strength grade of concrete, steel bar type, and wind load may be important construction characteristics. Allocate the characteristic data to multiple standard grid units according to the grid scale. Each grid unit represents a small area in the construction scene, carrying its specific spatial position and construction characteristics. For example, a certain grid unit may correspond to the bearing capacity value of the foundation, and another grid unit may contain the compressive strength of concrete. Then, based on the construction characteristic data of the grid units, calculate the characteristic gradient changes, which reflect the quantification of the characteristic differences between adjacent grid units and calculate the change rate of the differences. For example, during foundation construction, the foundation settlement amount may change significantly in some areas, and the areas with significant changes form gradient units. Gradient units are particularly crucial in the construction scene modeling, especially in the identification of uneven foundation settlement or high-risk areas. Finally, according to the changes in the characteristic gradients, screen out the areas with significant risks from the multiple characteristic gradient units through a preset threshold, record and mark the units. The marked units represent potential risk areas, which can provide accurate data support for subsequent risk prediction and Gaussian process regression modeling of the construction scene, ensuring that potential risks during the construction process are discovered in a timely manner and corresponding measures are taken.

[0021] Step S113: Preset a risk gradient threshold, and select N risk gradient units from the multiple characteristic gradient units according to the risk gradient threshold. Specifically, in order to accurately identify high-risk areas during the construction process, it is first necessary to set a risk gradient threshold. The threshold is usually based on construction characteristics, such as the change rate of parameters like soil settlement, concrete strength, or steel bar stress. For example, when the settlement rate of the soil is greater than a certain threshold, it may indicate the risk of uneven settlement, or when the steel bars show strength changes beyond the bearing range during the stress process, which may all lead to potential safety hazards during construction. When setting the threshold, the influence of the construction environment and historical data, such as geological environment, weather changes, etc., also needs to be considered. Then, based on the set risk gradient threshold, the system compares the calculated gradient change values of each construction area with the threshold and screens out the unit areas with excessive risks. Each unit represents a part of the construction site, and characteristics such as terrain, soil type, construction technology, etc. are all reflected in it. For example, if the settlement change rate of a specific area exceeds the threshold, this area will be marked as a high-risk unit. Through the screening mechanism, the system finally obtains N high-risk gradient units, and these areas need special attention. During the actual construction process, if a failure or accident occurs in an area, it will directly affect the construction progress and safety. Therefore, the selected high-risk units will be used as the basis for subsequent construction adjustments. Engineers can take measures such as reinforcement, adjusting the construction sequence, or other countermeasures to prevent potential construction risks and ensure the safe and smooth progress of the project.

[0022] Step S114: Use 1 / 2 of the standardized grid scale to grid the N risk gradient units into N groups of first updated grid units. Specifically, by using 1 / 2 of the standardized grid scale for gridding, the originally larger risk gradient units are subdivided into multiple smaller updated grid units, thereby improving the spatial resolution and enabling more refined capture of risk changes in the construction site. For example, in a building construction project, the original grid units cover a large area, and the risk changes in soil settlement are not obvious. However, by reducing the grid scale by half, the different settlement rates in local areas can be revealed in detail, helping to identify some potential hazard areas. The subdivided grid units not only reflect the spatial characteristics with higher resolution but also can better capture the minor risk changes that may occur during the construction process. By refining the grid, the risk of each small area can be predicted more accurately. For example, in areas with complex geological conditions, it will be found that the risk changes in some areas in the refined grid units are greater, so measures can be taken earlier to avoid potential safety problems. This method can provide higher prediction accuracy for the construction process and ensure more scientific risk assessment during the actual construction process.

[0023] Step S115: Calculate the characteristic gradient change for the N groups of first updated grid cells based on the first construction characteristic data to obtain N groups of updated gradient cells. Specifically, in the construction task, it is first necessary to collect construction characteristic data related to the first-phase task. The data includes soil type, material types (such as concrete grade, steel bar type), construction technology, environmental conditions (such as temperature, humidity, etc.). The characteristic data reflects the actual situation of the construction site. Next, calculate the characteristic gradient change for the N groups of first updated grid cells based on the data. The gradient calculation result of each grid cell represents the change rate of the construction characteristics in this area. For example, the change in soil density in different areas, or the difference in concrete strength at different construction locations. By calculating the gradient, the changes and potential risks in the construction process can be quantified. Subsequently, the calculated gradients are used to generate N groups of updated gradient cells, and the updated cells reflect the characteristic change situation at a higher-resolution grid scale.

[0024] Step S116: Select M second updated grid cells from the N groups of updated gradient cells according to the risk gradient threshold. Specifically, in the construction process, it is first necessary to set a risk gradient threshold, which is used to distinguish different risk areas of the construction site. The threshold is set according to factors such as soil density change, material strength, and terrain change. For example, if the soil density change in a certain area is too large, or the strength values of some concrete columns are lower than the standard requirements, then the risk of the above area is relatively high and needs to be processed preferentially. When calculating the gradient value of each grid cell, the risk change rate of each grid cell is obtained by collecting construction data and combining it with environmental data. Suppose there are some areas on the construction site where foundation settlement is caused by soil softening due to rainfall, and the calculated gradient value will reflect the change degree of this area. Then, compare the calculated gradient value with the set risk gradient threshold. Only when the gradient value is higher than the threshold is it considered a high-risk area. The grid cells with higher risk will be selected to form M second updated grid cells. The above areas need to be given special attention for reinforcement, monitoring, or re-planning. For example, if there is a strong settlement risk in an area during construction, the system will automatically screen out and mark this area to ensure that potential risk areas can be timely responded to and processed throughout the construction process.

[0025] Step S117, and so on, perform grid cell screening and update until the obtained updated grid cells meet the risk gradient threshold, and output the first grid scale. Specifically, during the construction process, first, according to the set risk gradient threshold, all possible high-risk areas are screened out. The high-risk areas are divided into multiple grid cells in a grid-like manner, where each grid cell represents an area with specific construction characteristics and risk gradient values. According to the gradient value, the system will identify the high-risk areas and give priority to their treatment. For example, if there is a high risk of settlement in a certain geological area, the system will mark this area as high-risk and include this grid cell in the screening scope. Next, the system will refine each screened high-risk grid cell, gradually reducing the grid scale to ensure that the risk areas can be accurately captured. If the risk gradient values of some areas are lower than the set threshold, they will be excluded in the subsequent screening, and only the areas with continuous high risk will be retained. This process is continuously iterated, and through repeated screening and update, the accuracy of risk assessment is gradually optimized. Finally, when the risk gradients of all grid cells meet the set threshold requirements, the final first grid scale is output. This scale accurately reflects the risk levels of each area during the construction process, providing effective support for subsequent construction decision-making, resource allocation, and risk response. For example, at a certain construction site, the risk of ground subsidence is high. After grid processing and scale refinement by the system, the areas that most need to be reinforced are determined and divided into high-risk areas, so as to give priority to reinforcement treatment to ensure the safety of construction.

[0026] In a possible implementation, Gaussian process regression modeling is performed on the multiple first task pixel points to generate multiple first-stage Gaussian models. Step S130 further includes step S131. After retrieving the first task execution time from the first-stage task, environmental data is collected according to the first task execution time to obtain first environmental prediction data. Specifically, before the construction task starts, the system first determines the start and end times of the construction based on the execution time of the first-stage task, which is crucial for subsequent environmental data collection. For example, if the first-stage task is foundation excavation, determining the execution time can provide a time window for the collection and analysis of relevant data. Then, the system collects relevant environmental data based on this time window, including climate information such as temperature, humidity, and wind speed, as well as geographical information related to the construction site such as soil humidity, groundwater level, and terrain height. For example, during the concrete pouring stage, changes in temperature and humidity will significantly affect the hardening process of the concrete, so collecting this data is crucial for the construction progress. Finally, the system processes and analyzes the collected environmental data to generate first environmental prediction data, which not only reflects the current environmental conditions but also predicts the environmental change trend in the future for a period of time. For example, based on real-time weather and historical data, the system may predict the precipitation in the next few days to make preparations for the construction in advance and avoid the adverse impact of rainfall on the construction progress and safety. Through the collection and prediction of these data, the system can provide support for risk assessment during the construction process, such as predicting the impact of extreme weather on the construction progress or warning of the landslide risk caused by excessive soil humidity.

[0027] Step S132, perform similar construction task matching based on the first construction characteristic data and the first environmental prediction data to obtain a first similar task model, where the first similar task model includes H sample Gaussian models for H sample pixel points, and the H sample pixel points have H sample pixel construction characteristic identifiers. Specifically, in the first stage of the construction task, the system first extracts construction characteristic data from the task, such as settlement control range, concrete grade, steel bar type, construction technology (such as formwork erection, pouring technology), and construction location (including terrain height, spatial coordinates, etc.). At the same time, environmental prediction data (such as temperature, humidity, wind speed, etc.) is also collected. These data will be used as a basis for the system to find similar construction tasks to the current task through similarity matching with historical data. This matching is based on the similarity metrics of multiple construction characteristics, such as concrete grade, settlement control range, etc., to ensure the selection of the most similar task scenario. Through the above method, the system generates a first similar task model, which contains multiple sample pixel points, each pixel point representing a construction area and carrying corresponding construction characteristic identifiers, such as design materials, construction technology, etc. Each sample pixel point generates a Gaussian model through Gaussian process regression, reflecting different characteristics and dynamic changes of the construction task. Through the Gaussian model, the system can dynamically predict the potential risks and performance of the construction task under various conditions, thus providing accurate data support for subsequent task execution and risk management.

[0028] Step S133, extract the multiple task construction characteristics of the multiple first task pixel points from the first stage task. Specifically, during the execution of the first stage task, the system extracts rich construction characteristic data from multiple task pixel points. Each pixel point represents an area of the construction site and contains various construction characteristics of that area. These characteristics include settlement control range, concrete grade, steel bar type, construction technology (such as concrete pouring technology or formwork installation technology), terrain height, spatial coordinates, etc. Taking the settlement control range as an example, in an earth excavation area, it may involve soil stability and settlement control measures; in a reinforced concrete structure area, the concrete grade and steel bar type are key construction characteristics. The construction technology may include pouring speed, formwork erection method, etc. in different areas. The characteristics of each area are clearly marked by identifiers. For example, a certain area may require a specific grade of concrete and steel bar type, while another area needs to consider different terrain heights and construction parameters. The extraction and identification of characteristic data provide basic data for the modeling and dynamic risk prediction of subsequent construction tasks, ensuring that the construction requirements and potential risks of each area can be accurately evaluated throughout the construction process.

[0029] Step S134, pre-construct a construction characteristic evaluation network. Specifically, in the process of pre-constructing the construction characteristic evaluation network, template pixel construction characteristics are first extracted from multiple construction tasks, covering multiple factors such as terrain, height, concrete grade, steel bar type, construction technology, settlement control range, and design materials, which reflect the physical and technical characteristics in different construction tasks. Then, all the collected characteristic data will be standardized to ensure comparison under a unified standard, thus eliminating the differences in different dimensions and units. For example, terrain changes may affect the concrete pouring process, while the steel bar type may affect the bearing capacity of the concrete. The standardized data will be used for characteristic combination and similarity measure calculation, and then the similarity between different construction tasks or regions will be evaluated. For example, the similarity between different construction characteristics is measured by calculating the Euclidean distance or cosine similarity. Then, through the similarity measure channel, the construction characteristics are mapped to a unified standard space to support subsequent task matching and Gaussian regression modeling. Finally, combining the characteristic similarity measure and the standardization channel, a construction characteristic evaluation network is constructed, which can analyze and evaluate the risks and adaptability of construction tasks in real time, helping the construction team effectively cope with various challenges in complex construction environments.

[0030] Step S135, after combining and enumerating the H sample pixel construction characteristics and multiple task construction characteristics, use the construction characteristic evaluation network to perform characteristic similarity measurement on the combination results to obtain multiple sample mapping relationships between the H sample pixel construction characteristics and multiple task construction characteristics. Specifically, first combine and enumerate the characteristics of multiple construction tasks with the construction characteristics of H sample pixels. The construction characteristics of each task cover multiple dimensions such as terrain, construction technology, and material selection, such as the grade of concrete, the type of steel bars, and the terrain height. By enumerating all characteristic combinations, a relatively large construction characteristic set can be generated to ensure the coverage of all potential construction scenarios. Then, use the pre-constructed construction characteristic evaluation network to perform similarity measurement on these enumerated characteristic combinations. The evaluation network measures the matching degree by calculating the similarity between the construction characteristics of each task and the sample pixel characteristics, such as using the Euclidean distance or cosine similarity. For example, if the concrete grade and steel bar type of a certain task are similar to the characteristics of a sample pixel, the network will output a relatively high similarity value. According to the similarity measurement results, mapping relationships between the construction characteristics of multiple tasks and the sample pixel characteristics are generated. The mapping relationships show which task construction characteristics are most similar to the sample pixel characteristics and provide a basis for subsequent construction scenario modeling.

[0031] Step S136: Migrate and adjust the H sample Gaussian models according to the multiple sample mapping relationships to generate the multiple first-stage Gaussian models adapted to the multiple first-task pixel points. Specifically, first, obtain multiple sample mapping relationships through the foregoing steps. These mapping relationships are established based on the similarity between the characteristics of the construction tasks and the characteristics of the sample pixels. Based on the mapping relationships, migrate and adjust the H sample Gaussian models. Specifically, the goal of the migration and adjustment is to enable these Gaussian models to adapt to the new construction task characteristics, especially the construction characteristics of the multiple first-task pixel points. Each task pixel point represents an area in the construction scene and has different construction characteristics, such as settlement control range, concrete grade, steel bar type, etc. The process of migration and adjustment includes adjusting the mean and covariance matrix of the model through mathematical methods based on the existing sample Gaussian models, so that the model can reflect the new construction characteristics. The adjustment not only retains the effectiveness of the original sample model but also enables the model to adapt to the new construction tasks, providing construction risk prediction for different regions. Finally, the adjusted multiple Gaussian models can accurately predict the risks that occur at each task pixel point during the actual construction process and effectively evaluate the construction progress of each region.

[0032] In a possible implementation, pre-construct a construction characteristic evaluation network. Step S134 further includes step S1341: interactively obtain the construction characteristics of multiple template pixels. Specifically, in the construction scene modeling, interactively obtaining the construction characteristics of multiple template pixels involves a detailed division of the construction area or task. Each division unit is a pixel point, representing the specific construction characteristics of the area. For example, the settlement control range may be obtained through geological exploration data, and information such as concrete grade and steel bar type is usually extracted from design drawings or construction reports, and the terrain height is obtained through on-site surveys or remote sensing technologies. Through real-time interaction with the site, combined with automatic sensor data and the construction management system interface, ensure that the characteristic data of each construction task area can be transmitted to the system in real time. After obtaining the characteristic data, the system performs standardization processing on the information. Categorical data (such as steel bar type, concrete grade) uses one-hot encoding, and continuous data (such as settlement range, terrain height) is processed through standardization methods, so that the data can enter the subsequent modeling stage uniformly. The points marked as the construction characteristics of template pixels contain a complete description of the construction tasks. Each construction characteristic of template pixels identifies various construction conditions and requirements of the tasks, including but not limited to settlement control, construction technology, material selection, task parameters, etc. In addition, the dynamic changes at the construction site will also affect the characteristic data. For example, changes in material selection or construction technology, so update these data through real-time feedback to ensure that the data throughout the construction process remains up-to-date and accurate. The continuous update method ensures the accuracy of the construction scene modeling and provides strong support for subsequent risk prediction and construction decision-making.

[0033] Step S1342, perform characteristic standardization annotation on the construction characteristics of the multiple template pixels to obtain multiple template characteristic vectors. Specifically, in the process of constructing the construction scenario model, it is first necessary to standardize the construction characteristics of the multiple template pixels to ensure that different types of data can be compared on the same scale. For continuous data, such as settlement range and terrain height, use the standardization formula to convert it into a value with zero mean and unit standard deviation. For example, for the settlement range data, its original values may be distributed in a large interval, and after standardization, each data point of the settlement range will be converted according to its mean and standard deviation, so that they fall within the same numerical range, thus avoiding the influence between different dimensions. For categorical data, such as concrete grade and steel bar type, it is processed by one-hot encoding or embedding vectors to ensure that discrete data can be converted into a digital format that can be processed by machine learning models. For example, the concrete grade C30 can be converted into [0,1,0] through one-hot encoding, and the steel bar type HRB400 is converted into the corresponding vector representation, and each category is mapped to a unique digital representation. The standardized characteristic data will be combined into a template characteristic vector, representing the construction characteristics of each task area. The vector will be used as the basis for subsequent construction task matching and modeling, enabling the construction characteristics of each task area to be compared and analyzed under a unified standard.

[0034] Step S1343, combinatorially enumerate the construction characteristics of the multiple template pixels to obtain F sets of template construction characteristics. Specifically, in the construction scenario modeling, first collect the construction characteristic data of multiple template pixels. Each pixel represents a specific area in the construction scenario and contains key characteristics of the area, such as settlement control range, material selection (such as concrete grade and steel bar type), construction process, terrain height, spatial coordinates, etc. For example, the settlement control range may be 1 - 5 mm, the concrete grade may have options of C30, C35, and C40, and the steel bar type is HRB400 or HRB500. After collecting the characteristics, next perform combinatorial enumeration to combine different characteristics into multiple possible construction scenarios. For instance, by combining the settlement control range, concrete grade, and steel bar type, multiple construction characteristic combinations can be obtained. The first combination is a settlement control range of 1 - 5 mm, a concrete grade of C30, and a steel bar type of HRB400, and the second combination is a settlement control range of 2 - 6 mm, a concrete grade of C35, and a steel bar type of HRB500. This process generates F sets of template construction characteristics by enumerating all possible combinations. Each set represents a specific construction scenario, and the combinations provide a basis for subsequent construction task matching and Gaussian regression modeling. For example, when the construction task requires judging applicable materials or processes based on these characteristics, the preset construction characteristic combinations can help optimize the construction plan and improve construction accuracy and efficiency. The F sets of template construction characteristics ultimately provide precise data support for the construction task, contributing to the dynamic adjustment and optimization of the risk prediction of the construction scenario.

[0035] Step S1344, perform similarity measurement on the F sets of template construction characteristics to obtain F template characteristic similarities. Specifically, when performing similarity measurement of template construction characteristics, first standardize the multiple template construction characteristics to eliminate the influence of different dimensions and units. For example, for continuous data (such as settlement range, terrain height), use the standardization processing formula , where is the data value after standardization processing, is the original data value, is the mean value, Let \(\sigma\) be the standard deviation to transform the data into a unified scale. For categorical data (such as concrete grade, steel bar type), the one-hot encoding or embedding vector representation method is used to convert it into a numerical form. Next, similarity measurement methods (such as Euclidean distance, Hamming distance, etc.) are used to calculate the similarity between the construction characteristics of each group of templates. For example, the settlement range and terrain height can be measured by the Euclidean distance, while the concrete grade and steel bar type can be measured by the Hamming distance. Through similarity calculation, the similarity values between each group of template characteristics and other group characteristics are obtained, and these values can provide decision-making support for construction tasks. Finally, the similarity values help to optimize the selection of construction plans, improve construction efficiency, reduce potential risks, and optimize the risk prediction of construction scenarios through a further Gaussian regression model to cope with unforeseen complex situations during the construction process.

[0036] Step S1345, train the feature standardization channel using the multiple template pixel construction characteristics and multiple template feature vectors. Specifically, to effectively process the multiple template pixel construction characteristic data, first collect the construction characteristic data related to the template. The data includes settlement control range, concrete grade, steel bar type, construction technology, terrain height, and spatial coordinates, etc. The characteristics represent multiple key factors in the construction process, such as construction methods, material selection, and site environment. Next, perform standardization processing on the collected construction characteristic data. For continuous data, such as the settlement control range and terrain height, use the standardization formula to transform the data into dimensionless data so that it can be compared on the same scale. For categorical data, such as concrete grade and steel bar type, use one-hot encoding or embedding vector representation to convert the categorical data into a numerical form for easy processing by the model. Then, based on the standardized characteristic data, construct a feature standardization channel. The goal of this channel is to learn how to uniformly transform different construction characteristic data into a standardized format through training. During the training process, input the standardized construction characteristic data and the corresponding feature vectors into the network. The system will adjust the weights through backpropagation so that when facing new construction characteristic data, it can automatically convert it into a standardized result. After training is completed, use new construction data for testing to evaluate the effect of the standardization channel and check whether it can accurately process different types of construction characteristic data in actual construction and improve the prediction accuracy.

[0037] Step S1346, train the similarity measurement channel using the construction characteristics of the F-group templates and the similarity of the F template characteristics. Specifically, to train the similarity measurement channel, first prepare F-group template construction characteristic data. Each group of template construction characteristics includes different data points related to construction, such as terrain height, concrete grade, steel bar type, construction technology, etc. Calculate the similarity between each group of template characteristics to form F template characteristic similarities, which represent the matching degree between different template construction characteristics. Then, based on the above data, construct a similarity measurement channel. This channel is trained through deep learning or machine learning algorithms to learn how to quantify the similarity between construction task characteristics. By inputting template construction characteristics and similarity data, use the backpropagation algorithm to adjust the network weights to optimize the accuracy of similarity prediction. During the training process, the system continuously adjusts the model parameters to improve the accuracy of similarity calculation, so that when new construction task data is input, the similarity measurement channel can evaluate the similarity between the new task and the existing templates according to the training results. For example, if a new construction task has similar terrain and material selection to a template task, the system can predict the risks and challenges that the task may face during construction through the similarity measurement channel, thus helping the project manager make more reasonable decisions. The well-trained similarity measurement channel can provide accurate similarity assessments for new construction tasks, effectively supporting risk prediction, resource allocation, and task planning during the construction process.

[0038] Step S1347, cascade the feature normalization channel and the similarity metric channel to complete the construction of the construction feature evaluation network. Specifically, in the process of constructing the construction feature evaluation network, first, the input construction feature data is normalized through the feature normalization channel, including normalizing continuous data such as settlement range and terrain height, and representing categorical data such as concrete grade and steel bar type using one-hot encoding or embedding vectors. The normalized data helps to eliminate the bias between data of different dimensions, making subsequent processing more consistent and accurate. Then, the normalized data is passed to the similarity metric channel. In this process, the system evaluates the matching degree between tasks by calculating the feature similarity between different tasks. The similarity metric channel will use deep learning or machine learning algorithms to calculate the similarity between tasks for construction features such as material selection and construction technology. For example, for a concrete pouring task in progress, the system can evaluate the potential risks during the construction process based on the similarity of historical construction tasks and provide real-time feedback. Then, the normalization channel and the similarity metric channel are cascaded so that the two can work together within a unified framework. The normalization channel first processes all input data and converts it into a consistent normalized format, and the similarity metric channel performs task matching and similarity measurement based on the normalized data, thereby achieving the accuracy of task evaluation. Finally, through training and optimization, the network can provide real-time support for the decision-making of construction tasks, help the construction team identify potential problems, optimize resource allocation, and predict risks during the construction process. For example, during deep foundation pit excavation, by evaluating construction features, the system can identify potential geological change problems in advance, thus providing effective risk warnings for the construction team.

[0039] In a possible implementation, multiple template pixel construction characteristics are obtained through interaction. Step S1341 further includes step S13411. The template pixel construction characteristics include settlement control range, concrete grade, steel bar type, construction technology, terrain height, and construction materials. Specifically, during the construction process, multiple template pixel construction characteristics directly affect the construction quality and progress. The key characteristics involved include settlement control range, concrete grade, steel bar type, construction technology, terrain height, and construction materials. First, the settlement control range involves the possible settlement changes of the soil or infrastructure during the construction process, which is crucial for ensuring the stability of the building. For example, in high-rise buildings, special attention needs to be paid to the settlement control range of the foundation to prevent the building from tilting or cracking due to excessive settlement. Secondly, the concrete grade refers to the strength grade of the concrete, which directly determines the compressive performance of the building structure. For example, in key structures such as high-rise buildings or bridges, high-strength concrete is usually required, which not only affects the scheduling of the construction process but also poses requirements for the material ratio and curing methods. The steel bar type is an important part of the building structure, which determines the type, specification, and layout of the steel bars, thereby affecting the stability of the load-bearing system. For example, high-strength steel bars may be required for complex structures or load-bearing parts, while ordinary steel bars are used for ordinary walls or non-load-bearing parts. The construction technology involves various technical methods from material selection to the actual construction process, which directly determines the construction quality and safety. For example, different mixing and pouring techniques during the concrete pouring process will directly affect the strength and durability of the concrete. The terrain height reflects the geographical conditions of the construction site. The undulation of the terrain may affect the design and construction process of the foundation. For example, when building in mountainous or hilly areas, special support structures or drainage system designs may be required. Finally, the construction materials cover all the raw materials used during the construction process. Different material types and qualities have important impacts on the construction time, cost, and later maintenance. For example, using fire-resistant and heat-insulating materials can improve the safety and energy efficiency of the building. All these characteristics comprehensively affect the decision-making and prediction models during the construction process. Through methods such as Gaussian regression for dynamic risk prediction, the smooth progress of the construction tasks can be ensured.

[0040] Step S200: During the execution of the first-phase task, according to the construction progress, gradually activate the multiple first-phase Gaussian models in the first task model, perform dynamic risk parallel prediction on the first-phase task, and generate the first-phase risk situation. Specifically, during the execution of the first-phase task, according to the construction progress, the system will gradually activate multiple Gaussian models in the first task model. These Gaussian models represent different risk factors that may occur during the construction process, such as environmental changes, material quality, equipment failures, etc. As the construction progress advances, the activation status of each model will change with different stages of the task, ensuring that these risks can be tracked and evaluated in real time. By processing these activated Gaussian models in parallel, the system can predict multiple potential risks simultaneously, ensuring a comprehensive risk assessment. During this process, the system generates a risk situation map for the first phase based on the output of the model. The risk situation map can display the risk levels at different time points or regions during the construction process. For example, when the construction progress approaches a specific node, the system will update the risk map for that phase, which may show an increase in risk in certain areas, such as abnormal changes in the settlement control range or non-compliance with the quality of construction materials in a specific area. This real-time update and dynamic adjustment of the risk situation enable construction managers to take countermeasures in a timely manner based on real-time data, avoiding possible delays or construction accidents. By gradually activating the model and dynamically updating the risk assessment, the system can effectively optimize the construction process and ensure the maximization of construction progress and quality.

[0041] In a possible implementation, as Figure 2 shown, during the execution of the first-phase task, according to the construction progress, gradually activate the multiple first-phase Gaussian models in the first task model, perform dynamic risk parallel prediction on the first-phase task, and generate the first-phase risk situation. Step S200 further includes step S210: preset the activation radius of adjacent pixels. Specifically, presetting the activation radius of adjacent pixels is a crucial step in construction scene modeling. The activation radius defines the influence range of each construction task pixel, which in turn affects the risk assessment of the surrounding area. In different construction tasks, the size of the activation radius needs to be adjusted according to the complexity of the task. For example, infrastructure construction tasks may require a larger activation radius to analyze the potential risks of the surrounding environment and other construction tasks, while small residential construction projects can set a smaller radius to reduce the calculation amount and focus on evaluating the main risk areas. During the construction process, as the project progresses and the construction scope expands, the activation radius should be dynamically adjusted to ensure that it gradually covers a wider range of potential risk areas, thereby improving the accuracy of risk prediction during the construction process and helping the project team to more effectively control the dynamic risk situation at each stage of the construction.

[0042] Step S220: According to the construction progress of the first-phase task, in the first-task model, gradually activate the multiple first-phase Gaussian models with the adjacent pixel activation radius as the constraint. Specifically, during the construction process, as the task progresses, the construction progress gradually advances, and the progress directly affects the risk prediction model of the construction area. To accurately capture the dynamic changes during the construction process, an adjacent pixel activation radius is set, and based on this constraint, the multiple first-phase Gaussian models are gradually activated. When the progress of the construction task reaches a predetermined node, the multiple Gaussian models in the first-task model will be gradually activated according to the construction progress, which not only considers the expansion of the construction area but also can reflect the gradual advancement of the construction task in real time. For example, in the initial stage of construction, the Gaussian model may only cover the core construction area, but as the task progresses, the model will gradually expand its influence range until it covers the entire construction site. By dynamically adjusting the activation radius and model range, the regional risk changes during the construction process can be more accurately reflected. When the progress of the construction task reaches a certain specific stage, the activated Gaussian model will further load the construction characteristics according to the real-time collected construction data to ensure that the task model can accurately evaluate the risk situation, thereby providing real-time and accurate support for construction decisions.

[0043] Step S230: Load the real-time collected construction data into the multiple first-phase Gaussian models that are gradually activated to obtain the multiple superimposed construction characteristic risk values of the multiple first-task pixel points. Specifically, during the construction process, the real-time collected construction data includes various environmental factors and construction dynamics, such as settlement data, equipment usage, environmental temperature, humidity changes, material properties, etc. The data is obtained in real time through sensors and monitoring systems and is loaded into the multiple first-phase Gaussian models that have been gradually activated. Each Gaussian model represents the construction characteristics of a small area (pixel point) in the construction site, such as foundation settlement, concrete strength, etc. During the data loading process, according to the changes in the real-time construction data, the construction characteristics and risk values of each pixel point are updated. Through the above method, not only can the risk value of each construction area be obtained, but also the problems that may occur during the construction process can be dynamically reflected, such as incomplete hardening of concrete due to temperature fluctuations or material expansion due to high humidity. Finally, the construction characteristics and risk values of multiple task pixel points are superimposed to generate a comprehensive construction risk situation, helping the construction team to monitor the risk changes in each area during the construction process in real time and adjust the construction strategy in a timely manner according to the prediction results. The above method ensures the safety and accuracy of the construction process and improves the risk management efficiency.

[0044] Step S240, use spatial interpolation to integrate the multiple superimposed construction characteristic risk values and output the risk situation in the first stage. Specifically, in the risk prediction during the construction process, spatial interpolation is used to integrate the construction characteristic risk values in different regions. When the risk values corresponding to each task pixel point of the construction task are generated, the scattered data often has discontinuities or blanks between regions. Especially in the actual construction environment, due to different equipment distributions, acquisition conditions, or construction characteristics, the risk data in some regions may be missing or incomplete. Therefore, using the spatial interpolation method to perform interpolation calculations on these risk data can smoothly transition the risk information of adjacent surrounding regions, enabling the originally isolated data points to be seamlessly connected to the overall model. The interpolation method not only fills the data gaps but also obtains a more accurate and continuous construction risk situation map by calculating the changes in risk values between adjacent regions. For example, in a construction area, the risk of certain construction processes is relatively high (such as deep foundation pit operations). After spatial interpolation of the risk values, they can be combined with the risk values of other surrounding processes to form a continuous risk distribution map. Finally, through further analysis and processing of the integrated risk values, the construction team can monitor the risk situation at the construction site in real time, accurately identify potential high-risk areas, such as settlement problems caused by soil inhomogeneity during the construction process in a certain area, and take timely measures to prevent the risk from expanding.

[0045] Step S300, use the first task model and the risk situation in the first stage as the stage modeling benchmark, and perform construction scenario modeling on the second-stage task through stage-by-stage progressive modeling to generate a second incremental model. Specifically, during the construction scenario modeling process, the task model and risk situation in the first stage are used as the benchmark, providing detailed data on the construction progress and potential risks in the first stage. The data reflects factors such as construction techniques, materials, and geological conditions during the construction process of the first-stage tasks such as foundation treatment and underground pipeline installation, and generates the corresponding risk situation. After completing the first-stage tasks, the information is used as the benchmark for the second-stage task modeling. The second-stage tasks may include structural construction or masonry work of the superstructure. Therefore, when modeling, in addition to inheriting the task model of the first stage, it is also necessary to analyze new construction techniques, materials, and technical requirements, etc. On this basis, through the stage-by-stage progressive modeling method, a second incremental model is generated to ensure that the construction scenario in the second stage can accurately reflect the characteristics of the new tasks and supplement the deficiencies in the first-stage model. The modeling process is not only based on the risk situation in the first stage but also can timely identify and address potential risks in the second stage to ensure the efficiency and safety of the construction process.

[0046] In a possible implementation, taking the first task model and the first-phase risk situation as the phase modeling benchmark, the construction scenario of the second-phase task is modeled through progressive phased modeling to generate a second incremental model. Step S300 further includes step S310 of extracting the second-phase task from the target construction task according to the task execution topological sequence. Specifically, in a construction project, the task execution topological sequence is used to clarify the sequence of different construction tasks, ensuring that each task is carried out after its dependencies are completed. In this process, first, the system needs to establish a topological sequence based on the dependencies between construction tasks, which means that the execution of each task must be carried out in a specific order. For example, the earth excavation task must be completed before the infrastructure laying task to ensure the smooth progress of subsequent work. Next, the system extracts the second-phase task from the target construction task. The task usually cannot start until the first phase is completed. For example, after the earth excavation is completed, it will involve infrastructure construction, installation of underground facilities, etc. The extraction of the second-phase task is not just a matter of sequence selection, but also takes into account the dependencies of the task on the first phase, ensuring that the second-phase task can only be started after the results of the previous phase are completed. For example, the construction of the structural framework can only start after the underground pipelines are laid. This process also requires the system to monitor the execution progress of the tasks to ensure that each task starts at the appropriate time and avoid construction delays caused by incorrect dependencies between tasks.

[0047] Step S320: Take the first task model as the benchmark for stage modeling, and perform construction scenario modeling according to the second-stage task to generate a second task model. Specifically, in construction scenario modeling, taking the first task model as the benchmark for stage modeling can utilize the construction characteristics, risk assessment, and prediction results generated by the first-stage task to provide support and guidance for the second-stage task. Specifically, the first task model contains the construction environment information, resource allocation, and risk situation of the first stage, and the above information provides a solid foundation for the implementation of the second-stage task. For example, in a certain construction project, the first-stage task involves foundation excavation and foundation treatment, and the task model of this stage details the soil settlement characteristics, construction equipment layout, and material usage. When entering the second-stage task, such as structural frame construction, the data of the first task model will be used as input to provide important references for the modeling of the second stage. At the same time, according to the specific construction characteristics of the second-stage task, such as the concrete grade required for the frame structure, the steel bar arrangement method, and the construction sequence, combined with the output results of the first task model, the construction scenario of the second stage is further refined. For example, the first-stage model may predict a high settlement risk in a certain area, and the second-stage model will adjust the concrete pouring volume or optimize the steel bar arrangement in this area to reduce the risk. In this process, the first task model not only provides a basic reference but also combines with the characteristics of the second-stage task through dynamic modeling to generate a targeted and forward-looking second task model, thus realizing the phased progression of construction scenario modeling. Such a modeling method not only ensures the close connection between the first stage and the second stage tasks but also lays a foundation for resource optimization and risk control in construction.

[0048] Step S330: Conduct an evolutionary difference analysis on the first task model and the second task model to obtain the second incremental model. Specifically, the process of evolutionary difference analysis begins with the clarification and refinement of the boundaries of the first task model and the second task model. The first task model usually describes the completed construction areas, such as the completed foundation pouring and the preliminary support structure, while the second task model covers all construction areas in the current stage, including the possible extended wall construction and frame erection parts. By comparing their spatial distributions and construction characteristics, the newly added areas in the second task model can be clearly determined. The newly added areas not only include the uncompleted construction parts but may also involve new construction characteristics, such as the need to use high-strength concrete, special types of steel bars, or the introduction of complex construction techniques to meet the increased load requirements. Subsequently, for the newly added areas, extract their construction characteristic data, such as material selection, terrain height, and design parameters, and update them in real-time in combination with the dynamic construction progress. For example, when the newly added area enters the frame welding stage, the model will dynamically reflect the changes in construction characteristics such as welding strength and weld positions. Finally, through the integrated analysis of the newly added areas, the second incremental model is generated. The above model accurately describes the construction characteristics and risk situations of the newly added areas, providing reliable data support and risk prediction guarantee for the next stage of construction, such as identifying high-risk areas or priority resource allocation points. The whole process ensures the combination of the dynamic and real-time nature of the model. Especially in multi-stage construction tasks, this analysis method effectively improves the accuracy of modeling and the scientific nature of decision-making.

[0049] Step S340, obtain the second grid scale by optimizing according to the construction task characteristics of the second-stage task. Specifically, to obtain the second grid scale by optimizing according to the construction task characteristics of the second-stage task, it is first necessary to clarify the range of characteristic data collection from the construction tasks. For example, in high-rise building construction, the characteristic data may include foundation load capacity, concrete pouring requirements, steel bar diameter distribution, and terrain height changes. Then, divide the entire construction area into initial standardized grid cells. For example, each grid cell is set to be 10 meters × 10 meters in size, which is used to quickly evaluate the basic characteristic distribution of the entire construction site. Subsequently, calculate the gradient changes of each grid cell according to the collected construction characteristic data, such as the settlement rate in different areas during the concrete pouring process or the change range of steel bar density. To ensure accuracy, screen out the grid cells with gradient changes higher than the preset threshold, such as identifying areas with significant foundation settlement or obvious differences in steel bar arrangement density. Further refine these high-gradient grid cells and reduce their size to 5 meters × 5 meters or smaller to improve the characteristic resolution. For example, in bridge construction, the high-gradient areas may be concentrated at the joints between the bridge piers and the girders. Refining these areas can ensure the modeling accuracy. After that, repeat the process of characteristic gradient calculation and screening until the refined grid cells meet the optimization conditions, such as the characteristic gradient change is lower than the set threshold or reaches the calculated resolution limit. Finally, generate the optimized grid scale, such as 1 meter × 1 meter, which is used to accurately capture the characteristic distribution within the construction area. This optimized grid scale provides a solid foundation for construction scene modeling and subsequent dynamic risk prediction, especially suitable for the refined management of complex construction areas, such as high-risk scenarios like large bridge foundations or subway tunnel excavations.

[0050] Step S350: Constrain by the second grid scale to construct the multiple second-stage Gaussian models for the multiple second-task pixel points in the second incremental model. Specifically, with the optimized second grid scale, the construction area is divided into multiple small grid units, and each unit corresponds to a pixel point. The pixel points form the basic units of the construction scenario. For example, in road construction, the construction area can be divided into grids of 2 meters × 2 meters, and each grid unit records the material properties, terrain height, and progress information of the area. Then, construction characteristic data, such as concrete strength grade, foundation bearing capacity, etc., are extracted for each pixel point, and these data will be used as the input for subsequent Gaussian process regression modeling. In the initialization stage, specific characteristic data, such as the foundation settlement rate, the real-time state of the construction progress, and the initial risk value, are assigned to each pixel point in the construction area to ensure complete input data for the Gaussian model. Subsequently, according to the construction characteristic data of the pixel points, a Gaussian process regression model is constructed for each pixel point separately. For example, in building foundation construction, the Gaussian model may predict the risk value of foundation settlement in a certain area. To capture the spatial correlation between pixel points, a radial basis function (RBF) kernel function is used for modeling, enabling the characteristics between adjacent pixel points to influence each other and ensuring that the model can reflect the continuity and consistency in the actual construction scenario. In addition, by adjusting the parameters of the kernel function (such as the length scale) and the mean function, the model is verified and optimized to accurately capture the dynamic construction risks. For example, in the foundation construction of high-rise buildings, adjusting the model parameters can more accurately predict the settlement degree of each foundation column. Finally, all Gaussian models are combined to form the second incremental model, comprehensively describing the dynamic characteristics and risk situation of the second-stage construction tasks. The incremental model can be used to show the risk changes in the key support areas during bridge construction, helping construction personnel adjust the construction strategy in real time.

[0051] Step S400, during the execution of the second-phase task, starting from the first task model and the first-phase risk situation, gradually activate multiple second-phase Gaussian models in the second incremental model according to the construction progress for dynamic risk prediction, and output the second-phase risk situation. Specifically, during the execution of the second-phase task, starting from the first task model and the first-phase risk situation, gradually activate multiple second-phase Gaussian models in the second incremental model in combination with the construction progress to achieve dynamic risk prediction. Specifically, the real-time monitoring of the construction progress can be achieved through the task completion ratio or time nodes. For example, in bridge construction, when the concrete pouring of the bridge pier reaches a certain ratio, the system will activate the Gaussian model corresponding to the bridge pier and associate the states of the surrounding pixels. Then, the real-time collected construction data, such as material status, weather changes, equipment operation parameters, etc., will be loaded into the corresponding Gaussian model for prediction analysis. For example, in high-rise building construction, the wind speed and the stability data of high-altitude construction equipment directly affect the risk value prediction, thus helping construction personnel to identify potential hidden dangers in advance. Subsequently, by integrating the output results of multiple Gaussian models, a comprehensive risk situation map is generated, such as marking the risk levels of different areas of the site: unstable foundation may be high risk, and relatively large stress on the support may be medium risk. Finally, the risk situation is output in the form of a dynamic cloud map, clearly presenting the risk changes and distributions in different areas of the construction site. For example, in subway tunnel construction, the cloud map can show the risk evolution of different soil layers in real time during the shield machine propulsion process, providing a decision-making basis for adjusting the construction plan and optimizing resource allocation.

[0052] Step S500, and so on, dynamically predict the risk situation of multi-stage construction scenarios of the target construction task according to the construction progress, and output a construction situation cloud map. Specifically, with the generation and dynamic prediction of the construction situation cloud map as the core, the whole process is gradually realized through the dynamic assessment of risks in multi-stage tasks. First, according to the construction progress, the Gaussian model corresponding to the stage is dynamically activated, based on the construction data collected in real time, such as equipment operation data, material properties, and weather information. In bridge construction, when the pile foundation is completed, the Gaussian model for pier construction is activated to evaluate risk areas, such as the settlement risk of the pier foundation or the risk of concrete cracking. Then, the risk situations of each stage are integrated into a global perspective through progressive analysis. For example, in tunnel construction, the soil layer stability during the advancement of the shield machine is used as the benchmark for track laying in the next stage. The construction situation cloud map integrates all risk prediction results through a spatial interpolation algorithm and displays the risk distribution in the construction area in the form of an intuitive cloud map. For example, in large dam construction, the cloud map intuitively shows the change in the pressure distribution of the dam body structure by color zoning, with high-risk areas marked in red and low-risk areas in green. In building construction, the cloud map dynamically shows the wind load risk of each floor structure of high-rise buildings, helping the construction party optimize resource allocation and construction strategies. Through the construction situation cloud map, the construction team can grasp the risk distribution and dynamic evolution of the whole stage in real time. For example, in water conservancy projects, it can timely identify the potential risks of flood impact on the under-construction dam body, guide the rapid adjustment of protection measures, so as to ensure construction safety and efficiency.

[0053] In the embodiment of the present application, the first-stage task is extracted from the target construction task, the first task model is constructed, and the construction scenario is modeled through Gaussian process regression. During the execution of the first-stage task, multiple Gaussian models are gradually activated according to the construction progress for dynamic risk prediction to generate a risk situation. Using the task model and risk situation of the first stage as a benchmark, an incremental model for the second stage is generated through progressive modeling, and Gaussian models are continuously activated during the execution of the second stage for risk prediction. The construction risk situation of multiple stages is gradually updated through the construction progress, and a dynamic construction situation cloud map is output, achieving the technical effect of accurately predicting the risk changes during the construction process in a complex construction environment by combining the Gaussian process regression model and selecting appropriate kernel functions and mean functions for parameter adjustment.

[0054] The above specific implementation manners do not constitute a limitation to the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present application shall be included within the protection scope of the present application.

Claims

1. A construction scenario modeling method based on Gaussian process regression, characterized in that, The method includes: Extracting the first-stage tasks from the target construction task, and performing Gaussian process regression modeling on the construction scenarios of the first-stage tasks to generate a first task model, where the first task model includes multiple first-stage Gaussian models; During the execution of the first-stage tasks, gradually activate the multiple first-stage Gaussian models in the first task model according to the construction progress, and perform dynamic risk parallel prediction on the first-stage tasks to generate a first-stage risk situation; Taking the first task model and the first-stage risk situation as the stage modeling benchmarks, perform construction scenario modeling on the second-stage tasks through phased progressive modeling to generate a second incremental model; During the execution of the second-stage tasks, taking the first task model and the first-stage risk situation as the prediction starting point, gradually activate the multiple second-stage Gaussian models in the second incremental model according to the construction progress for dynamic risk prediction, and output the second-stage risk situation; And so on, perform dynamic prediction of the multi-stage construction scenario risk situation of the target construction task according to the construction progress, and output a construction situation cloud map; Performing Gaussian process regression modeling on the construction scenarios of the first-stage tasks to generate a first task model, the method includes: Performing grid scale optimization according to the construction characteristics of the first-stage tasks to obtain a first grid scale; Using the first grid scale to grid the first task model into multiple first task pixels; Performing Gaussian process regression modeling on the multiple first task pixels to generate multiple first-stage Gaussian models; Taking the first task model and the first-stage risk situation as the stage modeling benchmarks, perform construction scenario modeling on the second-stage tasks through phased progressive modeling to generate a second incremental model, the method includes: Extracting the second-stage tasks from the target construction task according to the task execution topological sequence; Taking the first task model as the stage modeling benchmark, perform construction scenario modeling according to the second-stage tasks to generate a second task model; Performing evolutionary difference analysis on the first task model and the second task model to obtain the second incremental model; Optimizing according to the construction task characteristics of the second-stage tasks to obtain a second grid scale; Constrained by the second grid scale, construct the multiple second-stage Gaussian models of the multiple second task pixels in the second incremental model; During the execution of the first-stage tasks, gradually activate the multiple first-stage Gaussian models in the first task model according to the construction progress, and perform dynamic risk parallel prediction on the first-stage tasks to generate a first-stage risk situation, the method includes: Presetting an adjacent pixel activation radius; According to the construction progress of the first-stage tasks, in the first task model, gradually activate the multiple first-stage Gaussian models with the adjacent pixel activation radius as the constraint; Loading the real-time collected construction data into the gradually activated multiple first-stage Gaussian models to obtain multiple superimposed construction characteristic risk values of the multiple first task pixels; Using spatial difference to integrate the multiple superimposed construction characteristic risk values and output the first-stage risk situation.

2. The construction scenario modeling method based on Gaussian process regression according to claim 1, wherein Optimize the grid scale according to the construction characteristics of the first-stage task to obtain the first grid scale. The method includes: Grid the first task model into multiple standard grid cells based on the standardized grid scale; Collect the first construction characteristic data from the first-stage task, and calculate the characteristic gradient change of the multiple standard grid cells according to the first construction characteristic data to obtain multiple characteristic gradient cells; Preset a risk gradient threshold, and screen out N risk gradient cells from the multiple characteristic gradient cells according to the risk gradient threshold; Grid the N risk gradient cells into N groups of first updated grid cells using 1 / 2 of the standardized grid scale; Calculate the characteristic gradient change of the N groups of first updated grid cells according to the first construction characteristic data to obtain N groups of updated gradient cells; Screen out M second updated grid cells from the N groups of updated gradient cells according to the risk gradient threshold; And so on, perform grid cell screening and updating until the obtained updated grid cells meet the risk gradient threshold, and output the first grid scale.

3. The construction scenario modeling method based on Gaussian process regression according to claim 2, characterized in that, Perform Gaussian process regression modeling on the multiple first task pixel points to generate multiple first-stage Gaussian models. The method includes: After retrieving the first task execution time from the first-stage task, collect environmental data according to the first task execution time to obtain first environmental prediction data; Perform similar construction task matching according to the first construction characteristic data and the first environmental prediction data to obtain a first similar task model, where the first similar task model includes H sample Gaussian models of H sample pixel points, and the H sample pixel points have H sample pixel construction characteristic identifiers; Extract the multiple task construction characteristics of the multiple first task pixel points from the first-stage task; Pre-construct a construction characteristic evaluation network; After combining and enumerating the H sample pixel construction characteristics and the multiple task construction characteristics, use the construction characteristic evaluation network to measure the characteristic similarity of the combination results to obtain multiple sample mapping relationships between the H sample pixel construction characteristics and the multiple task construction characteristics; Perform migration adjustment on the H sample Gaussian models according to the multiple sample mapping relationships to generate the multiple first-stage Gaussian models adapted to the multiple first task pixel points.

4. The construction scenario modeling method based on Gaussian process regression according to claim 3, characterized in that, Pre-construct a construction characteristic evaluation network. The method includes: Interactively obtain multiple template pixel construction characteristics; Perform characteristic standardization annotation on the multiple template pixel construction characteristics to obtain multiple template characteristic vectors; Perform combination enumeration on the multiple template pixel construction characteristics to obtain F groups of template construction characteristics; Perform similarity measurement on the F groups of template construction characteristics to obtain F template characteristic similarities; Train a characteristic standardization channel using the multiple template pixel construction characteristics and the multiple template characteristic vectors; Train a similarity measurement channel using the F groups of template construction characteristics and the F template characteristic similarities; Cascade the characteristic standardization channel and the similarity measurement channel to complete the construction of the construction characteristic evaluation network.

5. The construction scenario modeling method based on Gaussian process regression according to claim 4, characterized in that, The construction characteristics of the template pixels include the settlement control range, concrete grade, steel bar type, construction technology, terrain height, and construction materials.

Images