Regional supervision scheduling method and system in intelligent construction

Abstract

The invention discloses a regional supervision scheduling method and system in intelligent construction, and relates to the technical field of equipment state monitoring, and the method comprises the steps: obtaining a three-dimensional model of a current construction site, carrying out the regional division of the three-dimensional model of the current construction site, determining a plurality of types of regions, the plurality of types of areas comprise a working area, a stored area and a storage prohibited area; material attribute information of a to-be-stored material is obtained, a plurality of candidate storage areas are determined according to the material attribute information and area constraint conditions, and the area constraint conditions are constructed based on the operation area, the stored area and the storage prohibited area; performing work efficiency, conflict influence and safety evaluation on the plurality of candidate storage areas, and determining a target storage area according to an evaluation result; and the to-be-stored materials are scheduled to the target storage area, and the three-dimensional model of the current construction site is dynamically updated. The technical problem of low material scheduling efficiency in the prior art is solved.

Description

Technical Field

[0001] This invention relates to the field of intelligent construction technology, specifically to a regional monitoring and scheduling method and system for intelligent construction. Background Art

[0002] Intelligent construction, as a core technology in the modern construction field, aims to improve construction efficiency and management through automation. Material scheduling, as a crucial link, directly impacts schedule, safety, and cost. Existing technologies primarily rely on human experience-based decision-making or static rule-based management systems, such as simply using two-dimensional planning software or pre-setting material storage locations. While these methods are effective for small-scale or low-variability projects, they struggle to meet the complex needs of large, dynamic construction sites. Their shortcomings include: First, scheduling decisions lack comprehensive analysis of three-dimensional spatial constraints, such as height restrictions, load-bearing capacity, or avoidance of restricted areas, leading to unreasonable material storage locations. For example, ignoring the actual distance between the work area and storage point may increase transportation time, or failure to dynamically avoid prohibited areas may cause safety violations. Second, they struggle to handle efficiency and conflict issues. Scheduling systems often neglect area usage frequency assessments; repeated occupation and clearing can cause conflicts in construction resources and real-time changes in work status, easily leading to transportation congestion, schedule delays, and resource waste. Third, safety mechanisms are weak, especially in emergency response. Existing technologies cannot automatically detect the risk of material storage obstructing rescue routes, increasing the probability of potential accidents. More fundamentally, static planning systems struggle to update their models in real time to adapt to changes in material usage, work progress, or external conditions, causing scheduling schemes to become detached from actual site conditions. This ultimately manifests as systemic problems such as low overall construction efficiency, frequent conflicts, and high safety risks.

[0003] To address the aforementioned shortcomings, this application proposes a regional monitoring and scheduling method for intelligent construction. This method integrates real-time 3D model zoning with intelligent evaluation to achieve dynamic optimization of material storage, thereby solving the technical problems of low material scheduling efficiency, frequent scheduling conflicts, and high safety risks in existing technologies. Summary of the Invention

[0004] This application provides a regional monitoring and scheduling method and system for intelligent construction, which is used to address the technical problems of low material scheduling efficiency, frequent scheduling conflicts and high safety risks in the prior art.

[0005] In view of the above problems, this application provides a regional monitoring and scheduling method and system for intelligent construction.

[0006] Firstly, this application provides a regional monitoring and scheduling method for intelligent construction, the method comprising: Obtain a 3D model of the current construction site, divide the 3D model of the current construction site into regions, and determine multiple types of regions, including work areas, already stored areas, and prohibited storage areas; Obtain the material attribute information of the material to be stored, and determine multiple candidate storage areas based on the material attribute information and regional constraints, wherein the regional constraints are constructed based on the work area, the already stored area, and the prohibited storage area; The multiple candidate storage areas are evaluated for work efficiency, conflict impact, and security, and the target storage area is determined based on the evaluation results. The materials to be stored are scheduled to the target storage area, and the current construction site 3D model is dynamically updated.

[0007] Secondly, this application provides a regional monitoring and scheduling system for intelligent construction, comprising: The area division module is used to acquire the current construction site 3D model, divide the current construction site 3D model into areas, and determine multiple types of areas, including work areas, already stored areas, and prohibited storage areas. The storage area confirmation module is used to obtain the material attribute information of the material to be stored, and determine multiple candidate storage areas based on the material attribute information and area constraints. The area constraints are constructed based on the work area, the already stored area, and the prohibited storage area. The regional assessment module is used to assess the work efficiency, conflict impact, and security of the multiple candidate storage regions, and determine the target storage region based on the assessment results. The material scheduling module is used to schedule the materials to be stored to the target storage area and dynamically update the current construction site 3D model.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages: This application proposes a regional monitoring and scheduling method and system for intelligent construction. By integrating 3D modeling, dynamic zoning, and multi-dimensional intelligent evaluation mechanisms, it significantly improves the scheduling accuracy, safety, and resource utilization efficiency of construction site materials. Compared with traditional technical solutions that rely on human experience or static rules, this application divides the site based on a real-time acquired 3D site model and intelligently filters candidate areas by combining material properties and regional rules. This effectively solves the conflict problem between material storage areas and work spaces, achieving global optimization of material scheduling without human intervention. This reduces resource conflict rates during construction, improves emergency response capabilities, and reduces ineffective handling, achieving the technical effects of global optimization of material scheduling, proactive suppression of safety risks, and systematic improvement of construction efficiency. It provides a highly robust scheduling and control method for intelligent construction. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 A flowchart illustrating a regional monitoring and scheduling method in intelligent construction provided in this application embodiment; Figure 2 This is a schematic diagram of the structure of a regional monitoring and scheduling system in intelligent construction, provided as an embodiment of this application.

[0011] The components represented by each number in the attached diagram are explained below: The module includes: 100 for area division, 200 for storage area confirmation, 300 for area evaluation, and 400 for material scheduling. Detailed Implementation

[0012] This application provides a regional monitoring and scheduling method and system for intelligent construction, which addresses the technical problems of low material scheduling efficiency, frequent scheduling conflicts, and high safety risks in existing technologies.

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0014] It should be noted that the terms "comprising" and "having" are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to these processes, methods, products, or devices.

[0015] Example 1, as Figure 1 As shown, this application provides a regional monitoring and scheduling method for intelligent construction, wherein the method includes: S10: Obtain the current construction site 3D model, divide the current construction site 3D model into regions, and determine multiple types of regions, including work areas, stored areas, and prohibited storage areas.

[0016] In traditional construction site management, the lack of continuous awareness of the three-dimensional spatial state leads to blurred boundaries between work and storage areas, often resulting in problems such as materials illegally occupying passageways, equipment interfering with work surfaces, or accidentally entering dangerous areas such as near high-voltage lines. Especially when multiple trades are working simultaneously, incomplete area information exacerbates the risk of site conflicts.

[0017] Step S10 in the method provided in this application embodiment includes: Extract the preset prohibited storage specifications, and set the prohibited storage area in the current construction site three-dimensional model according to the preset prohibited storage specifications; The construction operation status is obtained, and the construction operation space is identified in the current construction site 3D model based on the construction operation status to form the operation area; Detect the material occupancy status, and determine the material occupancy space in the current construction site three-dimensional model based on the material occupancy status to form the stored area; The prohibited storage areas, the work areas, and the already stored areas are summarized to form the multiple types of areas.

[0018] For example, a 3D model of the current construction site and preset prohibited storage guidelines are extracted from the current construction site database. These guidelines are pre-defined locations where construction materials are prohibited from being stored for production safety purposes, such as fire lanes, safety entrances and exits, dedicated equipment passages, and hazardous work areas. Based on these preset prohibited storage guidelines, prohibited spaces are set in the current construction site's 3D model to form prohibited storage areas.

[0019] To obtain the construction operation status, for example, a drone is used to scan the top view of the area, mark the area currently under construction, and mark the construction operation space in the current construction site 3D model according to the construction operation status, thus forming the operation area.

[0020] To check the occupancy status of materials, optionally, a drone can be used to scan an overhead view of the area and mark the areas where materials are currently stored. Among these areas, except for prohibited areas and work areas, areas with stored materials are considered stored areas, while the rest are not.

[0021] The prohibited storage areas, work areas, and areas where storage has already occurred are summarized to form multiple types of areas, enabling the confirmation of the current construction site type area.

[0022] In this embodiment, a 3D model of the current construction site is used to dynamically divide the work area, the already stored area, and the prohibited storage area, accurately constructing the spatial topology of the site. This method integrates construction status, material occupancy data, and safety regulations, laying a structured foundation for subsequent intelligent decision-making and avoiding site functional conflicts from the outset.

[0023] S20: Obtain the material attribute information of the material to be stored, and determine multiple candidate storage areas based on the material attribute information and regional constraints, wherein the regional constraints are constructed based on the working area, the already stored area, and the prohibited storage area.

[0024] Existing material scheduling systems suffer from three major shortcomings in the initial selection of storage locations: First, they only mechanically match the material foundation dimensions with site clearance, neglecting the dynamic envelope space of the equipment boom in the work area. For example, tower crane swings may encroach on unused areas, rendering candidate storage areas unusable. Second, they fail to consider the material's physical and chemical properties with the site's load-bearing characteristics. For instance, stacking large precast components in backfilled areas may cause settlement, creating structural hazards. Third, they ignore the temporal and spatial differences in prohibited storage rules. For example, areas permitted for storage during the day may become work areas at night. These deficiencies necessitate a secondary manual verification of the candidate storage areas output by the algorithm, extending the decision-making cycle and potentially generating invalid scheduling instructions in complex sites.

[0025] Step S20 in the method provided in this application embodiment includes: Based on the material property information, extract the size parameters, weight parameters, and quantity parameters of the material to be stored; Based on the work area, generate work interference constraints; based on the stored area, generate space occupancy constraints; based on the prohibited storage area, generate entry prohibition constraints. The region constraints are constructed by combining the operation interference constraints, space occupancy constraints, and no-entry constraints. In the remaining space of the current construction site 3D model, spatial matching is performed according to the size parameters, weight parameters and quantity parameters, and multiple storage subspaces that meet the regional constraints and pass the spatial matching are determined as the multiple candidate storage areas; Specifically, within the remaining space of the current construction site's 3D model, spatial matching is performed based on the size parameters, weight parameters, and quantity parameters. Multiple storage subspaces that satisfy the regional constraints and pass the spatial matching are identified as the multiple candidate storage areas, including: The remaining space is meshed according to the size parameters to generate multiple spatial mesh units; Retrieve the unit three-dimensional model of the material to be stored, and construct an adjustable material storage three-dimensional model by combining the quantity parameters and the maximum stacking height limit of the material to be stored. The material storage three-dimensional model is initially set according to the maximum stacking height, and is placed in a ground-level traversal according to the multiple spatial grid cells to form the first storage subspace set; The ground load-bearing capacity of each of the first storage subspaces in the first storage subspace set is verified, and the first storage subspaces that pass the ground load-bearing capacity verification are added to the plurality of storage subspaces; The stacking height of the material storage 3D model is sequentially reduced, and the ground-level traversal placement and ground bearing capacity verification of the storage subspace are performed. This process is repeated iteratively until the stacking height reaches the minimum stacking height, thus obtaining the multiple storage subspaces.

[0026] In this embodiment of the application, the size parameters, weight parameters, and quantity parameters of the material to be stored are extracted based on the material property information. The size parameters are parameters that measure the length, width, and height of the material to be stored, in meters; the weight parameters are parameters that measure the weight of the material to be stored, in kilograms; and the quantity parameters are parameters that measure the material to be stored, in pieces.

[0027] Based on the work area, generate work interference constraints; based on the already stored area, generate space occupancy constraints; and based on the prohibited storage area, generate no-entry constraints. A distance, such as 3 meters, is set outside the work area boundary as the work interference constraint; the space occupied by the already stored area is considered the stored area, serving as the space occupancy constraint; and the space occupied by the prohibited storage area is considered the prohibited storage area, serving as the no-entry constraint.

[0028] The constraints of operation interference, space occupancy and no-entry are combined and processed, and the spatial union of the three types of constraints is taken to construct the regional constraints.

[0029] Based on dimensional parameters, such as the length of the shortest side, the remaining space is meshed to generate multiple spatial mesh cells.

[0030] Retrieve a unit 3D model of the material to be stored. This 3D model is a scaled-down version of the unit material and has the same dimensional and weight parameters. Combining the quantity parameters and the maximum stacking height limit of the material, construct an adjustable 3D model for material storage. These 3D models are stacked according to the quantity parameters, with the stacking height ≤ the maximum stacking height.

[0031] The material storage 3D model is initially set according to the maximum stacking height, and then placed sequentially along the ground according to multiple spatial grid cells to form the first storage subspace set, in which multiple spatial grid cells satisfy the region constraint conditions.

[0032] The ground bearing capacity of each of the first storage subspaces in the first storage subspace set is verified. When the weight parameter of the material to be stored × the quantity parameter of the material to be stored is less than or equal to the maximum ground bearing capacity, the ground bearing capacity verification is passed. The first storage subspace that has passed the ground bearing capacity verification is added to the multiple storage subspaces.

[0033] If the ground load-bearing capacity verification fails, the stacking height of the 3D model of the material storage is reduced sequentially until the stacking height is less than or equal to the maximum stacking height. Then, the ground load-bearing capacity verification of the material storage subspace is performed by traversing the ground.

[0034] Repeat the placement and verification process iteratively until the minimum stacking height is reached, resulting in multiple storage subspaces.

[0035] In this embodiment, intelligent preliminary screening of candidate storage areas is achieved based on material properties such as material size, dynamic area constraints such as operational interference constraints, the space occupancy of quantified equipment activity intensity, and the space requirements of the materials to be stored. The method provided in this application can verify the load-bearing compatibility between the 3D model of the material storage and the remaining space to meet load-bearing limits, or automatically avoid planned excavation areas in the next 8 hours, outputting a set of physically acceptable, compliant, and safe candidate storage areas. The method provided in this application can reduce invalid candidate storage schemes and significantly improve decision-making efficiency.

[0036] S30: The multiple candidate storage areas are evaluated for work efficiency, conflict impact and security, and the target storage area is determined based on the evaluation results.

[0037] Traditional assessment methods suffer from a lack of dimensionality: for example, they rely solely on the shortest transportation distance as a single indicator, failing to quantify high-frequency material retrieval areas, such as the increased retrieval time that may result from temporary storage in the rebar processing area; or they lack dynamic modeling of safety factors, such as materials obstructing fire-fighting facilities or extending emergency response routes. This one-sided optimization leads the selected storage areas into an efficiency trap, potentially saving time on a single dispatch, but increasing overall costs due to frequent conflicts or safety rectifications.

[0038] Step S30 in the method provided in this application embodiment includes: Extract the first candidate storage region from the plurality of candidate storage regions; A first work efficiency coefficient is determined based on the transportation distance between the first candidate storage area and the work area; A first conflict impact coefficient is generated based on the usage frequency of the first candidate storage area within a preset historical period. The first conflict impact coefficient is generated based on the usage frequency of the first candidate storage area within a preset historical period, including: The preset historical period is uniformly discretized according to a preset time interval to generate multiple historical moment nodes; At each historical time point, the surveillance images of the first candidate storage area are retrieved to form a historical surveillance image set, and the total number of historical surveillance images in the historical surveillance image set is counted. The number of construction-free images generated is counted by identifying each historical monitoring image in the historical monitoring image set. Specifically, this involves identifying no construction traces in each historical monitoring image from the historical monitoring image set and counting the number of images without construction traces, including: Build a surveillance image recognizer and a construction trace verifier; The construction trace verification device includes: Collect a sample monitoring image set, annotate each sample monitoring image in the sample monitoring image set with construction traces, distinguish between images with construction traces and images without construction traces, and form a sample annotation set; The construction trace verifier is trained and generated based on the sample monitoring image set and the sample annotation set; Obtain an image of the idle state of the first candidate storage area and configure the monitoring image recognizer; Set a no-construction-trace counter, the initial value of which is the total number of historical monitoring images; The monitoring image recognizer is used sequentially to identify the first historical monitoring image in the historical monitoring image set to determine whether it is consistent with the idle state image; When the first historical monitoring image is inconsistent with the idle state image, the first historical monitoring image is input into the construction trace verifier for verification. If the verification confirms the existence of construction traces, the value of the no construction trace counter is decremented by 1. After the historical monitoring image set is identified, the final value of the no-construction-trace counter is obtained as the number of no-construction-trace images; The first conflict impact coefficient is obtained by calculating the ratio of the number of images without construction traces to the total number of historical monitoring images. An emergency impact analysis is performed on the first candidate storage area to determine the degree of impact of the first candidate storage area on emergency response and to generate a first safety factor. Specifically, an emergency response impact analysis is performed on the first candidate storage area to determine its impact on emergency response and generate a first safety factor, including: Determine the location coordinates of the entrance, exit, and work area of ​​the current construction site, and construct a preset rescue route from the entrance through the work area to the exit; Calculate the first passage time along the preset rescue path when there are no materials stored in the first candidate storage area; Simulate the second travel time along the preset rescue path after storing materials in the first candidate storage area; Calculate the difference between the second passage duration and the first passage duration, process the difference numerically, and take the reciprocal to obtain the first safety factor; The first work efficiency coefficient, the first conflict impact coefficient, and the first safety coefficient are weighted and summed according to preset weights to obtain the first evaluation result of the first candidate storage area. The evaluation process is repeated for the remaining candidate storage areas in turn to obtain multiple evaluation results. The candidate storage area with the best evaluation result is selected as the target storage area.

[0039] In this embodiment, a first candidate storage area is extracted from multiple candidate storage areas. A first work efficiency coefficient is determined based on the transportation distance between the first candidate storage area and the work area. The first work efficiency coefficient is the reciprocal of the transportation distance between the first candidate storage area and the work area. For example, if the transportation distance between the first candidate storage area and the work area is 10 meters, then the first work efficiency coefficient is 1 ÷ 10 = 0.1. The smaller the transportation distance, the greater the first work efficiency.

[0040] The preset historical period is uniformly discretized at time intervals to generate multiple historical time nodes. For example, the preset historical period is set to 168 hours, and the data is uniformly discretized at time intervals of 4 hours to generate multiple historical time nodes.

[0041] At each historical time point, retrieve the surveillance images from the first candidate storage area to form a historical surveillance image set, and count the total number of historical surveillance images in the historical surveillance image set.

[0042] The system identifies images without construction traces in each historical monitoring image set, counts the number of images without construction traces, and uses a construction trace verifier for identification.

[0043] Build a surveillance image recognizer and a construction trace verifier.

[0044] A surveillance image recognition system is constructed using a neural network. The system has a four-layer structure: an input layer with one neuron to receive the surveillance image to be recognized; a first hidden layer with 32 neurons using ReLU activation to scale the input image to the same size as the idle state baseline image; a second hidden layer with 64 neurons using ReLU activation to perform grayscale comparison on the data processed by the first hidden layer; and an output layer with one neuron using Linear activation to output a consistency judgment result, indicating whether the surveillance image to be recognized matches the idle state image. Sample surveillance images are collected and labeled as idle and non-idle, resulting in idle and non-idle sample surveillance images. These are then input into the surveillance image recognition system for supervised training. Parameters are adjusted and training continues until convergence is achieved, meaning the accuracy of the output in judging whether the image matches the idle state image reaches over 95%. This indicates that the surveillance image recognition system is successfully trained.

[0045] A construction trace verifier is constructed using a convolutional neural network. The construction trace verifier consists of four layers: an input layer, a first convolutional layer containing 32 3×3 convolutional kernels activated by the ReLU function, a second convolutional layer containing 64 3×3 convolutional kernels activated by the ReLU function, and an output layer using the Sigmoid function.

[0046] Collect a set of sample monitoring images, and annotate each sample monitoring image in the sample monitoring image set with construction traces. Distinguish between images with construction traces and images without construction traces, forming a set of annotated images with construction traces and a set of annotated images without construction traces.

[0047] Input the labeled sets of sample images with construction traces and the labeled sets of sample images without construction traces into the construction trace verifier for supervised training. Continuously adjust the parameters until the model converges, that is, the accuracy of the output images with and without construction traces reaches more than 95%, which means that the construction trace verifier training is complete.

[0048] Obtain the idle status image of the first candidate storage area and configure the monitoring image recognizer.

[0049] Set a counter for no construction traces. The initial value of the counter for no construction traces is the total number of historical monitoring images.

[0050] The monitoring image recognizer is used to identify the first historical monitoring image in the historical monitoring image set in turn, and to determine whether it is consistent with the idle state image.

[0051] When the first historical monitoring image is inconsistent with the idle state image, the first historical monitoring image is input into the construction trace verifier for verification. If the verification confirms the existence of construction traces, the value of the no construction trace counter is decremented by 1.

[0052] After the historical monitoring image set is identified, the final value of the no-construction-trace counter is obtained as the number of no-construction-trace images.

[0053] The first conflict impact coefficient is obtained by calculating the ratio of the number of images without construction traces to the total number of historical monitoring images. The first conflict impact coefficient = number of images without construction traces ÷ total number of historical monitoring images. For example, if the number of images without construction traces is 80 and the total number of historical monitoring images is 100, then the first conflict impact coefficient = 80 ÷ 100 = 0.8. The larger the first conflict impact coefficient, the more images without construction traces there are, and the lower the probability that the first candidate storage area may be affected by construction and therefore cannot be stored.

[0054] Determine the location coordinates of the entrance, exit, and work area of ​​the current construction site, and construct a pre-defined rescue route from the entrance, through the work area, to the exit.

[0055] Calculate the first passage time along the preset rescue route when there are no materials stored in the first candidate storage area.

[0056] The simulation measures the second travel time along the preset rescue path after materials are stored in the first candidate storage area.

[0057] Calculate the difference between the second and first passage times. After numerical processing, take the reciprocal of this difference to obtain the first safety factor. For example, if the second passage time is 20 minutes and the first passage time is 15 minutes, then the first safety factor = 1 ÷ |second passage time - first passage time| = 1 ÷ |20 - 15| = 0.2. When the first passage time equals the second passage time, the first safety factor is 1. A larger first safety factor indicates a smaller impact on the rescue route after materials are stored in the first candidate storage area, and thus a higher safety factor.

[0058] The first evaluation result of the first candidate storage area is obtained by weighting and summing the first work efficiency coefficient, the first conflict impact coefficient, and the first safety coefficient according to preset weights. The preset weights are pre-set weights reflecting the importance of each coefficient. The weight of the first work efficiency coefficient is α, the weight of the first conflict impact coefficient is β, and the weight of the first safety coefficient is γ, where α + β + γ = 1. For example, if α is set to 0.3, β to 0.3, γ to 0.4, the first work efficiency coefficient is 0.1, the first conflict impact coefficient is 0.8, and the first safety coefficient is 0.2, then the first evaluation result of the first candidate storage area = α × first work efficiency coefficient + β × first conflict impact coefficient + γ × first safety coefficient = 0.3 × 0.1 + 0.3 × 0.8 + 0.4 × 0.2 = 0.35.

[0059] The evaluation process is repeated for the remaining candidate storage areas in turn, resulting in multiple evaluation results. The candidate storage area with the best evaluation result, i.e. the largest evaluation result value, is selected as the target storage area.

[0060] This application's embodiments construct a three-dimensional evaluation system that combines a work efficiency coefficient weighted by transportation distance and material retrieval frequency, a conflict impact coefficient based on historical monitoring image analysis of area idle rate, and a safety coefficient that simulates changes in emergency passage time after material storage, thereby achieving globally optimal decision-making for candidate areas. By introducing time-dimensional parameters, such as the conflict coefficient which uses past idle state statistics and spatial dynamic simulation, and the safety coefficient which is calculated by extrapolating rescue path blockage scenarios, and employing a non-uniform weighting strategy for comprehensive evaluation, the construction delay rate can be significantly reduced, and the probability of major risk events can be lowered.

[0061] S40: The materials to be stored are moved to the target storage area, and the current construction site 3D model is dynamically updated; The current scheduling system suffers from a break in the execution-feedback chain: after a scheduling instruction is issued, the actual stacking location may deviate, such as storage personnel not storing materials according to precise coordinates, or changes in the material stacking shape, such as the natural collapse of bulk cargo stacks. This information cannot be quickly transmitted back to the management system, leading to the accumulation of errors between the 3D model and the actual site conditions, and a gradual increase in the error rate. Furthermore, scheduling instructions generated based on distorted models may exacerbate site chaos.

[0062] In this embodiment, the materials to be stored are precisely scheduled to the target storage area, and the current three-dimensional model of the construction site is dynamically updated. Optionally, the construction manager can upload the data, or a drone can be used to scan and set the target storage area as an already stored area, which facilitates subsequent planning.

[0063] In this embodiment, rapid synchronization of site status is achieved through a closed-loop linkage between scheduling commands and 3D model updates. The core mechanism lies in the real-time capture of material placement status while scheduling commands drive physical execution. Intelligent algorithms automatically update the spatial occupancy data of stored areas and trigger the next round of decision-making. This closed-loop mechanism ensures consistently high model fidelity, completely eliminating decision-making chain failures caused by outdated data, and forming an intelligent construction ecosystem where execution is the basis for correction.

[0064] Example 2, as Figure 2 As shown, based on the same inventive concept as the regional monitoring and scheduling method in intelligent construction provided in Embodiment 1, this embodiment of the invention also provides a regional monitoring and scheduling system in intelligent construction, including: The area division module 100 is used to acquire the current construction site three-dimensional model, divide the current construction site three-dimensional model into areas, and determine multiple types of areas, including work areas, stored areas and prohibited storage areas. The storage area confirmation module 200 is used to obtain the material attribute information of the material to be stored, and determine multiple candidate storage areas based on the material attribute information and area constraints. The area constraints are constructed based on the work area, the already stored area and the prohibited storage area. The area assessment module 300 is used to assess the work efficiency, conflict impact, and security of the multiple candidate storage areas, and determine the target storage area based on the assessment results. The material scheduling module 400 is used to schedule the materials to be stored to the target storage area and dynamically update the current construction site 3D model.

[0065] In one embodiment, the region partitioning module 100 is further configured to: Extract the preset prohibited storage specifications, and set the prohibited storage area in the current construction site three-dimensional model according to the preset prohibited storage specifications; The construction operation status is obtained, and the construction operation space is identified in the current construction site 3D model based on the construction operation status to form the operation area; Detect the material occupancy status, and determine the material occupancy space in the current construction site three-dimensional model based on the material occupancy status to form the stored area; The prohibited storage areas, the work areas, and the already stored areas are summarized to form the multiple types of areas.

[0066] In one embodiment, the storage area confirmation module 200 is further configured to: Based on the material property information, extract the size parameters, weight parameters, and quantity parameters of the material to be stored; Based on the work area, generate work interference constraints; based on the stored area, generate space occupancy constraints; based on the prohibited storage area, generate entry prohibition constraints. The region constraints are constructed by combining the operation interference constraints, space occupancy constraints, and no-entry constraints. In the remaining space of the current construction site 3D model, spatial matching is performed according to the size parameters, weight parameters and quantity parameters, and multiple storage subspaces that meet the regional constraints and pass the spatial matching are determined as the multiple candidate storage areas; Specifically, within the remaining space of the current construction site's 3D model, spatial matching is performed based on the size parameters, weight parameters, and quantity parameters. Multiple storage subspaces that satisfy the regional constraints and pass the spatial matching are identified as the multiple candidate storage areas, including: The remaining space is meshed according to the size parameters to generate multiple spatial mesh units; Retrieve the unit three-dimensional model of the material to be stored, and construct an adjustable material storage three-dimensional model by combining the quantity parameters and the maximum stacking height limit of the material to be stored. The material storage three-dimensional model is initially set according to the maximum stacking height, and is placed in a ground-level traversal according to the multiple spatial grid cells to form the first storage subspace set; The ground load-bearing capacity of each of the first storage subspaces in the first storage subspace set is verified, and the first storage subspaces that pass the ground load-bearing capacity verification are added to the plurality of storage subspaces; The stacking height of the material storage 3D model is sequentially reduced, and the ground-level traversal placement and ground bearing capacity verification of the storage subspace are performed. This process is repeated iteratively until the stacking height reaches the minimum stacking height, thus obtaining the multiple storage subspaces.

[0067] In one embodiment, the area assessment module 300 is further configured to: Extract the first candidate storage region from the plurality of candidate storage regions; A first work efficiency coefficient is determined based on the transportation distance between the first candidate storage area and the work area; A first conflict impact coefficient is generated based on the usage frequency of the first candidate storage area within a preset historical period. The first conflict impact coefficient is generated based on the usage frequency of the first candidate storage area within a preset historical period, including: The preset historical period is uniformly discretized according to a preset time interval to generate multiple historical moment nodes; At each historical time point, the surveillance images of the first candidate storage area are retrieved to form a historical surveillance image set, and the total number of historical surveillance images in the historical surveillance image set is counted. The number of construction-free images generated is counted by identifying each historical monitoring image in the historical monitoring image set. Specifically, this involves identifying no construction traces in each historical monitoring image from the historical monitoring image set and counting the number of images without construction traces, including: Build a surveillance image recognizer and a construction trace verifier; The construction trace verification device includes: Collect a sample monitoring image set, annotate each sample monitoring image in the sample monitoring image set with construction traces, distinguish between images with construction traces and images without construction traces, and form a sample annotation set; The construction trace verifier is trained and generated based on the sample monitoring image set and the sample annotation set; Obtain an image of the idle state of the first candidate storage area and configure the monitoring image recognizer; Set a no-construction-trace counter, the initial value of which is the total number of historical monitoring images; The monitoring image recognizer is used sequentially to identify the first historical monitoring image in the historical monitoring image set to determine whether it is consistent with the idle state image; When the first historical monitoring image is inconsistent with the idle state image, the first historical monitoring image is input into the construction trace verifier for verification. If the verification confirms the existence of construction traces, the value of the no construction trace counter is decremented by 1. After the historical monitoring image set is identified, the final value of the no-construction-trace counter is obtained as the number of no-construction-trace images; The first conflict impact coefficient is obtained by calculating the ratio of the number of images without construction traces to the total number of historical monitoring images. An emergency impact analysis is performed on the first candidate storage area to determine the degree of impact of the first candidate storage area on emergency response and to generate a first safety factor. Specifically, an emergency response impact analysis is performed on the first candidate storage area to determine its impact on emergency response and generate a first safety factor, including: Determine the location coordinates of the entrance, exit, and work area of ​​the current construction site, and construct a preset rescue route from the entrance through the work area to the exit; Calculate the first passage time along the preset rescue path when there are no materials stored in the first candidate storage area; Simulate the second travel time along the preset rescue path after storing materials in the first candidate storage area; Calculate the difference between the second passage duration and the first passage duration, process the difference numerically, and take the reciprocal to obtain the first safety factor; The first work efficiency coefficient, the first conflict impact coefficient, and the first safety coefficient are weighted and summed according to preset weights to obtain the first evaluation result of the first candidate storage area. The evaluation process is repeated for the remaining candidate storage areas in turn to obtain multiple evaluation results. The candidate storage area with the best evaluation result is selected as the target storage area.

[0068] In summary, the embodiments of this application have at least the following technical effects: This application proposes a regional monitoring and scheduling method and system for intelligent construction. By integrating 3D modeling, dynamic zoning, and a multi-dimensional intelligent evaluation mechanism, it significantly improves the scheduling accuracy, safety, and resource utilization efficiency of construction site materials. Compared to traditional technical solutions that rely on manual experience or static rules, this application partitions the site based on a real-time acquired 3D site model and intelligently selects candidate storage areas by combining material properties and regional rules, effectively solving the problem of material storage and work space conflicts. By introducing an evaluation system that includes transportation distance, historical usage frequency, and emergency route impact analysis, it achieves a leap from spatial adaptation to behavioral adaptation, dynamically avoiding occupancy conflicts and emergency route blockage risks, significantly reducing the probability of construction interruption and safety hazards. At the same time, through gridded load-bearing verification and an adjustable stacking model, it maximizes the utilization rate of site volume while meeting spatial constraints, avoiding the risk of over-stacking. It also utilizes a closed-loop update mechanism, such as automatically refreshing the occupancy status after material scheduling, to achieve rapid response to dynamic changes in the site, solving the scheduling failure problem caused by model lag in traditional methods. Ultimately, this application achieves global optimization of material scheduling without human intervention, resulting in reduced resource conflict rate, improved emergency response capability, and reduced ineffective handling during the construction process, providing a highly robust scheduling and control method for intelligent construction.

[0069] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0070] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0071] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A regional monitoring and scheduling method for intelligent construction, characterized in that, The method includes: Obtain a 3D model of the current construction site, divide the 3D model of the current construction site into regions, and determine multiple types of regions, including work areas, already stored areas, and prohibited storage areas; Obtain the material attribute information of the material to be stored, and determine multiple candidate storage areas based on the material attribute information and regional constraints, wherein the regional constraints are constructed based on the work area, the already stored area, and the prohibited storage area; The multiple candidate storage areas are evaluated for work efficiency, conflict impact, and security, and the target storage area is determined based on the evaluation results. The materials to be stored are scheduled to the target storage area, and the current construction site 3D model is dynamically updated.

2. The method according to claim 1, characterized in that, Obtain a 3D model of the current construction site, divide the 3D model into regions, and determine multiple types of regions, including work areas, already stored areas, and prohibited storage areas, including: Extract the preset prohibited storage specifications, and set the prohibited storage area in the current construction site three-dimensional model according to the preset prohibited storage specifications; The construction operation status is obtained, and the construction operation space is identified in the current construction site 3D model based on the construction operation status to form the operation area; Detect the material occupancy status, and determine the material occupancy space in the current construction site three-dimensional model based on the material occupancy status to form the stored area; The prohibited storage areas, the work areas, and the already stored areas are summarized to form the multiple types of areas.

3. The method according to claim 1, characterized in that, Obtain the material attribute information of the materials to be stored, and determine multiple candidate storage areas based on the material attribute information and regional constraints. The regional constraints are constructed based on the work area, already stored areas, and prohibited storage areas, and include: Based on the material property information, extract the size parameters, weight parameters, and quantity parameters of the material to be stored; Based on the work area, generate work interference constraints; based on the stored area, generate space occupancy constraints; based on the prohibited storage area, generate entry prohibition constraints. The region constraints are constructed by combining the operation interference constraints, space occupancy constraints, and no-entry constraints. In the remaining space of the current construction site 3D model, spatial matching is performed based on the size parameters, weight parameters, and quantity parameters to determine multiple storage subspaces that meet the regional constraints and pass the spatial matching as multiple candidate storage areas.

4. The method according to claim 3, characterized in that, In the remaining space of the current construction site 3D model, spatial matching is performed based on the size parameters, weight parameters, and quantity parameters. Multiple storage subspaces that satisfy the regional constraints and pass the spatial matching are identified as the multiple candidate storage areas, including: The remaining space is meshed according to the size parameters to generate multiple spatial mesh units; Retrieve the unit three-dimensional model of the material to be stored, and construct an adjustable material storage three-dimensional model by combining the quantity parameters and the maximum stacking height limit of the material to be stored. The material storage three-dimensional model is initially set according to the maximum stacking height, and is placed in a ground-level traversal according to the multiple spatial grid cells to form the first storage subspace set; The ground load-bearing capacity of each of the first storage subspaces in the first storage subspace set is verified, and the first storage subspaces that pass the ground load-bearing capacity verification are added to the plurality of storage subspaces; The stacking height of the material storage 3D model is sequentially reduced, and the ground-level traversal placement and ground bearing capacity verification of the storage subspace are performed. This process is repeated iteratively until the stacking height reaches the minimum stacking height, thus obtaining the multiple storage subspaces.

5. The method according to claim 1, characterized in that, The multiple candidate storage areas are evaluated for work efficiency, conflict impact, and security, and the target storage area is determined based on the evaluation results, including: Extract the first candidate storage region from the plurality of candidate storage regions; A first work efficiency coefficient is determined based on the transportation distance between the first candidate storage area and the work area; A first conflict impact coefficient is generated based on the usage frequency of the first candidate storage area within a preset historical period. An emergency impact analysis is performed on the first candidate storage area to determine the degree of impact of the first candidate storage area on emergency response and to generate a first safety factor. The first work efficiency coefficient, the first conflict impact coefficient, and the first safety coefficient are weighted and summed according to preset weights to obtain the first evaluation result of the first candidate storage area. The evaluation process is repeated for the remaining candidate storage areas in turn to obtain multiple evaluation results. The candidate storage area with the best evaluation result is selected as the target storage area.

6. The method according to claim 5, characterized in that, Based on the usage frequency of the first candidate storage area within a preset historical period, a first conflict impact coefficient is generated, including: The preset historical period is uniformly discretized according to a preset time interval to generate multiple historical moment nodes; At each historical time point, the surveillance images of the first candidate storage area are retrieved to form a historical surveillance image set, and the total number of historical surveillance images in the historical surveillance image set is counted. The number of construction-free images generated is counted by identifying each historical monitoring image in the historical monitoring image set. The first conflict impact coefficient is obtained by calculating the ratio of the number of images without construction traces to the total number of historical monitoring images.

7. The method according to claim 6, characterized in that, The system identifies no construction traces in each historical monitoring image in the historical monitoring image set and counts the number of images without construction traces, including: Build a surveillance image recognizer and a construction trace verifier; Obtain an image of the idle state of the first candidate storage area and configure the monitoring image recognizer; Set a no-construction-trace counter, the initial value of which is the total number of historical monitoring images; The monitoring image recognizer is used sequentially to identify the first historical monitoring image in the historical monitoring image set to determine whether it is consistent with the idle state image; When the first historical monitoring image is inconsistent with the idle state image, the first historical monitoring image is input into the construction trace verifier for verification. If the verification confirms the existence of construction traces, the value of the no construction trace counter is decremented by 1. After the historical monitoring image set is identified, the final value of the no-construction-trace counter is obtained as the number of no-construction-trace images.

8. The method according to claim 7, characterized in that, Constructing a construction trace verification tool includes: Collect a sample monitoring image set, annotate each sample monitoring image in the sample monitoring image set with construction traces, distinguish between images with construction traces and images without construction traces, and form a sample annotation set; The construction trace verifier is trained and generated based on the sample monitoring image set and the sample annotation set.

9. The method according to claim 5, characterized in that, An emergency response impact analysis is performed on the first candidate storage area to determine its impact on emergency response and generate a first safety factor, including: Determine the location coordinates of the entrance, exit, and work area of ​​the current construction site, and construct a preset rescue route from the entrance through the work area to the exit; Calculate the first passage time along the preset rescue path when there are no materials stored in the first candidate storage area; Simulate the second travel time along the preset rescue path after storing materials in the first candidate storage area; Calculate the difference between the second passage duration and the first passage duration, process the difference numerically, and take the reciprocal to obtain the first safety factor.

10. A regional monitoring and scheduling system for intelligent construction, characterized in that, The system is used to implement the regional monitoring and scheduling method for intelligent construction according to any one of claims 1-9, the system comprising: The area division module is used to acquire the current construction site 3D model, divide the current construction site 3D model into areas, and determine multiple types of areas, including work areas, already stored areas, and prohibited storage areas. The storage area confirmation module is used to obtain the material attribute information of the material to be stored, and determine multiple candidate storage areas based on the material attribute information and area constraints. The area constraints are constructed based on the work area, the already stored area, and the prohibited storage area. The regional assessment module is used to assess the work efficiency, conflict impact, and security of the multiple candidate storage regions, and determine the target storage region based on the assessment results. The material scheduling module is used to schedule the materials to be stored to the target storage area and dynamically update the current construction site 3D model.

Images