Construction management system and method for complex construction scene

Abstract

The invention relates to a construction management system and method for a complex construction scene, and relates to the technical field of construction management, and the system comprises a monitoring module which obtains the construction progress and construction monitoring information of the construction scene in real time, and a digital twinborn model recognition module which builds a three-dimensional digital twinborn model and determines the risk recognition result of equipment and personnel; the decision module determines a plurality of initial management strategies based on the risk identification result and a pre-constructed construction safety knowledge graph; and based on the construction progress, the construction plan, the environment information and a preset multi-objective optimization algorithm, optimizing the plurality of initial management strategies, determining an optimal management strategy, and sending the optimal management strategy to a corresponding management terminal in the optimal management strategy. The method has the effects of effectively identifying the construction risk, optimizing the management strategy, guaranteeing the construction safety and improving the construction efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of construction management, and in particular to a construction management system and method for complex construction scenes. BACKGROUND

[0002] In the field of building construction, with the rapid advancement of urbanization, various complex construction projects are emerging, and construction scenes are increasingly complex and diverse. This not only involves the coordinated operation of numerous large mechanical equipment, such as tower crane equipment, bucket wheel machines, excavators, and construction elevators, but also requires the cooperation of a large number of construction personnel. Effective management in terms of safety management, progress control, and equipment maintenance during the construction process is crucial to ensuring the smooth progress of the project, as it directly affects construction efficiency, personnel safety, and the overall quality of the project.

[0003] Existing construction management mainly relies on on-site patrols and experience-based judgments. For personnel management, it is difficult to obtain accurate personnel positions and detailed operation information in real time, mainly relying on on-site observation of personnel working conditions. For equipment management, it is based on reports from operating personnel and periodic equipment maintenance records to assess equipment conditions. In terms of safety risk identification, it relies on the experience and simple checks of management personnel, lacking a scientific and systematic approach.

[0004] Therefore, there is an urgent need for a construction management solution suitable for complex construction scenes to improve construction safety and efficiency. SUMMARY

[0005] The technical problem to be solved by the present application is to provide a construction management system and method for complex construction scenes, aiming to solve at least one of the above technical problems.

[0006] The technical solution of the present application to solve the above technical problems is as follows: In a first aspect, the present application provides a construction management system for complex construction scenes, adopting the following technical solution: A construction management system for complex construction scenes, comprising: a monitoring module for real-time acquisition of construction progress and construction monitoring information of the construction scene, the construction monitoring information including personnel information, equipment operating parameters, construction site image information, and environmental information, the personnel information including personnel positioning information, operation information, and safety helmet wearing information, the construction scene having a plurality of equipment including tower crane equipment, bucket wheel machines, excavators, and construction elevators; a digital twin model identification module for establishing a three-dimensional digital twin model of the construction scene and the equipment within the construction scene, integrating the construction progress, construction monitoring information, and equipment operation logic of the construction scene in the three-dimensional digital twin model; The digital twin model identification module is further configured to determine, for each of the devices, safety boundary information of the device based on the operation logic of the device, the operation parameter of the device, the environment information, and a preset risk algorithm; determine a risk identification result of the personnel based on the personnel information and the safety boundary information of each of the devices, the risk identification result including a risk level and a risk type; and determine a risk identification result of each of the devices based on the operation parameter of the device. The decision module is configured to determine a plurality of initial management strategies based on the risk identification result and a pre-constructed construction safety knowledge graph; optimize the plurality of initial management strategies based on the construction progress, the construction plan, the environment information, and a preset multi-objective optimization algorithm, to determine an optimal management strategy; and send the optimal management strategy to a management terminal corresponding to the optimal management strategy.

[0007] The present application has the following beneficial effects: by constructing a three-dimensional digital twin model that integrates real-time monitoring data and device operation logic, full-factor digital mapping and dynamic risk perception of complex construction scenes are achieved; the system can automatically identify the safety boundaries of personnel and devices based on multi-source heterogeneous data, and through the collaborative decision of the knowledge graph and the multi-objective optimization algorithm, an optimal management strategy that takes into account safety, progress, and cost is generated, the active safety early warning capability and the fine management level of the construction process are improved, and the safety accident rate is reduced.

[0008] Based on the above technical solutions, the present application can be further improved as follows.

[0009] Further, the system further comprises: The construction deduction and prediction module is configured to drive the three-dimensional digital twin model to perform forward construction deduction based on the construction progress of the current period, the optimal management strategy, and the construction plan and meteorological prediction data of the next work period of the current period, to predict the key path progress risk and the dynamic safety risk distribution map in the next work period. The decision module is further configured to determine the optimal scheduling strategy of the devices and the personnel in the next work period based on the key path progress risk, the dynamic safety risk distribution map in the next work period, and a preset optimization algorithm.

[0010] The beneficial effects of the above further solutions are that the construction deduction and prediction module can simulate the construction process of the next period in the digital twin environment based on the current progress, the optimal strategy, and future meteorological conditions, identify the progress bottlenecks on the key path and the dynamic safety risk of the whole field in advance, so that the decision module can develop a forward-looking optimal scheduling strategy according to the prediction result, thereby changing the risk control from passive response to active early warning and avoidance, and reducing the construction risk.

[0011] Further, the system further comprises: a personnel training module configured to obtain historical operation behaviors, risk records, and historical simulation information of the personnel, generate a simulation training task based on the historical operation behaviors, the risk records, the historical simulation information, and a scheduling strategy of the personnel in a next work cycle, determine a target case from the construction safety knowledge graph based on the simulation training task, and generate a virtual training scene in the three-dimensional digital twin model based on the target case, the virtual training scene including a virtual scene environment, a virtual device state, a virtual fault setting, a virtual risk trigger logic, and a plurality of training contents; During the personnel training, the personnel training module is configured to obtain multi-modal operation data of the personnel in the virtual training scene in real time, determine whether the personnel have operation errors based on a matching result of the multi-modal operation data and a standard operation model, obtain a historical operation frequency, a historical operation error frequency, a historical occurrence probability of the training content in a real construction scene, an impact level, and a relevance of the training content to a construction task in the next work cycle of the personnel on the training content if the personnel have operation errors, determine a priority of the training content based on the historical operation frequency, the historical operation error frequency, the historical occurrence probability of the training content in the real construction scene, the impact level, the relevance of the training content to the construction task in the next work cycle, and a preset weighting calculation method, and select a trigger content, a timing, and an intensity from a set of priority risks.

[0012] The personnel training module generates a highly realistic and targeted virtual training scene in a digital twin environment based on individual historical data and future tasks of the personnel, and performs accurate evaluation through real-time collection of multi-modal operation data. When an operation error is found, the personnel training module dynamically fuses the historical occurrence frequency, the actual risk level, and the relevance to subsequent tasks of the error to intelligently determine training focus and intensity in a weighted calculation manner, thereby realizing personalized training content and improving the real operation skills and risk response ability of the personnel in complex scenarios.

[0013] Further, the system further comprises: a device maintenance module configured to determine a plurality of devices to be maintained based on risk identification results, historical maintenance records, and real-time health monitoring data of each of the devices; determine a first maintenance weight of each of the devices to be maintained based on a criticality of a construction task of each of the devices to be maintained in a next work cycle; determine a second maintenance weight of each of the devices to be maintained based on a historical maintenance cost and a remaining service life of each of the devices to be maintained; determine a third maintenance weight of each of the devices to be maintained based on a risk level and a risk type of each of the devices to be maintained; determine a maintenance priority of each of the to-be-maintained devices based on the first maintenance weight, the second maintenance weight and the third maintenance weight of each of the to-be-maintained devices; group the to-be-maintained devices by using an improved clustering algorithm based on the position, the type, the expected maintenance time and the maintenance priority of each of the to-be-maintained devices, to obtain a plurality of sets of maintenance devices, and sort the to-be-maintained devices in each set of maintenance devices according to a comprehensive maintenance priority weight; determine a maintenance strategy based on the average maintenance priority, the average maintenance time, the central position and the type corresponding to each set of maintenance devices.

[0014] The beneficial effects of the above further scheme are that the device maintenance module determines the to-be-maintained devices in combination with the device risk identification result, the historical maintenance record and the real-time health monitoring data, determines the maintenance priority by comprehensively considering the construction task criticality, the historical maintenance cost, the remaining service life, the risk level and the type, sorts the to-be-maintained devices by using the improved clustering algorithm, and finally determines the maintenance strategy, so that the to-be-maintained devices can be more accurately identified, the maintenance priority and grouping can be reasonably arranged, the efficiency and pertinence of device maintenance can be improved, and the stable operation of the devices in the complex construction scene can be ensured.

[0015] Further, the digital twin model identification module is configured to determine a risk identification result of the personnel based on the personnel information and the safety boundary information of each of the devices, and specifically configured to: perform spatial collision detection on the personnel positioning information and the safety boundary information of each of the devices to obtain a collision detection result, wherein the personnel positioning information includes a personnel position and a motion trajectory; determine a safety helmet wearing requirement of an area where the personnel is located based on the personnel positioning information; determine a safety helmet wearing detection result of the personnel based on the safety helmet wearing information and the safety helmet wearing requirement; determine an operation behavior detection result of the personnel based on the operation information and a preset visual behavior recognition model; determine the risk identification result of the personnel based on the collision detection result, the safety helmet wearing detection result and the operation behavior detection result of the personnel.

[0016] The beneficial effects of the above further scheme are that the collision risk of the personnel and the devices is determined by using spatial collision detection, the safety helmet wearing requirement is determined according to the personnel positioning, the operation behavior of the personnel is detected by means of the visual behavior recognition model, and thus the risk identification result of the personnel is comprehensively and accurately determined, and the identification ability and management level of the risk of the construction personnel are improved.

[0017] Further, the decision module is configured to, when determining the plurality of initial management strategies based on the risk identification result and a pre-constructed construction safety knowledge graph, specifically configured to: traverse and query the construction safety knowledge graph based on the risk identification result as a query condition to obtain historical cases similar to the current risk identification result, relevant safety specifications and expert recommendation schemes; select, based on the plurality of historical cases, a successful disposal scheme corresponding to a historical case with a similarity greater than a set similarity threshold as a first strategy set; determine, based on the relevant safety specifications, a management scheme corresponding to the relevant safety specifications as a second strategy set; determine, based on the successful disposal scheme, the management scheme, the expert recommendation scheme and a preset logical reasoning rule, a third strategy set; determine, based on the first strategy set, the second strategy set and the third strategy set, the plurality of initial management strategies.

[0018] The beneficial effects of the above further scheme are that the decision module obtains similar historical cases, relevant safety specifications and expert recommendation schemes from the construction safety knowledge graph based on the risk identification result, then selects a suitable successful disposal scheme from the historical cases to form a first strategy set, determines a management scheme from the relevant safety specifications to form a second strategy set, combines the successful disposal scheme, the management scheme, the expert recommendation scheme and the logical reasoning rule to determine a third strategy set, and finally determines a plurality of initial management strategies based on the three strategy sets, thereby providing more choices for subsequent determination of an optimal management strategy and improving the accuracy and effectiveness of the construction management strategy.

[0019] Further, the decision module is configured to, when determining the optimal management strategy based on the construction progress, the construction plan, the environmental information and a preset multi-objective optimization algorithm, specifically configured to: determine constraint conditions and a plurality of optimization objective functions based on the construction progress, the construction plan and the environmental information, wherein the constraint conditions represent logical constraints between construction processes, equipment safety operation range constraints, personnel working time constraints and personnel skill constraints, and the plurality of optimization objective functions represent functions of minimizing schedule delay, minimizing safety risk and minimizing execution cost; use the plurality of initial management strategies as an initial population of the multi-objective optimization algorithm; for each individual in the current population, simulate the management strategy corresponding to the individual in the three-dimensional digital twin model based on the construction progress, the environmental information and the construction plan to obtain a simulation result, and calculate the fitness value of the individual for each optimization objective function based on the simulation result and the plurality of optimization objective functions; perform non-dominated sorting and selection on all individuals in the current population based on the fitness values corresponding to the individuals, to obtain parent individuals; generate a new generation population based on the parent individuals and preset crossover and mutation operation rules, take the new generation population as a current population of a next iteration of the current iteration, until a preset maximum iteration number is reached or a convergence condition of the optimization objective functions is met, to obtain a non-dominated solution set of the current population; determine the optimal management strategy based on the non-dominated solution set.

[0020] The beneficial effect of the above further scheme is that the decision module determines constraint conditions and optimization objective functions based on construction progress, construction plan and environment information, optimizes the initial management strategy by using a multi-objective optimization algorithm, obtains a non-dominated solution set through simulation, calculation of fitness values, non-dominated sorting, crossover and mutation and the like, and further determines the optimal management strategy, which can minimize schedule delay, minimize safety risk and minimize execution cost, while satisfying constraints in terms of construction process, equipment operation, personnel working time and skills and the like.

[0021] Further, when the decision module is used to determine the optimal scheduling strategy of equipment and personnel in the next work cycle based on the critical path progress risk in the next work cycle, the dynamic safety risk distribution map and the preset optimization algorithm, it is specifically used for: identifying potential delay points and delay probabilities on the construction critical path based on the critical path progress risk; determining risk interaction areas and risk values of personnel and equipment based on the dynamic safety risk distribution map; constructing a multi-objective scheduling optimization model aiming to minimize risk, minimize schedule delay and maximize resource utilization rate based on the state and position information of the equipment and personnel, preset scheduling rules and resource constraints; solving the multi-objective scheduling optimization model by using a multi-objective optimization algorithm to generate multiple scheduling schemes; simulating and evaluating each scheduling scheme in the three-dimensional digital twin model to obtain fitness values of the schemes; selecting an optimal scheduling strategy from a Pareto optimal solution set based on the fitness values of the schemes.

[0022] The beneficial effect of the above further scheme is that the decision module identifies potential delay points and delay probabilities on the construction critical path, determines risk interaction areas and risk values of personnel and equipment, and provides a basis for scheduling; a multi-objective scheduling optimization model is constructed, aiming to minimize risk, minimize schedule delay and maximize resource utilization rate, which can realize efficient utilization of resources and reduction of risk; a multi-objective optimization algorithm is used to solve and generate multiple scheduling schemes, which are simulated and evaluated in the three-dimensional digital twin model, so that an optimal scheduling strategy can be screened out to ensure smooth construction.

[0023] Further, the monitoring module comprises: a plurality of attitude sensors for detecting the attitude of each component of the equipment; a plurality of vibration sensors for detecting the vibration information of each component of the equipment; a plurality of displacement sensors for detecting the displacement of each movable component of the equipment; a plurality of pressure sensors for detecting the pressure information of the hydraulic system of the equipment; a plurality of position tags arranged on the equipment for obtaining the position information of the equipment based on a positioning system; a plurality of cameras arranged at the construction site for collecting personnel images to identify personnel operating behavior and safety helmet wearing information, and collecting construction site image information, based on the construction site image information and a preset construction progress BIM model, determining the construction progress of the construction scene; a plurality of intelligent safety helmets for detecting personnel safety helmet wearing information; a plurality of positioning tags arranged on personnel for obtaining personnel positioning information based on a positioning system; a plurality of wind speed sensors, temperature sensors, humidity sensors, rainfall sensors, dust sensors and noise sensors arranged at the construction site for detecting environmental information of the construction scene.

[0024] The beneficial effects of the above further scheme are: accurate and real-time acquisition of construction progress and construction monitoring information of the construction scene, including equipment attitude, vibration, displacement, pressure, position and other operating parameters, personnel positioning, operation, safety helmet wearing and other information, construction site image information and environmental information.

[0025] In a second aspect, the present application provides a construction management method for a complex construction scene, which adopts the following technical scheme: A construction management method for a complex construction scene comprises: real-time acquisition of construction progress and construction monitoring information of the construction scene, the construction monitoring information comprising personnel information, equipment operating parameters, construction site image information and environmental information, the personnel information comprising personnel positioning information, operating information and safety helmet wearing information, the construction scene comprising a plurality of equipment, including tower crane equipment, bucket wheel machines, excavators and construction elevators; integrating the construction progress, construction monitoring information and equipment operating logic of the construction scene into a pre-constructed three-dimensional digital twin model; For each device, based on the device's operating logic, operating parameters, environmental information, and a preset risk algorithm, the device's safety boundary information is determined; based on the personnel information and the safety boundary information of each device, the personnel risk identification result is determined, the risk identification result including risk level and risk type; based on the device's operating parameters, the risk identification result for each device is determined. Based on the risk identification results and the pre-constructed construction safety knowledge graph, multiple initial management strategies are determined. Based on the construction progress, construction plan, environmental information and a preset multi-objective optimization algorithm, the multiple initial management strategies are optimized to determine the optimal management strategy, and the optimal management strategy is sent to the management terminal corresponding to the optimal management strategy.

[0026] Additional aspects and advantages of this application will be set forth in part in the description which follows, and will become apparent from the description or may be learned by practice of this application. Attached Figure Description

[0027] Figure 1 A schematic diagram of the structure of a construction management system for a complex construction scenario provided in one embodiment of the present invention; Figure 2 This is a flowchart illustrating a construction management method for complex construction scenarios, provided as an embodiment of the present invention. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, 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 some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0029] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0030] like Figure 1 As shown, a construction management system 100 for complex construction scenarios mainly includes: The monitoring module 101 is configured to acquire construction progress and construction monitoring information of a construction scene in real time, the construction monitoring information comprising personnel information, equipment operation parameters, construction site image information and environmental information, the personnel information comprising personnel positioning information, operation information and safety helmet wearing information, and the construction scene comprising a plurality of equipment, the plurality of equipment comprising tower crane equipment, bucket wheel machines, excavators and construction elevators. The digital twin model identification module 102 is configured to establish a three-dimensional digital twin model of the construction scene and the equipment in the construction scene, and integrate the construction progress, the construction monitoring information and the operation logic of the equipment in the three-dimensional digital twin model. The digital twin model identification module 102 is further configured to, for each of the equipment, determine safety boundary information of the equipment based on the operation logic of the equipment, the operation parameters of the equipment, the environmental information and a preset risk algorithm, determine a risk identification result of personnel based on the personnel information and the safety boundary information of each of the equipment, the risk identification result comprising a risk level and a risk type, and determine a risk identification result of each of the equipment based on the operation parameters of the equipment. The decision module 103 is configured to determine a plurality of initial management strategies based on the risk identification result and a pre-constructed construction safety knowledge graph, optimize the plurality of initial management strategies based on the construction progress, a construction plan, the environmental information and a preset multi-objective optimization algorithm, determine an optimal management strategy, and send the optimal management strategy to a management terminal corresponding to the optimal management strategy.

[0031] By constructing a three-dimensional digital twin model that integrates real-time monitoring data and equipment operation logic, full-factor digital mapping and dynamic risk perception of a complex construction scene are achieved. The system can automatically identify safety boundaries of personnel and equipment based on multi-source heterogeneous data, and generate an optimal management strategy that takes into account safety, progress and cost through collaborative decision-making of a knowledge graph and a multi-objective optimization algorithm, thereby improving the active safety early warning capability and the fine management level of the construction process and reducing the incidence of safety accidents.

[0032] Optionally, the monitoring module 101 comprises: A plurality of attitude sensors configured to detect the attitude of each component of the equipment. A plurality of vibration sensors configured to detect vibration information of each component of the equipment. A plurality of displacement sensors configured to detect the displacement of each movable component of the equipment. A plurality of pressure sensors configured to detect pressure information of a hydraulic system of the equipment. A plurality of position tags arranged on the equipment and configured to acquire position information of the equipment based on a positioning system. A plurality of cameras arranged at the construction site are configured to collect personnel images to identify personnel operation behaviors and safety helmet wearing information, and collect construction site image information, and based on the construction site image information and a preset construction progress BIM model, determine the construction progress of the construction scene. A plurality of intelligent safety helmets are configured to detect safety helmet wearing information of personnel. A plurality of positioning tags arranged on personnel are configured to obtain personnel positioning information based on a positioning system. A plurality of wind speed sensors, temperature sensors, humidity sensors, rainfall sensors, dust sensors, and noise sensors arranged at the construction site are configured to detect environmental information of the construction scene.

[0033] In the embodiments of the present application, the attitude sensor can be a high-precision MEMS attitude sensor, which accurately detects the attitude of each component of the equipment. Of course, a fiber-optic gyroscope attitude sensor can also be used. The attitude sensor is fixed on the equipment component by means of bolts or glue, and is connected with the data acquisition system to transmit the detected attitude data in real time. The vibration sensor can be a piezoelectric vibration sensor, which can effectively detect the vibration information of the equipment component. The displacement sensor can be a magnetostrictive displacement sensor, which accurately detects the displacement of the moving parts of the equipment, or can be a grating displacement sensor. The displacement sensor is installed on the movement path of the moving parts of the equipment and is fixed by mechanical connection or magnetic adsorption.

[0034] The pressure sensor can be a diffused silicon pressure sensor, which is used to detect the pressure information of the hydraulic system of the equipment and is installed on the pipeline or container of the hydraulic system through a sealed interface. The position tag can be a UWB position tag, which obtains the position information of the equipment based on a positioning system. The position tag is installed at a suitable position of the equipment and interacts with the positioning base station through wireless communication. The camera can be a high-definition network camera, which is arranged at each key position of the construction site to collect personnel images and construction site image information.

[0035] The intelligent safety helmet is built-in with a detection chip and a sensor, which can detect the safety helmet wearing information of personnel. A new type of intelligent helmet with the same function can also be used. The positioning tag is arranged on the personnel and can be a small and light active RFID positioning tag, which obtains the positioning information of the personnel through a positioning system. The wind speed sensor, temperature sensor, and other environmental sensors are distributed in different areas of the construction site to detect environmental information in real time. These sensors transmit data to the data processing center of the monitoring module 101 through wired or wireless means, and are combined to realize comprehensive real-time monitoring of the construction scene, so as to obtain accurate construction progress and construction monitoring information in time to provide basic data for subsequent management.

[0036] Optionally, the digital twin model identification module 102 is configured to establish a three-dimensional digital twin model of the construction scene and the equipment in the construction scene. When establishing the model, a professional three-dimensional modeling software such as 3ds Max, Maya, etc. is used to construct a three-dimensional model in combination with the equipment size, position and other information obtained by the monitoring module 101. The construction progress of the construction scene, the construction monitoring information, and the operation logic of the equipment are integrated in the three-dimensional digital twin model, so that the model can truly reflect the actual situation of the construction scene. For each equipment, the safety boundary information of the equipment is determined based on the operation logic of the equipment, the operating parameters of the equipment, the environmental information and the preset risk algorithm. For example, the safe rotation angle, lifting height, etc. of the tower crane equipment under different wind speeds and different lifting weights are the safety boundary information. Based on the personnel information and the safety boundary information of each equipment, the risk identification result of the personnel is determined.

[0037] In the embodiments of the present application, the digital twin model identification module 102 is configured to determine the risk identification result of the personnel based on the personnel information and the safety boundary information of each equipment, and specifically configured to: perform spatial collision detection on the personnel positioning information and the safety boundary information of each equipment to obtain a collision detection result, wherein the personnel positioning information includes a personnel position and a motion trajectory; determine a safety helmet wearing requirement of an area where the personnel is located based on the personnel positioning information; determine a safety helmet wearing detection result of the personnel based on the safety helmet wearing information and the safety helmet wearing requirement; determine an operation behavior detection result of the personnel based on the operation information and a preset visual behavior recognition model; determine the risk identification result of the personnel based on the collision detection result, the safety helmet wearing detection result and the operation behavior detection result of the personnel.

[0038] In the above embodiments, laser scanning technology or millimeter wave radar technology can be used for more accurate spatial collision detection.

[0039] The digital twin model identification module 102 calculates the shortest distance from the personnel position point to the safety boundary geometric model of the equipment, and when the distance is lower than a preset safety threshold, it is determined that there is a spatial collision risk.

[0040] According to different functional areas (such as high-altitude operation area, heavy lifting area, material stacking area) where the personnel is located, the digital twin model identification automatically matches the safety equipment configuration required by the area, for example, a safety helmet must be worn in the lifting area. The digital twin model identification compares the wearing state information uploaded by the personnel intelligent safety helmet with the rule requirements, and outputs a detection result of whether it is in compliance.

[0041] The digital twin model recognition module 102 analyzes video frames in real time based on a preset visual behavior recognition model (for example, a classification model trained based on YOLO, OpenPose, etc.) according to video streams uploaded by cameras deployed in key areas. The model can identify typical unsafe operation behaviors such as "not directing according to the specified gesture", "entering the blind area of the equipment", "violating the warning line", etc., and output operation behavior detection results.

[0042] In the embodiments of the present application, the digital twin model recognition module 102 is configured to determine the risk identification result of each device based on the device operation parameters, and specifically configured to: The device operation parameters transmitted by the monitoring module 101 are received, for example, for a tower crane, the operation parameters include hoisting load data collected by a torque sensor in real time, amplitude data collected by an amplitude sensor, hook height data collected by a height sensor, tower top wind speed data collected by an anemometer, and rotation angle data collected by a rotation sensor. For a construction elevator, the operation parameters include load, running speed, floor position information. For excavators and other heavy machinery, the operation parameters include engine speed, hydraulic system pressure, hydraulic oil temperature, and vehicle body posture data collected by an inclination sensor.

[0043] The received original operation parameters are preprocessed, including data cleaning (such as processing noise and outliers) and standardization to obtain time series data. Key feature indicators are calculated from the time series data through a preset feature extraction algorithm. For example, for a tower crane, the ratio of actual load to rated load (load rate), theoretical rated load at the current amplitude, the ratio of actual wind speed to allowable working wind speed, and the acceleration of the rotation mechanism are calculated; for a hydraulic excavator, the fluctuation variance of the hydraulic system pressure, the duration of continuous high-load operation of the engine, and other features are extracted.

[0044] The extracted feature indicators are input into a preset risk identification model to determine the risk identification result. The risk identification model is composed of a rule base and a machine learning model. The safety thresholds of each device are pre-stored in the rule base, for example: the maximum wind speed threshold for the allowable working wind speed of the tower crane; the overload threshold and the overspeed threshold of the construction elevator. When any feature indicator exceeds its corresponding safety threshold, the rule is triggered, and it is determined that there is a risk of illegal operation. At the same time, the machine learning model comprehensively analyzes the feature vector to identify potential hidden risks, for example: by analyzing the subtle change trend of the hydraulic system pressure to predict potential hydraulic failure risks, or by analyzing the engine operation data and vibration data to assess the health degradation risk of the equipment.

[0045] The risk identification result at least includes a "risk level", which can be divided into "no risk", "low risk", "medium risk", and "high risk", and a "risk type", which is explicitly "overload risk", "capsizing risk", "mechanical failure risk", or "violation operation risk". Finally, the risk identification result is sent to the decision module 103 as a key basis for generating a management strategy.

[0046] Optionally, when the decision module 103 is used to determine a plurality of initial management strategies based on the risk identification result and a pre-constructed construction safety knowledge graph, it is specifically used for: The risk identification result is used as a query condition to perform traversal query in the construction safety knowledge graph to obtain similar historical cases, related safety specifications, and expert advice schemes to the current risk identification result. Based on a plurality of historical cases, a successful disposal scheme corresponding to a historical case with a similarity greater than a set similarity threshold is selected as a first strategy set. Based on the related safety specifications, a management scheme corresponding to the related safety specifications is determined as a second strategy set. Based on the successful disposal scheme, the management scheme, the expert advice scheme, and a preset logical reasoning rule, a third strategy set is determined. Based on the first strategy set, the second strategy set, and the third strategy set, a plurality of initial management strategies are determined.

[0047] In the embodiments of the present application, the construction safety knowledge graph is constructed based on a graph database, and the node types thereof include risk types, device types, environmental conditions, specification clauses, historical cases, disposal schemes, and expert advice. The edge relationship defines the semantic connection between the nodes, such as belonging to a risk type, violating a specification clause, having a disposal scheme, and being applicable to a device. When the decision module 103 receives the real-time risk identification result, the risk identification result is converted into a graph query statement. For example, the statement will find all "historical case" nodes and "specification clause" nodes directly associated with the "personnel intrusion into the forbidden area" risk node, and traverse to obtain the detailed scene data of these cases, the content of the disposal scheme node, and the specific requirements of the specification clause.

[0048] The decision module 103 uses the schemes in the first strategy set and the second strategy set together with the queried expert advice scheme as a basis for reasoning, runs a reasoning engine, deduces new, combined strategy options, and generates a strategy constituting a third strategy set.

[0049] Optionally, when the decision module 103 is used to determine an optimal management strategy based on the construction progress, the construction plan, the environmental information, and a preset multi-objective optimization algorithm, it is specifically used for: determine constraint conditions and multiple optimization objective functions based on the construction progress, the construction plan, and the environmental information, the constraint conditions representing logical constraints between construction processes, safe operation range constraints of equipment, working time constraints of personnel, and skill constraints of personnel, and the multiple optimization objective functions representing functions of minimizing construction duration delay, minimizing safety risk, and minimizing execution cost; use the multiple initial management strategies as initial populations of the multi-objective optimization algorithm; for each individual in the current population, simulate the management strategy corresponding to the individual in the three-dimensional digital twin model based on the construction progress, the environmental information, and the construction plan, obtain simulation results, and calculate fitness values of the individual for the multiple optimization objective functions according to the simulation results and the multiple optimization objective functions; based on the fitness values corresponding to the individuals, perform non-dominated sorting and selection on all individuals in the current population to obtain parent individuals; based on the parent individuals and preset crossover and mutation operation rules, generate a new generation of populations, use the new generation of populations as the current population of the next iteration of the current iteration, and perform iterations until a preset maximum number of iterations is reached or a convergence condition of the multiple optimization objective functions is met, to obtain a non-dominated solution set of the current population; determine an optimal management strategy based on the non-dominated solution set.

[0050] In the embodiments of the application, constraint conditions are defined, for example, construction process constraints (a subsequent process B must start after a preceding process A is completed), equipment safety constraints (a working range of a tower crane must not exceed a rated range), and personnel constraints (a working time of a single day must not exceed a legal upper limit, and an operator must have corresponding qualifications). The multi-objective optimization algorithm is a non-dominated sorting genetic algorithm.

[0051] As an optional implementation manner of the embodiments of the application, the construction management system for a complex construction scene further includes: a construction deduction and prediction module configured to drive the three-dimensional digital twin model to perform forward construction deduction based on the construction progress of the current period, the optimal management strategy, and the construction plan and meteorological prediction data of a next working period of the current period, and predict a critical path progress risk and a dynamic safety risk distribution map in the next working period; The decision module 103 is further configured to determine an optimal scheduling strategy of equipment and personnel in the next working period based on the critical path progress risk and the dynamic safety risk distribution map in the next working period and a preset optimization algorithm.

[0052] The personnel training module is configured to acquire historical operation behaviors, risk records, and historical simulation information of personnel, generate a simulation training task based on the historical operation behaviors, the risk records, the historical simulation information, and a scheduling strategy of the personnel in a next work cycle, determine a target case from the construction safety knowledge graph based on the simulation training task, and generate a virtual training scene in the three-dimensional digital twin model based on the target case, wherein the virtual training scene includes a virtual scene environment, a virtual device state, a virtual fault setting, a virtual risk trigger logic, and multiple training contents. During the personnel training, the personnel training module is configured to acquire multi-modal operation data of the personnel in the virtual training scene in real time, determine whether the personnel have operation errors based on a matching result of the multi-modal operation data and a standard operation model, acquire a historical operation frequency, a historical operation error frequency, a historical occurrence probability of the training content in a real construction scene, an impact level, and a correlation of the training content with a construction task in a next work cycle of the personnel on the training content if the personnel have operation errors, determine a priority of the training content based on the historical operation frequency, the historical operation error frequency, the historical occurrence probability of the training content in the real construction scene, the impact level, the correlation of the training content with the construction task in the next work cycle, and a preset weighting calculation method, and select a trigger content, a timing, and an intensity from a set of priority risks.

[0053] The device maintenance module is configured to determine multiple to-be-maintained devices based on risk identification results, historical maintenance records, and real-time health monitoring data of each of the devices. The device maintenance module is configured to determine a first maintenance weight of each of the to-be-maintained devices based on a criticality of a construction task of each of the to-be-maintained devices in a next work cycle. The device maintenance module is configured to determine a second maintenance weight of each of the to-be-maintained devices based on a historical maintenance cost and a remaining service life of each of the to-be-maintained devices. The device maintenance module is configured to determine a third maintenance weight of each of the to-be-maintained devices based on a risk level and a risk type of each of the to-be-maintained devices. The device maintenance module is configured to determine a maintenance priority of each of the to-be-maintained devices based on the first maintenance weight, the second maintenance weight, and the third maintenance weight of each of the to-be-maintained devices. The device maintenance module is configured to group the to-be-maintained devices by using an improved clustering algorithm based on locations, types, expected maintenance durations, and maintenance priorities of the to-be-maintained devices to obtain multiple maintenance device sets, and sort the to-be-maintained devices in the maintenance device sets according to comprehensive maintenance priority weights. The device maintenance module is configured to determine a maintenance strategy based on an average maintenance priority, an average maintenance duration, a central location, and a type corresponding to each of the maintenance device sets.

[0054] In the embodiments of the present application, the construction deduction and prediction module simulates the operation of the tower crane equipment in the rain in the three-dimensional digital twin model according to the rainfall in the meteorological prediction data, and predicts possible progress delays and safety risks. The construction deduction and prediction module drives the three-dimensional digital twin model, uses a discrete event simulation engine, takes "day" or "hour" as a time step, simulates the construction activities in the next working period. The simulation engine simulates according to the process logic of the construction plan and injects the instructions of the optimal management strategy, counts the probability of delay and the distribution of delay degree of each node on the critical path, and identifies potential progress bottlenecks and delay risk points.

[0055] In the embodiments of the present application, the decision module 103 is configured to determine the optimal scheduling strategy of equipment and personnel in the next working period based on the critical path progress risk in the next working period, the dynamic safety risk distribution map, and a preset optimization algorithm, and specifically configured to: identify potential delay points and delay probabilities on the construction critical path based on the critical path progress risk; determine the risk interaction area and risk value of personnel and equipment based on the dynamic safety risk distribution map; construct a multi-objective scheduling optimization model with the objectives of minimizing risk, minimizing schedule delay, and maximizing resource utilization rate based on the state and position information of equipment and personnel, preset scheduling rules, and resource constraints; solve the multi-objective scheduling optimization model based on a multi-objective optimization algorithm to generate multiple scheduling schemes; simulate and evaluate each scheduling scheme in the three-dimensional digital twin model to obtain the fitness value of each scheme; select the optimal scheduling strategy from the Pareto optimal solution set based on the fitness value of each scheme.

[0056] In the above embodiments, based on the critical path progress risk, the critical path in the construction scene is identified, and the task nodes with delay probability exceeding a preset threshold are located. By image recognition or direct processing of the heat map data grid, high-risk human-machine interaction areas and their corresponding risk values that will appear in different time periods in the next working period are determined. The real-time state of equipment and personnel (such as whether the equipment is available and whether the personnel are on duty), position information, preset scheduling rules (such as special operation personnel qualification matching rules), and resource constraints (such as total number of equipment and total number of personnel) are integrated to construct a multi-objective scheduling optimization model.

[0057] The multi-objective optimization algorithm is used to solve the multi-objective scheduling optimization model. Possible scheduling schemes are encoded into individuals to form an initial population. In each generation of evolution, each individual is injected into a three-dimensional digital twin model to simulate the entire process of the next work cycle. The simulation engine simulates personnel movement, equipment operation, task execution process, and interaction with environmental risks under the scheduling scheme. After the simulation ends, the objective function value corresponding to the scheduling scheme is calculated as the fitness value of the individual. After multiple iterations, the Pareto optimal solution set is output.

[0058] Finally, the final optimal scheduling strategy is selected from the Pareto solution set based on the preset decision rule. The decision rule can be a user-configured preference setting, for example, during a strict safety management period, the risk target is given a high weight; during the rush period, the duration target is given a high weight.

[0059] In the embodiments of the present application, the personnel training module obtains the historical operation behavior data, past risk records, performance information in previous simulation training of the target personnel, and receives the expected task scheduling strategy of the next work cycle of the personnel from the decision module 103.

[0060] By analyzing the model to evaluate the skill deficiencies of the personnel and the risk points of future tasks, simulation training tasks are automatically generated. The simulation training tasks are used as query requests to match in the construction safety knowledge graph to find target cases with similar risk characteristics, equipment types, and work environments. Then, the server drives the three-dimensional digital twin model to instantiate a virtual training scene according to the scene data, equipment parameters, and accident logic of the selected target case. The scene not only contains a real site environment model and equipment state, but also preinstalls specific virtual failures (such as sudden jamming of a steel cable) and risk triggering logic (such as triggering a risk event when the personnel enters a specific dangerous area), and encapsulates multiple targeted training content units.

[0061] The personnel training module realizes the specific deficiencies of each personnel and the future task requirements, provides highly personalized and precise training, and thus effectively improves the safety awareness and operation skills of the personnel from the source, and reduces human risks in actual operation.

[0062] In the embodiments of the present application, for each device to be maintained, the device maintenance module inputs the first, second and third maintenance weights into a weighted summation formula to calculate the comprehensive maintenance priority of each device. Then, the GIS geographic location information, device type and estimated maintenance time of all devices to be maintained are obtained. An improved clustering algorithm is used to cluster and analyze these devices, and devices with similar geographic locations, similar types and equivalent maintenance priorities are divided into the same maintenance device set. For each generated maintenance device set, a maintenance strategy is automatically generated according to the average maintenance priority, total estimated maintenance time, geographic center position and main device type of the set. The devices to be maintained are accurately identified, the maintenance priority and grouping are reasonably arranged, the efficiency and pertinence of device maintenance are improved, and the stable operation of the devices in complex construction scenes is ensured.

[0063] Figure 2 A flowchart of a construction management method for a complex construction scene is shown.

[0064] As shown in Figure 2 A construction management method for a complex construction scene mainly includes: S201, real-time acquisition of construction progress and construction monitoring information of a construction scene, wherein the construction monitoring information includes personnel information, device operation parameters, construction site image information and environmental information, the personnel information includes personnel positioning information, operation information and safety helmet wearing information, there are multiple devices in the construction scene, and the multiple devices include tower crane devices, bucket wheel machines, excavators and construction elevators; S202, integration of the construction progress, construction monitoring information and device operation logic of the construction scene in a pre-constructed three-dimensional digital twin model; S203, for each device, determination of safety boundary information of the device based on the operation logic of the device, the operation parameters of the device, the environmental information and a preset risk algorithm; determination of risk identification results of personnel based on the personnel information and the safety boundary information of each device, wherein the risk identification results include risk levels and risk types; and determination of risk identification results of each device based on the device operation parameters; S204, determination of multiple initial management strategies based on the risk identification results and a pre-constructed construction safety knowledge graph; optimization of the multiple initial management strategies based on the construction progress, construction plan, environmental information and a preset multi-objective optimization algorithm to determine an optimal management strategy, and sending of the optimal management strategy to a management terminal corresponding to the optimal management strategy.

[0065] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described method can refer to the corresponding process in the foregoing system embodiments, which will not be described herein.

[0066] Those skilled in the art can understand that the modules and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software manner depends on specific application and design constraints of the technical solutions. Those skilled in the art can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0067] The term "comprising" or any other variation thereof is intended to cover a non-exclusive inclusion, so that a process, method, article, or apparatus that comprises a list of elements does not only include those elements but can also include other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0068] The above description is merely preferred embodiments of the present application and a description of the principles of the technology used. Those skilled in the art should understand that the scope of the application involved in the present application is not limited to the technical solutions formed by the specific combinations of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above application concept. For example, the technical solutions formed by mutually replacing the above features with the technical features applied in the present application (but not limited to) having similar functions.

Claims

1. A construction management system for complex construction scenarios, characterized in that, include: The monitoring module is used to acquire construction progress and construction monitoring information in the construction scene in real time. The construction monitoring information includes personnel information, equipment operating parameters, construction site image information and environmental information. The personnel information includes personnel positioning information, operation information and safety helmet wearing information. There are multiple devices in the construction scene, including tower cranes, bucket wheel excavators, excavators and construction elevators. The digital twin model recognition module is used to establish a three-dimensional digital twin model of the construction scene and the equipment in the construction scene, and to integrate the construction progress, construction monitoring information and equipment operation logic of the construction scene into the three-dimensional digital twin model. The digital twin model identification module is also used to determine the safety boundary information of each device based on the device's operating logic, operating parameters, environmental information, and a preset risk algorithm. Based on the personnel information and the security boundary information of each device, the risk identification result of the personnel is determined, and the risk identification result includes risk level and risk type; Based on the equipment operating parameters, the risk identification result for each of the aforementioned devices is determined; The decision-making module is used to determine multiple initial management strategies based on risk identification results and a pre-built construction safety knowledge graph; Based on construction progress, construction plan, environmental information, and a preset multi-objective optimization algorithm, multiple initial management strategies are optimized to determine the optimal management strategy, and the optimal management strategy is sent to the management terminal corresponding to the optimal management strategy.

2. The construction management system for complex construction scenarios according to claim 1, characterized in that, The system also includes: The construction simulation and prediction module is used to drive the three-dimensional digital twin model to perform forward construction simulation based on the current cycle's construction progress, optimal management strategy, and the construction plan and meteorological forecast data for the next work cycle, predicting the critical path schedule risk and dynamic safety risk distribution map in the next work cycle. The decision-making module is also used to determine the optimal scheduling strategy for equipment and personnel in the next work cycle based on the critical path schedule risk, dynamic safety risk distribution map and preset optimization algorithm in the next work cycle.

3. The construction management system for complex construction scenarios according to claim 2, characterized in that, The system also includes: The personnel training module is used to acquire personnel's historical operational behaviors, risk records, and historical simulation information. Based on the historical operational behaviors, risk records, historical simulation information, and the personnel's scheduling strategy for the next work cycle, it generates simulation training tasks. Based on the simulation training tasks, it determines target cases from the construction safety knowledge graph. Based on the target cases, it generates virtual training scenarios in the three-dimensional digital twin model. The virtual training scenarios include virtual scene environments, virtual equipment states, virtual fault settings, virtual risk triggering logic, and multiple training contents. During personnel training, multimodal operation data of personnel in training content within a virtual training scenario is acquired in real time. Based on the matching results of the multimodal operation data and the standard operation model, it is determined whether the personnel have made operational errors. If the personnel have made operational errors, the historical operation frequency, historical operation error frequency, historical occurrence probability and impact level of the training content in a real construction scenario, and the relevance of the training content to the construction tasks in the next work cycle are acquired. Based on the historical operation frequency, historical operation error frequency, historical occurrence probability and impact level of the training content in a real construction scenario, the relevance of the training content to the construction tasks in the next work cycle, and a preset weighted calculation method, the priority of the training content is determined, and triggering content, timing, and intensity are selected from a set of priority risk factors.

4. The construction management system for complex construction scenarios according to claim 2, characterized in that, The system also includes: The equipment maintenance module is used to identify multiple devices to be maintained based on the risk identification results, historical maintenance records, and real-time health monitoring data of each device. Based on the criticality of the construction task for each piece of equipment to be maintained in the next work cycle, determine the first maintenance weight for each piece of equipment to be maintained; Based on the historical maintenance costs and remaining useful life of each piece of equipment to be maintained, a second maintenance weight is determined for each piece of equipment to be maintained; Based on the risk level and risk type of each piece of equipment to be maintained, determine the third maintenance weight for each piece of equipment to be maintained; Based on the first maintenance weight, second maintenance weight, and third maintenance weight of each device to be maintained, the maintenance priority of each device to be maintained is determined; Based on the location, type, estimated maintenance duration, and maintenance priority of each device to be maintained, an improved clustering algorithm is used to group the devices to be maintained into multiple sets of maintenance devices, and the devices to be maintained within each set of maintenance devices are sorted according to the comprehensive maintenance priority weight. Maintenance strategies are determined based on the average maintenance priority, average maintenance duration, central location, and type of each set of maintenance equipment.

5. A construction management system for complex construction scenarios according to claim 1, characterized in that, The digital twin model recognition module, when determining the risk identification result of a person based on the personnel information and the security boundary information of each device, is specifically used for: Spatial collision detection is performed on the personnel positioning information and the safety boundary information of each of the aforementioned devices to obtain the collision detection result. The personnel positioning information includes the personnel's position and movement trajectory. Based on the personnel location information, determine the helmet wearing requirements for the area where the personnel are located; Based on the helmet wearing information and the helmet wearing requirements, the helmet wearing test results of the personnel are determined; Based on the operational information and the preset visual behavior recognition model, the detection result of the person's operational behavior is determined; Based on the collision detection results, helmet wearing detection results, and operational behavior detection results of the personnel, the risk identification results of the personnel are determined.

6. The construction management system for complex construction scenarios according to claim 1, characterized in that, The decision-making module, when determining multiple initial management strategies based on risk identification results and a pre-built construction safety knowledge graph, is specifically used for: Using the risk identification results as query conditions, a comprehensive search is conducted in the construction safety knowledge graph to obtain historical cases, relevant safety specifications, and expert recommendations that are similar to the current risk identification results. Based on multiple historical cases, successful handling solutions corresponding to historical cases with similarity greater than a set similarity threshold are selected as the first strategy set; Based on relevant security standards, the management schemes corresponding to the relevant security standards are determined as the second set of strategies; Based on successful handling plans, management plans, expert advice plans, and pre-set logical reasoning rules, a third set of strategies is determined; Based on the first set of strategies, the second set of strategies, and the third set of strategies, multiple initial management strategies are determined.

7. A construction management system for complex construction scenarios according to claim 6, characterized in that, The decision-making module is used to optimize multiple initial management strategies based on construction progress, construction plan, environmental information, and a preset multi-objective optimization algorithm. Specifically, when determining the optimal management strategy, it is used for: Based on construction progress, construction plan, and environmental information, constraints and multiple optimization objective functions are determined. The constraints represent the logical constraints between construction procedures, the safe operating range constraints of equipment, the constraints of personnel working time, and the constraints of personnel skills. The multiple optimization objective functions represent functions that minimize construction delays, minimize safety risks, and minimize execution costs. The initial management strategies are used as the initial population for the multi-objective optimization algorithm; For each individual in the current population, the management strategy corresponding to the individual is simulated in a three-dimensional digital twin model based on the construction progress, environmental information and construction plan, and the simulation results are obtained. Based on the simulation results and multiple optimization objective functions, the fitness value of the individual for each optimization objective function is calculated. Based on the fitness value of each individual, a non-dominated sorting and selection is performed on all individuals in the current population to obtain the parent individuals; Based on the parent individuals and the preset crossover and mutation operation rules, a new generation population is generated. The new generation population is used as the current population of the next iteration of the current iteration until the preset maximum number of iterations is reached or the convergence conditions of multiple optimization objective functions are met, and the non-dominated solution set of the current population is obtained. The optimal management strategy is determined based on the non-dominated solution set.

8. A construction management system for complex construction scenarios according to claim 2, characterized in that, The decision-making module, when determining the optimal scheduling strategy for equipment and personnel in the next work cycle based on the critical path schedule risk, dynamic safety risk distribution map, and preset optimization algorithm, is specifically used for: Based on critical path schedule risk, identify potential delay points and delay probabilities on the critical path of construction. Based on the dynamic safety risk distribution map, the risk interaction areas and risk values ​​between personnel and equipment are determined; Based on the status and location information of equipment and personnel, preset scheduling rules and resource constraints, a multi-objective scheduling optimization model is constructed with the goals of minimizing risk, minimizing project delays and maximizing resource utilization. The multi-objective scheduling optimization model is solved based on a multi-objective optimization algorithm to generate multiple scheduling schemes. Each scheduling scheme is simulated and evaluated in a three-dimensional digital twin model to obtain the fitness value of each scheme; Based on the fitness values ​​of each scheme, the optimal scheduling strategy is selected from the Pareto optimal solution set.

9. A construction management system for complex construction scenarios according to claim 1, characterized in that, The monitoring module includes: Multiple attitude sensors are used to detect the attitude of various components of the device; Multiple vibration sensors are used to detect vibration information of various components of the device; Multiple displacement sensors are used to detect the displacement of each moving part of the device; Multiple pressure sensors are used to detect pressure information in the equipment's hydraulic system; Multiple location tags are set on the device to obtain the device's location information based on the positioning system; Multiple cameras are installed at the construction site to capture images of personnel to identify their operational behavior and helmet wearing information, as well as to collect construction site image information. Based on the construction site image information and a preset construction progress BIM model, the construction progress of the construction scene is determined. Multiple smart safety helmets are used to detect whether a person is wearing a safety helmet; Multiple location tags are placed on personnel to obtain personnel location information based on the location system; Multiple wind speed sensors, temperature sensors, humidity sensors, rainfall sensors, dust sensors, and noise sensors are installed at the construction site to detect environmental information of the construction scene.

10. A construction management method for complex construction scenarios, characterized in that, include: The system acquires real-time construction progress and monitoring information for the construction scene. The construction monitoring information includes personnel information, equipment operating parameters, construction site image information, and environmental information. The personnel information includes personnel positioning information, operation information, and safety helmet wearing information. The construction scene contains multiple pieces of equipment, including tower cranes, bucket wheel excavators, excavators, and construction elevators. The construction progress, construction monitoring information, and equipment operation logic of the construction scenario are integrated into a pre-built three-dimensional digital twin model; For each device, the safety boundary information of the device is determined based on the device's operating logic, operating parameters, environmental information, and a preset risk algorithm. Based on the personnel information and the security boundary information of each device, the risk identification result of the personnel is determined, and the risk identification result includes risk level and risk type; Based on the equipment operating parameters, the risk identification result for each of the aforementioned devices is determined; Based on the risk identification results and the pre-constructed construction safety knowledge graph, several initial management strategies were determined; Based on construction progress, construction plan, environmental information, and a preset multi-objective optimization algorithm, multiple initial management strategies are optimized to determine the optimal management strategy, and the optimal management strategy is sent to the management terminal corresponding to the optimal management strategy.

Images