Construction progress intelligent cooperative control method and system driven by multi-source data

Abstract

The invention relates to the technical field of construction progress collaborative management, in particular to a multi-source data driven construction progress intelligent collaborative control method and system. The method comprises the following steps: acquiring multi-source data of a construction site; dynamically fusing the multi-source data to obtain a three-dimensional construction map; constructing a construction plan planning model, and generating an optimal construction plan by using the construction plan planning model based on the three-dimensional construction map; and performing dynamic resource scheduling and allocation according to the optimal construction plan to obtain a resource allocation scheme. According to the method, the three-dimensional construction map is constructed by fusing the multi-source data, then the construction plan planning model is constructed to obtain the optimal construction plan, and finally resource allocation is performed based on the optimal construction plan, so that the problems of data islands, progress lag and response lag in traditional highway engineering construction are solved, and the construction efficiency and safety are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of collaborative management of construction progress, and particularly to a method and system for intelligent collaborative control of construction progress driven by multi-source data. Background Art

[0002] During the construction process of a building project, construction progress control is a key link to ensure the project is completed on time, arrange resources reasonably, and control costs. Traditional construction progress control methods mainly rely on on-site manual observation, recording, and analysis, which have problems such as untimely, inaccurate, and incomplete data collection, as well as unsmooth information sharing and low collaborative efficiency among all participating parties. With the development of information technology, although some information-based means such as project management software have been applied to construction progress control, these methods often only utilize data from a single source, unable to fully integrate various data resources at the construction site, and it is difficult to achieve intelligent, precise, and collaborative control of construction progress. In actual construction, there are a large number of multi-source data at the construction site, including but not limited to construction drawings, progress plans, resource allocation data, on-site monitoring data, weather data, geological data, etc. These data are scattered in different systems and departments and have not been effectively integrated and utilized, resulting in a lack of comprehensive and real-time data support for construction progress control, making it difficult to timely discover and solve problems that occur during the construction process and affecting the smooth progress of the project. Therefore, there is an urgent need for a method that can integrate multi-source data and achieve intelligent collaborative control of construction progress. Summary of the Invention

[0003] In view of the deficiencies of existing methods and the requirements of practical applications, in order to solve the problems of data islands, progress lags, and response lags in traditional highway engineering construction and significantly improve construction efficiency and safety. On the one hand, the present invention provides a method for intelligent collaborative control of construction progress driven by multi-source data, including the following steps: obtaining multi-source data at the construction site; dynamically fusing the multi-source data to obtain a three-dimensional construction map; constructing a construction plan planning model, and based on the three-dimensional construction map, using the construction plan planning model to generate an optimal construction plan; performing dynamic resource scheduling and allocation according to the optimal construction plan to obtain a resource allocation plan.

[0004] The present invention constructs a three-dimensional construction map by fusing multi-source data, then constructs a construction plan planning model to obtain an optimal construction plan, and finally performs resource allocation based on the optimal construction plan, solving the problems of data islands, progress lags, and response lags in traditional highway engineering construction and significantly improving construction efficiency and safety.

[0005] Optionally, the dynamically fusing the multi-source data to obtain a three-dimensional construction map satisfies the following formula: , where represents Time coordinate The fused data value at represents the dynamic weight coefficient and satisfies: , represents the mapping of the point cloud intensity value of the lidar at the coordinate ; represents the grayscale normalization value of the UAV image at the coordinate ; represents the state vector of the sensor at the moment. By fusing multi-source data, a three-dimensional construction map of the construction site can be obtained, which is beneficial to the rapid generation of the construction progress plan and further improves the collaborative control ability of the construction progress. Through the dynamic weight the contribution of different data sources is adjusted in real time, and multi-source data such as lidar point cloud intensity, UAV image grayscale, and sensor state data are dynamically fused to generate a high-precision three-dimensional construction map, which can ensure the real-time and accuracy of the map. The fused three-dimensional map provides spatial environmental features (such as terrain undulation, obstacle distribution) for the generation of the construction plan in the subsequent steps and provides a position reference for resource scheduling, solving the problem of progress lag caused by data isolation in traditional methods.

[0006] Optionally, the construction plan planning model is constructed, and based on the three-dimensional construction map, the optimal construction plan is generated by using the construction plan planning model, including the following steps: Construct a construction plan planning model and a construction progress plan reward function; analyze the risks of the construction environment, and based on the three-dimensional construction map, use the risks, the construction plan planning model, and the construction progress plan reward function to generate the optimal construction plan. The present invention obtains the optimal construction plan by comprehensively considering various on-site factors through the construction progress plan reward function and the risk assessment of the construction environment, which is further beneficial to improving the collaborative control ability of the construction progress.

[0007] Optionally, the construction progress plan reward function satisfies the following formula: , where represents the construction progress plan reward at the moment, represents the number of days ahead of schedule, represents the amount of cost overrun, represents the variance of resource utilization rate, represents the risk probability at the moment.

[0008] Optionally, the analysis of the risks of the construction environment satisfies the following formula: , where represents the risk probability at the moment, denote activation function denote the weights of the convolutional kernel in the time dimension denote historical risk data at time denote the weights of the convolutional kernel in the spatial dimension denote adjacent construction segments real-time status data of

[0009] Optionally, the dynamic scheduling and allocation of resources according to the optimal construction plan to obtain a resource allocation plan includes the following steps: Construct an optimization objective function for resource allocation and schedule adjustment and a resource scheduling constraint model; combine the optimization objective function for resource allocation and schedule adjustment, the resource scheduling constraint model, and the optimal construction plan to obtain the resource allocation plan. The present invention further improves the rationality of the construction schedule plan by optimizing resource allocation and schedule adjustment.

[0010] Optionally, the optimization objective function for resource allocation and schedule adjustment satisfies the following formula: , where denote resources unit cost of denote resources allocation volume on the current day denote the shortage penalty coefficient, default value 1.2 denote resources demand threshold in the current construction stage (dynamically calculated according to the plan).

[0011] Optionally, the resource scheduling constraint model includes: Fluctuation limit of resource allocation volume, satisfying: Upper and lower limit constraints of resource allocation volume, satisfying: , where denote resources daily change in allocation volume of denote resources allocation volume of the previous day denote resources allocation volume denote the upper limit of total resource supply denote resources minimum safety inventory of

[0012] Optionally, the multi-source data-driven intelligent collaborative control method for construction progress further includes the following steps: Set the risk threshold of the construction environment; complete risk warning according to the risk threshold.

[0013] The present invention completes risk early warning by evaluating the risks of the construction environment and comparing them with risk thresholds, which is beneficial to improving the efficiency of safety production.

[0014] In a second aspect, in order to efficiently execute a multi-source data-driven intelligent collaborative control method for construction progress provided by the present invention, the present invention also provides a multi-source data-driven intelligent collaborative control system for construction progress, including a processor, an input device, an output device, and a memory. The processor, the input device, the output device, and the memory are interconnected. Among them, the memory is used to store a computer program, the computer program contains program instructions, and the processor is configured to call the program instructions to execute a multi-source data-driven intelligent collaborative control method as described in the first aspect of the present invention. The multi-source data-driven intelligent collaborative control system for construction progress of the present invention has a compact structure and stable performance, and can stably execute a multi-source data-driven intelligent collaborative control method provided by the present invention, further improving the overall applicability and practical application ability of the present invention. Description of the Drawings

[0015] Figure 1 It is a flowchart of a multi-source data-driven intelligent collaborative control method for construction progress provided by an embodiment of the present invention; Figure 2 It is a framework diagram of a multi-source data-driven intelligent collaborative control system for construction progress provided by an embodiment of the present invention. Detailed Embodiments

[0016] The specific embodiments of the present invention will be described in detail below. It should be noted that the embodiments described here are only for illustrative purposes and are not used to limit the present invention. In the following description, in order to provide a thorough understanding of the present invention, a large number of specific details are set forth. However, it is obvious to those of ordinary skill in the art that the present invention does not have to be practiced with these specific details. In other instances, well-known circuits, software, or methods have not been specifically described to avoid obscuring the present invention.

[0017] Throughout the specification, the reference to "an embodiment", "embodiments", "an example", or "examples" means that a particular feature, structure, or characteristic described in connection with the embodiment or example is included in at least one embodiment of the present invention. Thus, the phrases "in an embodiment", "in embodiments", "an example", or "examples" appearing throughout the specification do not necessarily all refer to the same embodiment or example. In addition, the specific features, structures, or characteristics may be combined in any suitable combination and / or sub-combination in one or more embodiments or examples. In addition, those of ordinary skill in the art should understand that the drawings provided herein are for illustrative purposes only and are not necessarily drawn to scale.

[0018] Please refer to Figure 1 , to solve the problems of data islands, lagging progress, and lagging response in traditional highway engineering construction, and significantly improve construction efficiency and safety. The present invention provides a multi-source data-driven intelligent collaborative control method for construction progress, as Figure 1 shown. In one embodiment, the method includes the following steps: S1. Obtain multi-source data at the construction site.

[0019] In an embodiment, obtaining multi-source data at the construction site includes obtaining lidar data, drone cruise data, and sensor data, and receiving multi-source data through the Apache Kafka stream processing platform, including lidar point clouds, drone images, sensor status, and manually filled logs.

[0020] Specifically, deploy a high-precision lidar (model: Velodyne VLP-16) every 50 meters along the longitudinal axis of the construction area, with a scanning frequency of 20 Hz, covering the entire cross-section of the tunnel (diameter ≤ 15 meters), point cloud accuracy of ±2 cm, and real-time generation of three-dimensional terrain data; use a DJI multi-rotor drone to perform 2 full-area cruises per day (8:00 am, 6:00 pm), with a flight altitude of 50 meters, equipped with a Zenmuse H20T hybrid sensor, collect image data with a resolution of 5 cm / px, and achieve centimeter-level spatial alignment through RTK positioning; deploy IoT sensor nodes on each construction machine (excavator, roller, etc.) in a 1:1 ratio, and monitor the mechanical fuel consumption, rotation speed, vibration intensity, and ambient temperature and humidity in real time, with a sampling frequency of 100 Hz, and transmit the data to the edge node through LoRa wirelessly. Further, deploy an NVIDIA Jetson AGX Xavier edge computing device at the construction site, with a built-in 512-core Volta GPU, supporting TensorRT acceleration, responsible for real-time processing of lidar point clouds, drone images, and sensor data, with a processing delay < 100 ms, communicate using a 5G private network, with a bandwidth ≥ 1 Gbps, end-to-end delay < 20 ms, supporting parallel transmission of multi-source data and real-time interaction with the remote cloud server.

[0021] Further, after obtaining the data, it is also necessary to map the lidar point cloud intensity value to [0, 1] and standardize the grayscale of the drone image.

[0022] Specifically, mapping the lidar point cloud intensity value to [0, 1] satisfies the following formula: , where represents the original point cloud intensity value of the lidar at the coordinate , represents the minimum reflection intensity calibrated by the lidar device, Represents the maximum reflection intensity for lidar device calibration.

[0023] The gray-scale normalization of the UAV image satisfies the following formula: , where, represents the gray-scale value of the UAV image at the pixel point.

[0024] S2. Dynamically fuse the multi-source data to obtain a 3D construction map.

[0025] In the embodiment, the dynamically fusing the multi-source data to obtain a 3D construction map satisfies the following formula: , where, represents the coordinate at the time of the fused data value, represents the dynamic weight coefficient and satisfies: , represents the point cloud intensity value mapping of the lidar at the coordinate , represents the gray-scale normalization value of the UAV image at the coordinate , represents the state vector of the sensor at the time

[0026] Furthermore, the dynamic weight coefficient of the lidar data is updated in real time through Kalman filtering, the value ranges are all , and the state vector includes the mechanical fuel consumption percentage, rotational speed, environmental temperature and humidity, etc., which are dimensionless.

[0027] S3. Construct a construction plan planning model, and generate an optimal construction plan based on the 3D construction map.

[0028] Specifically, the constructing a construction plan planning model and generating an optimal construction plan based on the 3D construction map includes the following steps: S31. Construct a construction plan planning model and a construction progress plan reward function.

[0029] Specifically, the construction plan planning model uses the PyTorch framework to construct a deep reinforcement learning model. The Actor-Critic framework is adopted. The Actor network is a 3-layer fully connected neural network (256-128-64 nodes), and the Critic network has a dual-stream structure, which processes spatial map features and temporal state features respectively.

[0030] Furthermore, the Actor network includes: Input layer: Receives the state vector, including the rasterized data of the 3D construction map (resolution 10m×10m, encoded as a 2D matrix of 64×64), the resource inventory (such as concrete, steel, etc., normalized to 0,1), the real-time risk probability (calculated by ST-CNN), and the weather index (such as rainfall probability, wind speed, normalized to 0,1). Network structure: A 3-layer fully connected neural network with the number of nodes being 256-128-64 in sequence, the activation function being ReLU (Rectified Linear Unit), and the output layer using the Softmax function to generate the priority probability distribution of the construction tasks.

[0031] The Critic network includes: Spatial feature stream: Processes the raster data of the 3D construction map, extracts spatial features through a convolutional layer (Conv2D, kernel size 3×3, number of channels 32), and the output is a 128-dimensional vector. Temporal feature stream: Processes the temporal state data (resource inventory, risk probability, weather index), extracts temporal features through an LSTM layer (number of hidden units 64), and the output is a 64-dimensional vector. Feature fusion: Concatenates the spatial features and temporal features into a 192-dimensional vector, and outputs the state value score through a fully connected layer (number of nodes 128→64).

[0032] Furthermore, the construction schedule reward function satisfies the following formula: , where represents the construction schedule reward at time represents the number of days ahead of schedule, represents the amount of cost overrun, represents the variance of resource utilization rate, and satisfies: , reflecting the volatility of resource allocation, dimensionless), represents the risk probability at time

[0033] S32. Analyze the risks of the construction environment, and based on the 3D construction map, use the risks, the construction plan planning model, and the construction schedule reward function to generate the optimal construction plan.

[0034] In the embodiment, the analysis of the risks of the construction environment satisfies the following formula: , where represents the risk probability at time ), represents Activation function Represents the weight of the time - dimension convolutional kernel (dimension: 3×1), capturing the trend of historical data in the past 3 days Represents The historical risk data at time (including the number of mechanical failures, geological settlement, etc., dimensionless), Represents the weight of the spatial - dimension convolutional kernel (dimension: 3×3), correlating data of 8 adjacent construction sections Represents the adjacent construction section 's real - time status data (such as vibration intensity (mm / s), temperature and humidity (%))

[0035] Risk refers to factors that may have a negative impact on progress, cost, or safety during the construction process, including but not limited to: ①Mechanical failures (such as abnormalities in excavators, rollers); ②Environmental interferences (such as construction stagnation caused by rainfall, strong winds); ③Resource shortages (such as insufficient supply of concrete, steel); ④Safety accidents (such as collapses, geological settlements); ⑤Coordination conflicts (such as resource competition or process conflicts among multiple construction sections).

[0036] Specifically, by using ST - CNN (Spatio - temporal Convolutional Neural Network), through the combined action of the time - dimension convolutional kernel and the spatial - dimension convolutional kernel, combining historical risk data and the status data of adjacent construction sections, the risk probability is finally output

[0037] First, input data: 1) Time - dimension input: Historical risk data in the past 3 days (such as the number of mechanical failures, geological settlement, etc.); 2) Spatial - dimension input: Real - time status data of 8 adjacent construction sections (such as vibration intensity, temperature and humidity).

[0038] Secondly, convolution operation: 1) Time convolution: Feature extraction is performed on historical risk data through a 3×1 convolutional kernel to capture the trend in the time series (such as the periodic fluctuations of failure frequencies); 2) Spatial convolution: Feature extraction is performed on data of adjacent construction sections through a 3×3 convolutional kernel to analyze spatial correlation (such as the impact of vibration in one section on the geological stability of adjacent sections); 3) Activation function ( ) The weighted sum of the convolution output is mapped to a risk probability value in the range of [0, 1] through the Sigmoid function ( )), representing the likelihood of risk occurring in the current construction section at the next time step

[0039] Through the joint analysis of spatio-temporal convolution, the spatio-temporal correlation of risks in a dynamic construction environment can be quantified. For example: (1) Temporal correlation: Mechanical failures for consecutive days may indicate systemic risks; (2) Spatial correlation: Excessive vibration in a certain construction section may cause geological loosening in adjacent sections.

[0040] Furthermore, based on the rasterized data (10m×10m resolution) of the three-dimensional construction map, using the risks, the construction plan planning model, and the construction schedule reward function, the optimal construction plan is generated, including the following steps: Define the state space. The inputs include construction progress (0 - 100%), resource inventory (real-time inventory of concrete, steel, etc., unit: tons), weather index (normalized to 0 - 1, such as rainfall probability, wind speed), and risk probability assessment value. Furthermore, based on the rasterized data (10m×10m resolution) of the three-dimensional construction map, use the MADDPG (Multi-Agent Deep Deterministic Policy Gradient) algorithm for iterative optimization according to the state space. The training period is 7 days, the optimizer is Adam (learning rate 0.001), and the convergence condition is that the loss function drops to <5%. During the training process, the model robustness is improved through the experience replay mechanism and the exploration-exploitation strategy. Offline training is carried out using 7-day historical data (including 200 construction scenarios), and the optimization goal is set to maximize the cumulative reward based on the reward function. After training, the model outputs the construction task priority sequence and resource allocation suggestions to generate the optimal construction plan.

[0041] Specifically, divide the three-dimensional construction map into construction units according to 10m×10m grids. Each grid contains terrain elevation, obstacle position, construction progress status (0 - 100%), resource requirements (such as concrete, steel, etc., unit: tons), real-time risk probability (0 - 1), and weather index (such as rainfall probability normalized to 0 - 1), which are encoded into a state vector (including progress, resource, risk, and environmental data). Adopt the Actor-Critic framework. The Actor network receives the rasterized map features and temporal data such as resources and risks, and outputs the task priorities of each construction unit; the Critic network evaluates the value of the construction plan, and combines the reward function to evaluate the pros and cons of the plan to generate the optimal construction plan. The optimal construction plan is output in the form of a Gantt chart, clarifying the time window, resource requirements, and risk response measures of each task.

[0042] Exemplarily, when the risk probability of a certain grid exceeds the threshold (such as 0.7), reduce the priority of this grid, and reallocate resources to low-risk areas. At the same time, punish the lagging progress through the reward function to ensure the robustness of the construction plan. If the terrain undulation degree of a certain grid > 5%, the model automatically marks it as a "high-difficulty construction area" and more mechanical resources need to be allocated; the risk data of adjacent grids (such as excessive vibration intensity) affects the risk assessment of the current grid through spatial convolution.

[0043] Furthermore, the training dataset is generated by fusing multi-source real-time data and historical data, including: (1) Real-time data: 1) Construction progress log (manually filled in, format: timestamp + completion percentage); 2) Resource inventory (concrete and steel inventory monitored by sensors); 3) Weather data (rain probability, wind speed, etc. provided by meteorological stations, normalized to 0-1); 4) Risk probability assessment value.

[0044] (2) Historical data: 1) Records of progress, resource consumption, and weather impacts of past construction projects; 2) Historical risk events (such as mechanical failures, geological settlements) and coping strategies.

[0045] Even further, data preprocessing is also required, including: 1) Normalization (such as weather index standardization), missing value imputation, and time series alignment; 2) Dynamically updating the weights of multi-source data through Kalman filtering to ensure input consistency.

[0046] Even further, the training set includes historical project data (such as the daily progress log, concrete consumption record, and mechanical failure events of a certain highway project) and simulated scenario data (such as coping strategies for a 3-day suspension of work due to rainfall).

[0047] The label of each training sample is the actual construction plan executed (task sequence and resource allocation optimized by manual experience). The model calculates the cross-entropy loss by comparing the predicted plan with the label , driving the update of the network weights of the construction plan planning model.

[0048] S4. Perform dynamic scheduling and allocation of resources according to the optimal construction plan to obtain a resource allocation plan.

[0049] In the embodiment, the performing dynamic scheduling and allocation of resources according to the optimal construction plan to obtain a resource allocation plan includes the following steps: S41. Construct an optimization objective function for resource allocation and schedule adjustment and a resource scheduling constraint model.

[0050] Specifically, the optimization objective function for resource allocation and schedule adjustment satisfies the following formula: , where represents the unit cost of resource , represents resource The allocation quantity on the current day, represents the shortage penalty coefficient, with a default value of 1.2, represents the resource demand threshold in the current construction stage (dynamically calculated according to the plan).

[0051] The resource scheduling constraint model includes: Resource allocation quantity fluctuation limit, satisfying: , Upper and lower limit constraints of the resource allocation quantity, satisfying: , where, represents the resource daily allocation quantity change, represents the resource allocation quantity of the previous day, represents the resource allocation quantity, represents the total resource supply upper limit, represents the resource minimum safety inventory.

[0052] S42. Combine the resource allocation and schedule adjustment optimization objective function, the resource scheduling constraint model, and the optimal construction plan to obtain the resource allocation plan.

[0053] Specifically, based on the Gurobi solver, optimize the resource allocation with a non-linear programming algorithm. Regarding the resource allocation problem as a multi-objective optimization problem, obtaining the optimal solution by setting the objective function and constraint conditions and using the Gurobi solver, i.e., the non-linear programming algorithm, is a conventional existing technology and will not be elaborated here.

[0054] In some other embodiments, the generated construction plan and real-time sensor data can also be saved on the blockchain platform, and the progress log hash value is written into the consortium chain, and the following formula is satisfied: , where, represents the construction plan hash value (a 256-bit string), represents the Secure Hash Algorithm, ensuring that the data is irreversible and unique.

[0055] Specifically, adopt the Hyperledger Fabric consortium chain, deploy 10 nodes (5 construction parties, 3 supervisors, and 2 suppliers), and the consensus algorithm is PBFT (Practical Byzantine Fault Tolerance) to ensure data consistency.

[0056] Performance indicators: Support automatic execution progress acceptance and payment of smart contracts, throughput ≥ 1000 TPS, data upload chain latency < 2 seconds, and SHA-256 encryption algorithm is used for hash value generation.

[0057] Furthermore, a risk threshold for the construction environment can be set, and based on the risk threshold, risk early warning can be completed.

[0058] In the embodiment, when occurs, a high-risk warning is pushed through the visual interface, and the response time is less than 2 hours.

[0059] In an example, the progress error of the generated construction plan is controlled within 8%, and the efficiency is increased by 60% compared with the traditional manual planning; at the same time, the resource utilization rate of concrete, steel, etc. exceeds 90%, and the out-of-stock rate drops to less than 5%.

[0060] Secondly, the security and reliability are significantly enhanced. The prediction accuracy of the ST-CNN model for risks such as collapses and mechanical failures reaches more than 95%, the response time is less than 2 hours, the accident rate is reduced by 85%, and the blockchain platform ensures that the progress logs and acceptance data cannot be tampered with, improving the multi-party collaboration efficiency by 40%.

[0061] Furthermore, the real-time performance and adaptability are optimized. When the construction plan planning model faces interferences such as rainfall and mechanical failures, the response speed of adaptively adjusting the construction plan is increased by 50%, and it can handle more than 90% of real-time fluctuations.

[0062] Finally, the embodiment achieves significant economic and environmental benefits. The resource scheduling algorithm reduces the total construction cost by 15% - 20%. The specific data is shown in Table 1.

[0063]

[0064] Table 1 Comparison table of economic and environmental benefits Exemplarily, in a certain tunnel construction project, the 5th - 7th grid (length 30m) is identified as a high-risk collapse area through the rasterized map ( ), the task sequence is dynamically adjusted, and the optimal construction plan is generated: 1. Pause the excavation in this area and give priority to constructing the 14th grid with low risk; 2. Call the standby concrete resources (increase from 80 tons in stock to 100 tons) to ensure the progress in other areas.

[0065] Through the calculation of the reward function, the adjusted progress is advanced by 2 days ( ), the variance of resource utilization rate drops by 30%, and the total reward value rises to +15.6, verifying the effectiveness of the method.

[0066] Please refer toFigure 2 , in an embodiment, to efficiently execute a multi-source data-driven intelligent collaborative control method for construction progress provided by the present invention, the present invention further provides a multi-source data-driven intelligent collaborative control system for construction progress, including: an input device, an output device, a processor, and a memory. The input device, the output device, the processor, and the memory are interconnected. The memory contains program instructions for the steps of the multi-source data-driven intelligent collaborative control method for construction progress. The multi-source data-driven intelligent collaborative control system of the present invention has a compact structure and stable performance, and can stably execute the multi-source data-driven intelligent collaborative control method of the present invention, further improving the overall applicability and practical application ability of the present invention.

[0067] In an embodiment, the so-called processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The input device may be used to obtain data information. The output device may be used to output the result obtained from the program instructions included in the computer program stored in the memory provided by the present invention. The memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. A part of the memory may also include a non-volatile random access memory.

[0068] In a possible implementation, the memory may include a program storage area and a data storage area. Among them, the program storage area may store an operating system and application programs required for at least one function, etc.; the data storage area may store data created during use. In addition, the memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. A part of the memory may also include NVRAM. The memory stores an operating system, operation instructions, executable modules, or data structures, or subsets thereof, or extended sets thereof. Among them, the operation instructions may include various operation instructions for implementing various operations. The operating system may include various system programs for implementing various basic tasks and processing hardware-based tasks.

[0069] In an embodiment, a storage medium is further provided. A computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the above multi-source data-driven intelligent collaborative control method for construction progress are implemented.

[0070] The storage medium may include: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc.

[0071] In summary, the present invention constructs a three-dimensional construction map by integrating multi-source data, then constructs a construction plan planning model to obtain an optimal construction plan, and finally performs resource allocation based on the optimal construction plan, solving the problems of data islands, progress lag, and response lag in traditional highway engineering construction, and significantly improving construction efficiency and safety.

[0072] Therefore, the present invention effectively overcomes various disadvantages in the prior art and has high industrial utilization value.

[0073] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered by the scope recorded in the present invention.

Claims

1. A multi-source data-driven intelligent collaborative control method for construction progress, characterized in that, Including the following steps: Obtain multi-source data of the construction site; Dynamically fuse the multi-source data to obtain a three-dimensional construction map; Construct a construction plan planning model, and based on the three-dimensional construction map, use the construction plan planning model to generate an optimal construction plan; According to the optimal construction plan, perform dynamic resource scheduling and allocation to obtain a resource allocation plan.

2. The intelligent collaborative control method for construction progress driven by multi-source data according to claim 1, wherein The dynamically fusing the multi-source data to obtain a three-dimensional construction map satisfies the following formula: , where represents the fusion data value at the time coordinate represents the dynamic weight coefficient and satisfies: , represents the point cloud intensity value mapping of the lidar at the coordinate and represents the gray-scale normalization value of the UAV image at the coordinate , represents the state vector of the sensor at time 3. The intelligent collaborative control method for construction progress driven by multi-source data according to claim 1, wherein The constructing a construction plan planning model, based on the three-dimensional construction map, using the construction plan planning model to generate an optimal construction plan includes the following steps: Construct a construction plan planning model and a construction progress plan reward function; Analyze the risks of the construction environment, and based on the three-dimensional construction map, use the risks, the construction plan planning model and the construction progress plan reward function to generate the optimal construction plan.

4. The intelligent collaborative control method for construction progress driven by multi-source data according to claim 3, characterized in that The construction progress plan reward function satisfies the following formula: , where represents the construction progress plan reward at time represents the number of days ahead of schedule, represents the amount of cost overrun, represents the variance of resource utilization rate, represents the risk probability at time 5. The intelligent collaborative control method for construction progress driven by multi-source data according to claim 3, characterized in that, The analyzing the risks of the construction environment satisfies the following formula: , where represents the risk probability at time represents the activation function represents the weight of the time - dimension convolutional kernel represents the historical risk data at time represents the weight of the space - dimension convolutional kernel represents the adjacent construction section of the real - time status data 6. The intelligent collaborative control method for construction progress driven by multi-source data according to claim 1, wherein The performing dynamic resource scheduling and allocation according to the optimal construction plan to obtain a resource allocation plan includes the following steps: Construct an optimization objective function for resource allocation and schedule adjustment and a resource scheduling constraint model; Combine the optimization objective function for resource allocation and schedule adjustment, the resource scheduling constraint model and the optimal construction plan to obtain the resource allocation plan.

7. The intelligent collaborative control method for construction progress driven by multi-source data according to claim 6, characterized in that The optimization objective function for resource allocation and schedule adjustment satisfies the following formula: , where represents the unit cost of the resource , represents the allocation volume of the resource on the current day, represents the shortage penalty coefficient, with a default value of 1.2, represents the resource demand threshold in the current construction stage.

8. The intelligent collaborative control method for construction progress driven by multi-source data according to claim 6, characterized in that, The resource scheduling constraint model includes: Fluctuation limit of resource allocation quantity, satisfying: ； Upper and lower limit constraints of resource allocation quantity, satisfying: ； Among them, represents the daily allocation change of the resource , represents the previous day's allocation of the resource , represents the allocation of the resource , represents the total resource supply upper limit, represents the resource 's minimum safety stock.

9. The intelligent collaborative control method for construction progress driven by multi-source data according to claim 1, characterized in that It also includes the following steps: Set a risk threshold for the construction environment; Complete risk early warning according to the risk threshold.

10. A multi-source data-driven intelligent collaborative control system for construction progress, characterized in that, The multi-source data-driven intelligent collaborative control system for construction progress includes: an input device, an output device, a processor, and a memory. The input device, the output device, the processor, and the memory are interconnected. The memory includes program instructions, and the program instructions are used to execute the multi-source data-driven intelligent collaborative control method for construction progress according to any one of claims 1-9.

Images