A combined operating platform rapid dispatching system and method for frame column construction

Abstract

The present invention provides a combined operating platform rapid scheduling system and method for frame column construction, which relates to the field of rapid scheduling technology. The system includes: a mapping platform module, which is used to use a dual-depth Q network and combine target site image data and construction plans to formulate a platform movement plan. A column-specific adaptation segmentation module, which is used to formulate an adaptability strategy based on the operating platform function and frame column feature data, and divide the frame column building into multiple stages according to the flow section according to the construction plan. A work type position matching module, which is used to determine the work type according to the different work types required in different stages, and formulate a personnel allocation strategy based on the real-time work type personnel position. An adaptive scheduling module, which is used to adjust the platform movement plan according to the adaptability strategy and the personnel allocation strategy to obtain a platform scheduling method. The present invention has achieved the beneficial effects of improving scheduling accuracy and safety, improving construction efficiency and resource utilization, and enhancing dynamic environment adaptability and intelligence level.

Description

Technical Field

[0001] The present invention relates to the field of rapid dispatching technology, and in particular to a combined operating platform rapid dispatching system and method for frame column construction. Background Art

[0002] During frame column construction, the combined operating platform serves as the core equipment supporting key processes such as rebar binding, formwork installation, and concrete pouring. Its scheduling efficiency and rationality are directly related to the speed of construction progress and the level of operational safety. However, the current platform scheduling model in the industry is still dominated by manual experience, resulting in numerous pain points that need to be addressed, seriously restricting the improvement of construction efficiency.

[0003] First, the platforms are severely inadequately adapted to construction needs. Existing scheduling methods lack a systematic link between frame column characteristics and platform functionality. Construction workers often choose platform types based on experience. For example, when constructing special-shaped columns, the misuse of rectangular platforms can lead to narrow working space at corners, making it difficult for workers to efficiently tie rebar. For large, heavily loaded columns, using conventional lightweight platforms can cause deformation due to insufficient load-bearing capacity, potentially leading to safety accidents.

[0004] Secondly, there was a lack of linkage between platform scheduling, personnel allocation, and flow section construction. For example, after the platform was moved to the work point as planned, the carpentry team was found to be still working in another area, leaving the platform idle. Alternatively, the concrete of the columns in the previous flow section had not yet set, but the scheduling system had already moved the platform to a subsequent section, disrupting the process logic. Furthermore, there was a significant mismatch between personnel skills and platform operation requirements. For example, having novices operate complex platforms with angle adjustment functions not only prolonged commissioning time but also potentially affected construction quality due to improper operation.

[0005] Furthermore, it has poor adaptability to dynamic construction environments. Construction site conditions are constantly changing: temporary material storage areas occupy the originally planned path, tower crane operating radius suddenly overlaps the platform movement route, and workers are temporarily absent due to emergencies. Traditional fixed scheduling solutions are unable to respond to these changes in real time, often leading to platform movement path conflicts and construction delays.

[0006] In addition, the traditional scheduling model lacks a systematic constraint system and insufficiently integrates key factors such as platform space parameters (such as operating radius), process requirements (such as the width of the prefabricated column lifting channel), and personnel skill levels. The decision-making process is highly subjective and has low accuracy, making it difficult to form replicable scheduling standards.

[0007] Therefore, constructing a combined operating platform rapid scheduling system and method for frame column construction has become an urgent need to improve the efficiency and safety of frame column construction. Summary of the Invention

[0008] The present invention provides a combined operating platform rapid scheduling system and method for frame column construction, which are used to solve the defects of the prior art such as poor scheduling adaptability, disconnected process connection and insufficient scheduling flexibility.

[0009] In one aspect, the present invention provides a combined operating platform rapid dispatching system for frame column construction, comprising:

[0010] The mapping platform module is used to collect target site image data in real time and obtain the construction plan. It uses a dual deep Q network and combines the target site image data and construction plan to formulate a platform movement plan.

[0011] The column-specific adaptation segmentation module is used to extract the characteristic data of frame columns according to the construction plan, and formulate an adaptability strategy based on the operating platform functions and the characteristic data of frame columns, and divide the frame column building into multiple stages according to the flow section according to the construction plan.

[0012] The job type and position allocation module is used to determine the job type according to the different jobs required at different stages, and to formulate personnel allocation strategies based on the real-time job type and personnel location.

[0013] The adaptive scheduling module is used to adjust the platform movement plan according to the adaptability strategy and the personnel allocation strategy to obtain the platform scheduling method.

[0014] The present invention provides a combined operating platform rapid dispatching system for frame column construction, wherein the drawing platform module includes:

[0015] The state feature extraction unit is used to preprocess the target site image data, extract high-level features through convolutional neural networks, and extract key constraints from construction data and convert them into state variables.

[0016] The space definition unit is used to define the action space according to the physical characteristics of the combined operation platform and meet the physical constraints.

[0017] Function formulation unit, used to formulate reward functions based on efficiency, safety, plan matching and task priority.

[0018] The decision model generation unit is used to train the intelligent agent through a dual network structure to learn the mapping relationship between state variables, action space and reward function to generate a movement decision model.

[0019] The plan generation unit is used to generate a platform movement plan based on the movement decision model, combined with the target site image and construction plan.

[0020] The present invention provides a combined operating platform rapid dispatching system for frame column construction, wherein the decision model generating unit generates a mobile decision model in the following steps:

[0021] Two deep Q networks with the same structure are constructed, divided into an evaluation network and a target network. Both take state variables as input and output the action value of each action space.

[0022] Execute the action space, obtain the reward function and enter the new state to store as the experience pool.

[0023] Randomly sample batches of data from the experience pool and select the optimal action in the current state space through the target network.

[0024] The target network is used in conjunction with the reward function to evaluate the value of the optimal action in the new state to obtain the target value.

[0025] The mean square error function is defined as the loss function, and the parameters of the evaluation network are updated through gradient descent to narrow the gap between the target value and the actual value. The updated parameters are transmitted to the target network at preset intervals to generate a mobile decision model.

[0026] The present invention provides a combined operating platform rapid dispatching system for frame column construction, wherein the steps of extracting frame column feature data by a column feature adaptation segmentation module include:

[0027] The cross-sectional characteristics, spatial position characteristics, quantity and number of the frame columns are extracted from the construction plan as spatial and morphological parameters.

[0028] Concrete strength grade, reinforcement configuration, embedded parts information, seismic structure and construction process type are extracted as structural and material parameters.

[0029] The process time, process connection relationship, area priority and supporting equipment association of each frame column are taken as timing features.

[0030] The spatial and morphological parameters, structural and material parameters and time series characteristics are converted into a unified format to obtain the frame column characteristic data, and a database is established for storage.

[0031] The present invention provides a combined operating platform rapid dispatching system for frame column construction, wherein the steps of formulating an adaptability strategy by a column characteristic adaptation segmentation module include:

[0032] The functions of the combined operating platform are defined in terms of structural form, load-bearing capacity, movement and positioning capability, and steering function, and a platform capability library is established by classifying it according to functional scenarios and adaptive column types.

[0033] Based on the frame column feature data and platform capability library, the specific requirements of different features for platform functions are clarified to form a demand-capability mapping table.

[0034] Determine adaptability goals based on construction efficiency, construction safety, and construction costs, and formulate adaptability strategies based on the demand-capability mapping table.

[0035] The present invention provides a combined operating platform rapid scheduling system for frame column construction, wherein the column-specific adaptation segmentation module divides the frame column construction into multiple stages, including the following steps:

[0036] The guiding principles for division are determined based on the principles of engineering quantity balance, spatial continuity, process connection, structural integrity and resource adaptation.

[0037] The plan distribution and quantity of frame columns, structural constraints, and resource and construction period constraints are extracted from the construction plan to determine the basis for division.

[0038] The division units are determined according to the division guidelines, division basis and frame column characteristic data, and the flow section boundaries are clearly defined to divide the frame column building into multiple stages.

[0039] The present invention provides a combined operating platform rapid dispatching system for frame column construction, wherein the steps of determining the type of work by the work type allocation module include:

[0040] According to the operation content at different stages, the skill requirements are determined from steel bar layout and connection, template assembly and reinforcement, concrete vibration and measurement and positioning.

[0041] Match the corresponding construction job types according to skill requirements, and refine the job responsibilities of different job types during this stage.

[0042] According to the process combinations in different stages, the required types and configuration quantities of construction work are integrated, and the coordination nodes between the work types are clarified.

[0043] The present invention provides a combined operating platform rapid dispatching system for frame column construction, wherein the steps of formulating a personnel allocation strategy by a work type and position allocation module include:

[0044] Update the basic information of all construction workers according to job type, skill level, real-time location and current status, thereby building a real-time personnel information database.

[0045] Based on the real-time personnel information database and job requirements, the demand area is defined, personnel with the same job type and suitable status are screened, and the distance between the personnel and the target area is calculated to evaluate the personnel's suitability.

[0046] Sort the personnel suitability and select personnel from high to low. Based on the real-time location, plan the shortest movement path for each assigned personnel and ensure that the positions of personnel in the same process are coordinated to obtain the personnel allocation strategy.

[0047] The present invention provides a combined operating platform rapid scheduling system for frame column construction, wherein the steps of adapting the scheduling module to obtain a platform scheduling method include:

[0048] Adaptation constraints are extracted from the adaptability strategy based on platform functions, platform space and processes, and personnel constraints are extracted from the personnel allocation strategy based on personnel location, time coordination and personnel skills.

[0049] Compare the platform mobility plan with the adaptation constraints and personnel constraints in terms of functional adaptability, spatial and path conflicts, and temporal coordination to identify conflict points.

[0050] The platform scheduling method is obtained by adjusting the four dimensions of platform type, path, time and location according to the conflict points.

[0051] In another aspect, the present invention provides a method for rapid dispatching of a combined operating platform for frame column construction, comprising:

[0052] The target site image data is collected in real time, and the construction plan is obtained. The platform movement plan is formulated by using a dual deep Q network and combining the target site image data and the construction plan.

[0053] The frame column characteristic data is extracted according to the construction plan, and an adaptability strategy is formulated based on the operating platform function and the frame column characteristic data. The frame column building is divided into multiple stages according to the flow section according to the construction plan.

[0054] Determine the type of work based on the different types of work required at different stages, and formulate a personnel allocation strategy based on the real-time location of the workers in each type of work.

[0055] The platform mobility plan is adjusted according to the adaptability strategy and personnel allocation strategy to obtain the platform scheduling method.

[0056] The present invention provides a combined operating platform rapid scheduling system and method for frame column construction. By using a dual-depth Q network, the system achieves rapid scheduling of the operating platform, significantly improving construction efficiency. The system automatically formulates adaptability strategies and personnel allocation strategies based on the construction plan and frame column characteristic data, achieving optimal resource allocation. By real-time monitoring of the operating platform and construction environment, the system can promptly detect and warn of potential safety risks, enhancing construction safety. By optimizing construction scheduling and resource allocation, construction costs are effectively controlled and the risk of project overruns is reduced. The configuration of the operating platform and personnel allocation can be flexibly adjusted according to different construction stages and needs, improving the adaptability of construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0058] Figure 1 This is one of the flow diagrams of a combined operating platform rapid dispatching system and method for frame column construction provided by an embodiment of the present invention;

[0059] Figure 2 This is the second flow chart of a combined operating platform rapid dispatching system and method for frame column construction provided by an embodiment of the present invention;

[0060] Figure 3 yes Figure 1 Schematic diagram of the structure of the special adaptation segment module of the center column;

[0061] Figure 4 This is the third flow chart of a combined operating platform rapid dispatching system and method for frame column construction provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0062] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0063] The following combination Figure 1-Figure 3 The present invention describes a combined operating platform rapid dispatching system and method for frame column construction.

[0064] like Figure 1 and Figure 2 As shown, an embodiment of the present invention provides a combined operating platform rapid dispatching system for frame column construction, comprising:

[0065] The mapping platform module is used to collect target site image data in real time and obtain the construction plan. It uses a dual deep Q network and combines the target site image data and construction plan to formulate a platform movement plan.

[0066] High-definition network cameras (resolution ≥1080P, frame rate ≥25fps) are installed at key locations on the site (such as crane tops, scaffolding poles, and fence edges) to provide a panoramic view of the construction area. Real-time footage is synchronized to a backend server via wired (Ethernet) or wireless (5G / Wi-Fi) transmission. Industrial-grade drones (such as multi-rotors) equipped with high-definition cameras or laser radar (LiDAR) are used to capture aerial photography of the site along pre-set routes. Using the drone's built-in GPS positioning and attitude sensor, the spatial coordinates of the images are simultaneously recorded to obtain image data of the target site.

[0067] like Figure 3 As shown, the mapping platform module includes:

[0068] The state feature extraction unit is used to preprocess the target site image data, extract high-level features through convolutional neural networks, and extract key constraints from construction data and convert them into state variables.

[0069] High-level features include: Obstacle features: location coordinates and dimensions of static obstacles (material piles, scaffolding), and real-time location and movement speed of dynamic obstacles (cranes, personnel).

[0070] Traversable area features: safe path areas determined by semantic segmentation.

[0071] Target frame column characteristics: location and cross-sectional dimensions of the column to be constructed.

[0072] Key constraints include: time constraints: the deadline of the current process and the connection interval with the next process.

[0073] Priority constraint: The priority weight of the target column in the construction plan.

[0074] The space definition unit is used to define the action space according to the physical characteristics of the combined operation platform and meet the physical constraints.

[0075] Define a discrete or continuous action space based on the physical characteristics of the combined operating platform (such as movement direction, speed limit):

[0076] Discrete motion (suitable for small and medium-sized platforms):

[0077] A={move forward 1m, move backward 1m, move left 1m, move right 1m, stay 10s}

[0078] (The range of motion can be adjusted according to the scale of the site, such as using a 0.5m step length in areas with dense columns).

[0079] Continuous action (suitable for large self-propelled platforms):

[0080] A={moving direction θ (0-360°), moving distance d (0- )}

[0081] ( is the maximum single moving distance of the platform, limited by the power system).

[0082] Physical constraints may include: the center of the platform must not exceed the passable area after movement, and the safe distance from obstacles must be ≥ 0.5m (adjusted according to the platform size).

[0083] Function formulation unit, used to formulate reward functions based on efficiency, safety, plan matching and task priority.

[0084] The efficiency reward encourages rapid approach to the target and is negatively correlated with the current distance. The formula is:

[0085]

[0086] Where, It is an efficiency reward. is the distance weight coefficient, is the position coordinate of the current operating platform, are the coordinates of the target location.

[0087] Safety rewards include safe state rewards and dangerous state penalties, and the formula is expressed as:

[0088]

[0089] Where, When it is safe, When it is dangerous, is the clear distance between the platform and the nearest obstacle, is the minimum safe distance, and is the safety weight, It's a safety bonus.

[0090] Plan matching rewards on-time completion and penalizes delays. The formula is:

[0091]

[0092] Where, is the current time, It is a plan match. Ahead of time or on time, It's a delay. It's planning time. and is the time weight.

[0093] The decision model generation unit is used to train the intelligent agent through a dual network structure to learn the mapping relationship between state variables, action space and reward function to generate a movement decision model.

[0094] The steps to generate a mobile decision model include:

[0095] Two deep Q networks with the same structure are constructed, divided into an evaluation network and a target network. Both take state variables as input and output the action value of each action space.

[0096] Execute the action space, obtain the reward function and enter the new state to store as the experience pool.

[0097] Randomly sample batches of data from the experience pool and select the optimal action in the current state space through the target network.

[0098] Use the target network and the reward function to evaluate the value of the optimal action in the new state to obtain the target value, which is expressed as:

[0099]

[0100] Where, is the target value, is the immediate reward obtained by the current action, is the discount factor, is the target network, It's a new state. It is the optimal action.

[0101] The mean square error function is defined as the loss function, and the formula is expressed as:

[0102]

[0103] Where, is the number of samples of the sampled batch data, It is The target value of samples, Is the evaluation network for the The predictive value of a sample.

[0104] The network parameters are updated and evaluated through gradient descent to narrow the gap between the target value and the actual value. The updated parameters are transmitted to the target network at preset intervals to generate a mobile decision model.

[0105] The plan generation unit is used to generate a platform movement plan based on the movement decision model, combined with the target site image and construction plan.

[0106] The state variables are updated based on the real-time image of the target site. This is then fed into the evaluation network, which selects the action space with the highest value and outputs a movement plan, which may include the platform's movement path and time. If a new obstacle appears, the new state is updated and the target network makes a new decision.

[0107] The column-specific adaptation segmentation module is used to extract the characteristic data of frame columns according to the construction plan, and formulate an adaptability strategy based on the operating platform functions and the characteristic data of frame columns, and divide the frame column building into multiple stages according to the flow section according to the construction plan.

[0108] The steps of extracting frame column feature data by the column-specific adaptation segmentation module include:

[0109] The cross-sectional characteristics, spatial position characteristics, quantity and number of frame columns are extracted as spatial and morphological parameters from the design drawings, BIM models, construction schedules and technical briefing documents related to the construction plan.

[0110] Concrete strength grade, reinforcement configuration, embedded parts information, seismic structure and construction process type are extracted as structural and material parameters.

[0111] The process time, process connection relationship, area priority and supporting equipment association of each frame column are taken as timing features.

[0112] Section features include: Section type: rectangular, circular, L-shaped, T-shaped, etc. (read from the column plan annotation of the structural construction drawing or the BIM model properties).

[0113] Cross-sectional dimensions: length × width of rectangular columns (e.g. 600mm × 800mm), diameter of circular columns (e.g. φ500mm), length and angle of each side of special-shaped columns (e.g. L-shaped columns with a long side of 500mm, a short side of 300mm, and an angle of 90°).

[0114] Spatial position features include: Plane coordinates: positioning in the building axis coordinate system (such as the corresponding coordinates of "axis grid A-3 to A-4, axis 1-2 to 1-3" (X=3500mm, Y=2800mm), which can be directly exported from the BIM model or converted through the drawing scale).

[0115] Elevation information: The starting and ending heights of the column (such as "top of foundation to 3.600m" and "3.600m to 7.200m"), which determines the construction floor and height range.

[0116] Distribution density: The number and spacing of frame columns in a unit area (e.g., "two columns per bay, 8m apart"), used to determine the degree of site congestion.

[0117] Quantity and numbering include: extracting the total number of frame columns, numbering rules (such as "KZ1-KZ20"), and clarifying the unique characteristics corresponding to each number (such as KZ1 for standard columns and KZ5 for special-shaped columns).

[0118] Material characteristics include: Concrete strength grade: such as "C30" and "C60" (used to determine the platform's load-bearing requirements, high-strength concrete casting platforms require higher stability).

[0119] Rebar configuration: longitudinal reinforcement specifications and quantity, stirrup spacing and the extent of the reinforcement area affect the working space requirements of the rebar tying platform.

[0120] Special structural features include: embedded parts information: pipelines, equipment brackets, and connection nodes embedded in the column require operating space to be reserved in platform scheduling.

[0121] Earthquake-resistant structure: such as the anchorage length of longitudinal reinforcement, the reinforced area at the column root, and the corner reinforcement measures for special-shaped columns, which affect the accuracy requirements of formwork installation.

[0122] The type of construction technology: cast-in-place columns (requiring formwork support) or prefabricated columns (requiring hoisting and alignment) determines the functional requirements of the platform (e.g. prefabricated columns require a hoisting auxiliary platform).

[0123] Construction time nodes include: process time of each frame column: such as "KZ3 steel bar binding start time: 5th day, completion time: 7th day" and "KZ3 concrete pouring time: 8th day".

[0124] Process connection relationship: Dependency with beam and slab construction (such as "beam reinforcement binding can only be carried out 24 hours after KZ3 pouring is completed"), which is used to match the time window of platform scheduling.

[0125] Regional priority: Frame columns on critical routes (such as core columns that affect the overall construction period) are marked as high priority (such as priority coefficient ω=1.0), and non-critical route columns are marked as low priority (ω=0.5).

[0126] Association with supporting equipment: For example, "KZ8 pouring requires the use of No. 2 pump truck" and "Precast column KZ15 lifting relies on a 50t tower crane", and the coordination time requirements between the extraction platform and other equipment (such as avoiding the busy hours of the tower crane).

[0127] The spatial and morphological parameters, structural and material parameters and time series characteristics are converted into a unified format to obtain the frame column characteristic data, and a database is established for storage.

[0128] Each record in the database corresponds to a column and contains fields such as number, section type (code), section size, center coordinates, concrete strength, steel bar specifications, embedded parts location, construction start / completion time, flow section, priority coefficient, etc.

[0129] The steps for the adaptability strategy of the column-specific adaptation segmentation module include:

[0130] The functions of the combined operating platform are defined in terms of structural form, load-bearing capacity, movement and positioning capability, and steering function, and a platform capability library is established by classifying it according to functional scenarios and adaptive column types.

[0131] Based on the frame column feature data and platform capability library, the specific requirements of different features for platform functions are clarified to form a demand-capability mapping table.

[0132]

[0133] Determine adaptability goals based on construction efficiency, construction safety, and construction costs, and formulate adaptability strategies based on the demand-capability mapping table.

[0134] Adaptability strategies can include determining the corresponding operating platform based on the cross-sectional type and size of the frame column, and selecting different strategies for cast-in-place columns, precast columns, and columns with embedded pipelines, including cast-in-place columns (requiring formwork support):

[0135] The platform needs to reserve a template installation operation port (such as a removable baffle) to facilitate the template to be lifted into place.

[0136] During the formwork reinforcement stage, the platform is equipped with temporary fixings (such as clips tied to the formwork supports) to improve overall stability.

[0137] Prefabricated columns (need to be hoisted and aligned): select a platform with a leveling function (such as hydraulic legs for fine-tuning the height) and use a laser line projector to assist in correcting the verticality of the column.

[0138] Anti-fall baffles (height ≥ 1.5m) are installed on the edge of the platform to prevent tools from falling during lifting.

[0139] Columns with embedded parts / pipelines: An openable observation window (e.g. 300×300mm) is set on the platform corresponding to the position of the embedded parts to facilitate the positioning and inspection of the embedded parts.

[0140] The platform modules in the pipeline pre-buried area adopt a lightweight design (such as aluminum alloy material) to reduce the squeeze on the pre-buried pipelines.

[0141] The steps of the column-specific adaptation segmentation module to divide the frame column building into multiple stages include:

[0142] The guiding principles for division are determined based on the principles of engineering quantity balance, spatial continuity, process connection, structural integrity and resource adaptation.

[0143] Principle of balanced engineering workload: The number of frame columns and construction difficulty (such as the proportion of special-shaped columns and cross-sectional dimensions) in each flow section should be roughly balanced to avoid excessive workload in one section leading to concentration of resources and idleness in other sections.

[0144] Principle of spatial continuity: The flow sections should be relatively independent and continuous on the plane (such as adjacent areas divided by axes) to facilitate efficient turnover of material transportation and equipment movement (such as combined operating platforms and tower cranes) within the section.

[0145] Principle of process connection: The flow section must match the rhythm of the frame column construction process (rebar → formwork → concrete → maintenance) to ensure that after the previous section of the column completes a certain process, the next section of the column can be constructed continuously, forming a "flow operation" (for example, when the first section of the column is being cast, the formwork of the second section of the column is being installed).

[0146] Principle of structural integrity: avoid destroying the overall load-bearing units of the frame structure (for example, the core tube area should be treated as an independent flow section and should not be split. The area with dense beam-column nodes should be included in the same flow section to avoid node quality problems caused by cross-section construction).

[0147] Resource adaptation principle: The number of production lines must match the on-site resources (such as the number of combined operating platforms, the number of tower cranes, and the number of worker teams) (for example, 3 teams correspond to 3-4 production lines to ensure continuous operation of the teams).

[0148] The plan distribution and quantity of frame columns, structural constraints, and resource and construction period constraints are extracted from the construction plan to determine the basis for division.

[0149] Planar distribution and quantity of frame columns: Count the total number, number, and axis position of frame columns in the building plane.

[0150] Mark special columns: such as core columns (need to be constructed first), special-shaped columns (highly difficult to construct), and large-section columns (require special equipment) as the key considerations when dividing.

[0151] Structural constraints: Location of post-casting strips / construction joints: According to design requirements, the boundary of the flow section should coincide with the post-casting strip to avoid the impact of secondary construction on the structure.

[0152] Floor height and construction joint setting: In multi-story buildings, each floor can be used as an independent large flow section, or subdivided according to floor height (such as ≤6m) (for example, if a floor is 3.6m high, it can be divided into two small flow sections according to the column construction height).

[0153] Resource and construction period constraints: Upper limit of available resources: If there are 3 sets of combined operating platforms and 2 reinforcement teams on site, the number of flow sections should not exceed 4 (1 buffer section is reserved).

[0154] Total construction period requirement: Reversely calculate the number of flow sections based on the construction period of key lines (such as "the total construction period of frame columns is 30 days").

[0155] The division units are determined according to the division guidelines, division basis and frame column characteristic data, and the flow section boundaries are clearly defined to divide the frame column building into multiple stages.

[0156] The job type and position allocation module is used to determine the job type according to the different jobs required at different stages, and to formulate personnel allocation strategies based on the real-time job type and personnel location.

[0157] The steps to determine the type of work include:

[0158] According to the operation content at different stages, the skill requirements are determined from steel bar layout and connection, template assembly and reinforcement, concrete vibration and measurement and positioning.

[0159] Match the corresponding construction job types according to skill requirements, and refine the job responsibilities of different job types during this stage.

[0160] Rebar worker, matching processes: reinforcement layout, longitudinal reinforcement connection, stirrup binding, and embedded parts fixing.

[0161] Core Responsibilities: Check steel bar specifications and quantity according to drawings, cut and bend steel bars.

[0162] Use mechanical connection (threading, sleeve tightening) or welding (electroslag pressure welding) to connect the longitudinal reinforcement to ensure the quality of the joint.

[0163] Tie stirrups according to the spacing, paying special attention to the arrangement of stirrups in seismic reinforcement areas (such as at both ends of columns).

[0164] Cooperate with surveyors to locate embedded parts and ensure that the position deviation is ≤10mm.

[0165] Carpentry (formwork), matching process: formwork selection, assembly, support reinforcement, and calibration.

[0166] Core Responsibilities:

[0167] Select the formwork type (steel formwork / wooden formwork) according to the column section size (such as 600×600mm) and assemble it into the column shape.

[0168] Set up the support system (vertical poles, horizontal poles, diagonal braces) to ensure the rigidity of the formwork (to resist the lateral pressure of concrete).

[0169] Use a total station to calibrate the template verticality (deviation ≤ H / 1000) and axis position (deviation ≤ 5mm).

[0170] Install tension bolts (for large-section columns) to prevent the mold from expanding during pouring.

[0171] Concrete worker, matching processes: concrete spreading, vibration, plastering, and maintenance assistance.

[0172] Core Responsibilities:

[0173] Check the slump of concrete (e.g. C30 requires 180±20mm) to determine whether it meets the pouring requirements.

[0174] Cooperate with the pump truck to ensure that the concrete is evenly poured into the mold and vibrated in layers (the thickness of each layer is ≤500mm).

[0175] After vibration is completed, the surface should be smoothed in time to eliminate surface bubbles and cracks.

[0176] Assist maintenance workers to cover the film and record the start and end time of maintenance.

[0177] Surveyor, matching process: positioning and layout at each stage (throughout the entire process).

[0178] Core Responsibilities:

[0179] Before construction, lay out the column axis and section control lines on the working surface according to the BIM model or drawings.

[0180] After the steel bars are tied, the position of the longitudinal reinforcement should be checked, and the verticality and elevation should be calibrated after the formwork is installed.

[0181] Check the formwork position again before pouring concrete to avoid excessive deviation.

[0182] Scaffolder (platform operator), matching process: setting up, moving and dismantling the operating platform at each stage.

[0183] Core Responsibilities:

[0184] According to the column height and working range, set up or assemble a combined operating platform.

[0185] Ensure that platform guardrails and scaffolding comply with safety regulations (load ≥ 2kN / m²).

[0186] Move the platform to the target column according to the scheduling instructions, and put it into use after fixing.

[0187] Quality Inspector (part-time / full-time), matching process: acceptance at each stage (rebar, formwork, concrete).

[0188] Core Responsibilities:

[0189] Rebar acceptance: Check the number, spacing, connection quality, and position of embedded parts of longitudinal reinforcement.

[0190] Formwork acceptance: check stiffness, verticality, and tightness of joints.

[0191] Concrete acceptance: check slump, test block preparation, and implementation of maintenance measures.

[0192] According to the process combinations in different stages, the required types and configuration quantities of construction work are integrated, and the coordination nodes between the work types are clarified.

[0193] Phase 1 (Rebar Binding):

[0194] Required types of work: reinforcement workers (3-4 people / team, including 1 welder), surveyor (1 person, for preliminary positioning), scaffolder (1 person, for setting up the reinforcement work platform).

[0195] Collaborative Relationship: After the surveyors complete the layout, the scaffolders set up the platform, and the rebar workers tie the steel bars according to the lines. The welders are responsible for welding the longitudinal reinforcement, and upon completion, the quality inspector inspects and accepts the work.

[0196] Phase 2 (Template Installation)

[0197] Required trades: Carpenters (4-5 people / team, including 1 team leader), surveyors (1 person, to calibrate the template), scaffolders (1 person, to adjust the platform position to fit the template).

[0198] Collaborative relationship: As carpenters assemble formwork, surveyors calibrate verticality in real time. Scaffolders adjust the operating platform based on the formwork position to ensure adequate working space for carpenters.

[0199] Phase 3 (Concrete pouring)

[0200] Required jobs: Concrete workers (3 persons, including 1 vibrator), signalman (1 person, directing the pump truck), carpenter (1 person, standing by to observe whether the formwork is expanding).

[0201] Collaborative relationship: The signalman directs the pump truck to place concrete, while the concrete worker is responsible for vibrating and plastering. The carpenter is on hand throughout the entire process and immediately calls for the pouring to be stopped and corrected if any formwork deformation is detected.

[0202] Phase 4 (Demolding / Curing)

[0203] Required types of work: Carpenter (2 people, dismantling formwork), Concrete worker (1 person, maintenance), Handyman (1 person, cleaning formwork and site).

[0204] Collaborative relationship: After the carpenters remove the formwork, the handymen clean up the surface debris. The concrete workers cover the concrete with film, sprinkle water regularly, and record the curing status.

[0205] Steps in developing a staffing strategy include:

[0206] Update the basic information of all construction workers according to job type, skill level, real-time location and current status, thereby building a real-time personnel information database.

[0207] Based on the real-time personnel information database and job requirements, the demand area is defined, personnel with the same job type and suitable status are screened, and the distance between the personnel and the target area is calculated to evaluate the personnel's suitability.

[0208] Sort the personnel suitability and select personnel from high to low. Based on the real-time location, plan the shortest movement path for each assigned personnel and ensure that the positions of personnel in the same process are coordinated to obtain the personnel allocation strategy.

[0209] The adaptive scheduling module is used to adjust the platform movement plan according to the adaptability strategy and the personnel allocation strategy to obtain the platform scheduling method.

[0210] The steps of adjusting the platform scheduling method include:

[0211] Adaptation constraints are extracted from the adaptability strategy based on platform functions, platform space and processes, and personnel constraints are extracted from the personnel allocation strategy based on personnel location, time coordination and personnel skills.

[0212] Platform functional constraints may include: "L-shaped frame columns must be matched with customized corner module platforms" and "large-section columns (1.2m×1.2m) require heavy-load platforms (load ≥5kN / m²)".

[0213] Spatial constraints may include: the safe distance between the platform and the column (for example, the inner edge of the platform should be ≥0.3m away from the column when the formwork is installed), and the platform operating range coverage (for example, the steel bar binding platform needs to surround the column within a range of 1.5m).

[0214] Process constraints may include: "During the precast column hoisting stage, the platform needs to reserve a hoisting channel (width ≥ 2m)" and "During concrete pouring, the platform needs to avoid the pump truck's distribution radius (≥ 3m)."

[0215] Personnel location constraints may include: the platform must be docked at the current or upcoming work area of ​​the personnel (e.g., "4 carpenters have been assigned to the frame columns of Section 2, and the platform must be moved to the perimeter of Section 2").

[0216] Time coordination constraints may include: the platform's arrival time must be earlier than the personnel's expected arrival time (e.g., if the personnel is expected to arrive in 10 minutes, the platform must be in place within 8 minutes).

[0217] Personnel skill adaptation may include: "Customized formwork platforms operated by skilled carpenters must be equipped with quick adjustment functions to match the personnel's operating habits."

[0218] Compare the platform mobility plan with the adaptation constraints and personnel constraints in terms of functional adaptability, spatial and path conflicts, and temporal coordination to identify conflict points.

[0219] Functional adaptability verification includes: checking whether the platform type complies with the adaptability strategy: if the initial plan is "Use a rectangular platform for the L-shaped column in Section 2", it conflicts with the constraint of "L-shaped columns require customized corner platforms" and is marked as "Platform type needs to be changed".

[0220] Verify that the platform parameters meet the requirements: For example, in the initial plan for large-section column construction, the platform load is 3kN / m², which is lower than the constraint requirement of 5kN / m², and is marked as "Platform load capacity needs to be upgraded."

[0221] The space and path conflict check includes: checking whether the platform movement path matches the work area assigned to the personnel: for example, in the initial plan, the platform path passes through the "Section 1 Material Stacking Area", but the personnel have been assigned to "Section 2", so the path needs to be adjusted to the vicinity of Section 2.

[0222] Verify the operating space after the platform is in place: For example, in the initial plan, the distance between the platform and the column after docking is only 0.2m, which is less than the constraint requirement of 0.3m. As a result, carpenters cannot carry out operations, and it is marked as "need to adjust the docking position."

[0223] Time coordination verification includes: comparing the platform's estimated arrival time with the personnel's arrival time: for example, if the platform's initial plan estimates arrival at segment 2 in 15 minutes, but the personnel are expected to arrive in 10 minutes, there is a waste of time "people waiting for the platform", which is marked as "the platform movement time needs to be shortened".

[0224] Check the matching of the platform and process rhythm: For example, in the initial plan, the platform's residence time during the concrete pouring stage is 2 hours, but the process only takes 1.5 hours, resulting in the platform being idle, and marked as "the residence time needs to be compressed."

[0225] The platform scheduling method is obtained by adjusting the four dimensions of platform type, path, time and location according to the conflict points.

[0226] Adjust the platform type and parameters: Replace the platform that does not meet the functional constraints: for example, replace the platform of the L-shaped column in section 2 from a rectangular module to an L-shaped corner module to ensure that it fits the shape of the column.

[0227] Upgraded platform parameters: For example, the platform load for large-section columns has been increased from 3kN / m² to 6kN / m², and anti-slip pedals have been added (to meet the safety requirements of concrete workers during vibrating).

[0228] Optimize the movement path and docking position: Correct the path to match the personnel work area: Based on the coordinates of Section 2 assigned by the personnel (x=15-20m, y=10-15m), adjust the platform path from "via Section 1 → Section 3 → Section 2" to "directly via the east side channel of Section 2", shortening the distance (from 50m to 30m).

[0229] Adjust the docking position to meet space constraints: For example, increase the distance between the platform and the column from 0.2m to 0.4m to ensure sufficient operating space for carpenters. Also, avoid the pump truck's material distribution radius by adjusting the platform's docking point from the south to the north of the column.

[0230] Coordinate time nodes and stay duration: Compress platform movement time: By optimizing the path (such as shortening the distance as mentioned above) and increasing the movement speed (such as changing manual pushing to tower crane-assisted lifting), the platform arrival time can be shortened from 15 minutes to 7 minutes, which is earlier than the arrival time of personnel (10 minutes).

[0231] Adjust the dwell time to match the process rhythm: For example, the dwell time of the concrete pouring platform is shortened from 2 hours to 1.5 hours, which is consistent with the process duration, and then immediately moves to the next section (Section 3) to reduce idleness.

[0232] Adapt to personnel skills and operating habits: Adjust the platform operation interface: For example, for the template platform commonly used by skilled carpenters, add quick adjustment buttons (such as one-click vertical locking) to reduce personnel operation time.

[0233] Synchronize platform and personnel tool configuration: If the personnel allocation strategy includes "surveyor with laser line projector", the platform will add a laser receiving device to assist in rapid positioning.

[0234] like Figure 4 As shown, based on the same general inventive concept, the present invention also protects a method for rapid scheduling of a combined operating platform for frame column construction, the rapid scheduling method comprising:

[0235] The target site image data is collected in real time, and the construction plan is obtained. The platform movement plan is formulated by using a dual deep Q network and combining the target site image data and the construction plan.

[0236] The frame column characteristic data is extracted according to the construction plan, and an adaptability strategy is formulated based on the operating platform function and the frame column characteristic data. The frame column building is divided into multiple stages according to the flow section according to the construction plan.

[0237] Determine the type of work based on the different types of work required at different stages, and formulate a personnel allocation strategy based on the real-time location of the workers in each type of work.

[0238] The platform mobility plan is adjusted according to the adaptability strategy and personnel allocation strategy to obtain the platform scheduling method.

[0239] This embodiment provides a rapid scheduling system and method for a combined operating platform for frame column construction. By extracting frame column characteristics (cross-section, material, and process), and integrating them with platform functions to establish a "demand-capacity mapping table," this system eliminates the subjectivity of manual judgment and ensures precise alignment between the platform and construction requirements. Furthermore, by dynamically matching work types and personnel locations, the platform plan is adjusted based on time-based coordination constraints, achieving three-dimensional collaboration across the "platform, personnel, and process" framework, avoiding idle resources and disconnected connections. By utilizing a dual-depth Q network, movement plans are dynamically generated, combining real-time image data with construction plans. A "constraint adjustment method" is used to verify and optimize paths and times in real time, ensuring that the scheduling plan adapts to site changes. This system achieves the beneficial effects of improving scheduling accuracy and safety, enhancing construction efficiency and resource utilization, and enhancing adaptability and intelligence in dynamic environments.

[0240] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0241] 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 it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A combined operating platform rapid dispatching system for frame column construction, characterized in that: include: A mapping platform module is used to collect target site image data in real time and obtain a construction plan. It uses a dual deep Q network to combine the target site image data and the construction plan to formulate a platform movement plan. The mapping platform module includes: a state feature extraction unit, configured to pre-process the target site image data, extract high-level features through a convolutional neural network, and extract key constraints from the construction plan and convert them into state variables; A space definition unit, used to define the action space according to the physical characteristics of the combined operation platform and satisfy physical constraints; Function formulation unit, used to formulate reward functions based on efficiency, safety, plan matching and task priority; A decision model generating unit, configured to generate a movement decision model by training an intelligent agent through a dual-network structure to learn a mapping relationship between the state variables, the action space, and the reward function; The step of the decision model generating unit generating the movement decision model includes: Construct two deep Q networks with the same structure, divided into an evaluation network and a target network, both of which take the state variables as input and output the action value of each action space; Execute the action space, obtain the reward function and enter a new state to store as an experience pool; Randomly sampling batch data from the experience pool and selecting the optimal action in the current state space through the target network; Using the target network and the reward function to evaluate the value of the optimal action in the new state to obtain a target value; defining a mean square error function as a loss function, and updating the parameters of the evaluation network by gradient descent to reduce the gap between the target value and the actual value, transmitting the updated parameters to the target network at preset intervals to generate the movement decision model; a plan generating unit, configured to generate a plan for moving the platform according to the movement decision model, in combination with the target site image and the construction plan; A column-specific adaptation segmentation module is used to extract frame column feature data according to the construction plan, formulate an adaptability strategy based on the operating platform function and the frame column feature data, and divide the frame column building into multiple stages according to the flow section according to the construction plan; The step of formulating the adaptability strategy by the column-specific adaptation segmentation module includes: The functions of the combined operating platform are defined in terms of structural form, load-bearing capacity, movement and positioning capability, and steering function, and a platform capability library is established by classifying it according to functional scenarios and adaptable column types; Based on the frame column feature data and the platform capability library, specific requirements of different features on platform functions are clarified to form a requirement-capability mapping table; Determine the adaptability target based on construction efficiency, construction safety and construction cost, and formulate the adaptability strategy in combination with the demand-capacity mapping table; The job type allocation module is used to determine the job type according to the different jobs required at different stages, and formulate a personnel allocation strategy based on the real-time location of the workers in the job type; The adaptation scheduling module is used to adjust the platform movement plan according to the adaptability strategy and the personnel allocation strategy to obtain a platform scheduling method.

2. A combined operating platform rapid dispatching system for frame column construction according to claim 1, characterized in that: The step of extracting the frame column feature data by the column feature adaptation segmentation module includes: Extracting cross-sectional characteristics, spatial position characteristics, quantity and number of frame columns from the construction plan as spatial and morphological parameters; Extract concrete strength grade, reinforcement configuration, embedded parts information, seismic structure and construction process type as structural and material parameters; The process time, process connection relationship, regional priority and supporting equipment association of each frame column are used as time series features; The spatial and morphological parameters, structural and material parameters and the temporal characteristics are converted into a unified format to obtain the frame column characteristic data, and a database is established for storage.

3. The combined operating platform rapid dispatching system for frame column construction according to claim 1 is characterized in that: The column-specific adaptation segmentation module divides the frame column building into multiple stages, including: Determine the guiding principles for division based on the principles of engineering quantity balance, spatial continuity, process connection, structural integrity, and resource adaptation; Determining the basis for division based on the plane distribution and quantity of frame columns extracted from the construction plan, resource and construction period constraints, and structural constraints; The division units are determined according to the division guiding principles, the division basis and the frame column characteristic data, and the flow section boundaries are clearly defined to divide the frame column building into multiple stages.

4. The combined operating platform rapid dispatching system for frame column construction according to claim 1 is characterized in that: The step of determining the type of work by the job type allocation module includes: Determine the skill requirements for different stages of operations, including steel bar layout and connection, formwork assembly and reinforcement, concrete vibration, and measurement and positioning; Match the corresponding construction job types according to the skill requirements, and refine the job responsibilities of different job types in this stage; According to the process combinations in different stages, the required types and configuration quantities of construction work are integrated, and the coordination nodes between the work types are clarified.

5. The combined operating platform rapid dispatching system for frame column construction according to claim 1 is characterized in that: The step of formulating the personnel allocation strategy by the job type allocation module includes: Update the basic information of all construction workers based on their job type, skill level, real-time location, and current status, thereby building a real-time personnel information database; Based on the real-time personnel information database and job requirements, the demand area is defined, and personnel with the same job type and suitable status are screened. The distance between the personnel and the target area is calculated to evaluate the personnel's suitability; The personnel suitability is sorted, and personnel are selected from high to low. Based on the real-time position, the shortest moving path is planned for each assigned personnel, and the positions of personnel in the same process are ensured to be coordinated to obtain the personnel allocation strategy.

6. The combined operating platform rapid dispatching system for frame column construction according to claim 1, characterized in that: The step of adjusting the adaptation scheduling module to obtain the platform scheduling method includes: Extracting adaptation constraints from the adaptability strategy based on platform functions, platform space, and processes, and extracting personnel constraints from the personnel allocation strategy based on personnel location, time coordination, and personnel skills; Comparing the platform movement plan with the adaptation constraints and the personnel constraints in terms of functional adaptability, spatial and path conflicts, and time coordination to identify conflict points; The platform scheduling method is obtained by adjusting the conflict points from four dimensions: platform type, path, time and location.

7. A method for rapid dispatching of a combined operating platform for frame column construction, which uses a rapid dispatching system for a combined operating platform for frame column construction as claimed in any one of claims 1 to 6, characterized in that: The rapid scheduling method includes: Collect target site image data in real time and obtain a construction plan, and use a dual deep Q network to combine the target site image data and the construction plan to formulate a platform movement plan; Extracting frame column feature data according to the construction plan, formulating an adaptability strategy based on the operating platform function and the frame column feature data, and dividing the frame column building into multiple stages according to the flow sections according to the construction plan; Determine the type of work based on the different types of work required at different stages, and formulate a personnel allocation strategy based on the real-time location of workers in each type of work; The platform movement plan is adjusted according to the adaptability strategy and the personnel allocation strategy to obtain a platform scheduling method.

Images