High-rise building construction path dynamic planning method based on artificial intelligence

Abstract

The invention discloses a high-rise building construction path dynamic planning method based on artificial intelligence, and the method comprises the following steps: S1, obtaining data, and generating a construction task sequence and initial resource supply relation graph; s2, generating a current resource supply state, a construction rhythm change state and an operation interference factor, and forming a current actual construction state; s3, constructing a construction-resource adversarial graph model; s4, initializing a path search tree structure; s5, performing path sampling and evaluation through an improved heuristic Monte Carlo search algorithm, and outputting a plurality of feasible path branches and evaluation scores thereof; s6, executing path branch comparison and dynamic optimization to obtain a current optimal construction path; s7, generating a construction step adjustment suggestion and a resource allocation adjustment instruction; and S8, recording the optimal construction path and the evaluation score, and forming a construction path optimization result. The intelligent, self-adaptive and real-time optimization capabilities of high-rise building construction path planning are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of construction management and artificial intelligence optimization, and particularly relates to a high-rise building construction path dynamic planning method based on artificial intelligence. BACKGROUND

[0002] With the acceleration of urbanization and the continuous improvement of the level of building industrialization, high-rise building construction is facing multiple challenges such as complex construction tasks, dynamic resource allocation, and frequent environmental disturbances. How to achieve efficient and reliable construction path dynamic planning has become a key problem in construction organization design. Existing construction path planning methods are mostly based on static task scheduling, critical path analysis or linear resource scheduling model, mainly relying on fixed process logic and experience parameters for scheduling optimization, but there are generally the following problems in actual application:

[0003] The current method generally fails to effectively integrate the complex dependency relationship between construction tasks and the dynamic resource supply and demand matching, resulting in a lack of adaptability in path decision-making. The construction site has limited environmental perception capability, lacks identification and modeling of potential job interference factors in sensor data and log information, and is difficult to dynamically respond to unexpected events and schedule deviations in the construction process. Traditional optimization methods mostly use single heuristic or static scheduling algorithms, which cannot fully simulate the construction state evolution and resource interaction process, resulting in resource conflicts, sequence deviations and low efficiency in the actual execution of the generated path scheme. At the same time, the feedback of the construction state is not fully utilized, and there is a lack of continuous updating mechanism for the path scheme, resulting in unstable construction path optimization effect and difficulty in realizing intelligent scheduling control throughout the process.

[0004] Therefore, how to provide a high-rise building construction path dynamic planning method based on artificial intelligence is a problem that needs to be solved by those skilled in the art. SUMMARY

[0005] One object of the present application is to provide a high-rise building construction path dynamic planning method based on artificial intelligence, which uses an improved heuristic Monte Carlo search algorithm, integrates a multi-strategy hybrid simulation mechanism, a construction rhythm adaptive path adjustment mechanism and a construction environment risk dynamic modeling mechanism, and combines construction task sequences, resource supply states and job interference factors to simulate and optimize the path, having the advantages of more adaptive path planning results, strong response capability to construction dynamic environment and more reasonable construction resource allocation.

[0006] According to the high-rise building construction path dynamic planning method based on artificial intelligence, the following steps are included: S1, obtain the construction task sheet and the component installation sequence diagram, generate the construction task sequence and the initial resource supply relationship diagram; S2, acquire the current construction progress state data, combine the initial resource supply relationship graph to form the current resource supply state, generate the construction rhythm change state according to the task completion rate of each construction node, and combine the field sensing data and the construction log to extract the operation interference factor, and generate the current actual construction state; S3, based on the construction task sequence and the current resource supply state, a construction-resource confrontation graph model is constructed; S4, initialize the path search tree structure based on the construction-resource confrontation graph model, and input it into the improved heuristic Monte Carlo search algorithm; S5, using the improved heuristic Monte Carlo search algorithm, introducing a multi-strategy hybrid simulation mechanism in the simulation stage to execute construction state evolution and resource response simulation, and in the selection and backtracking stage, the construction rhythm adaptive path adjustment mechanism and the construction environment risk dynamic modeling mechanism are fused, the path evaluation value is dynamically corrected, the path sampling and evaluation are completed, and multiple feasible path branches and their corresponding evaluation scores are output; S6, based on the multiple feasible path branches and the evaluation scores, combining the construction-resource confrontation graph model, the construction rhythm change state and the operation interference factor, the path branch comparison and dynamic optimization are executed, and the optimal construction path under the current state is obtained; S7, compare the optimal construction path with the current actual construction state in node order, resource allocation and time window, and output resource allocation adjustment instructions; S8, record the optimal construction path and the evaluation score to form the construction path optimization result.

[0007] Optionally, the S1 specifically comprises: S11, acquire the construction task list, and according to the task number, operation type, duration and resource demand type recorded in the construction task list, sequentially arrange and build a construction task item table, and form a task initial arrangement set according to the task number order; S12, acquire the component installation sequence graph, read the component number, floor elevation and structure connection direction information of each component from the component installation sequence graph, map the component number to the corresponding construction task number, and build the directed connection relationship between tasks according to the connection direction to obtain a construction priority relationship graph; S13, according to the directed edge direction in the construction priority relationship graph, the topological sorting algorithm is used to sort the task initial arrangement set to generate a construction task sequence that meets the structure connection logic; S14, count the resource demand type of each task in the construction task item table, arrange to generate a resource demand sequence according to the construction task sequence order, and combine the resource number in the standard construction resource configuration table provided by the construction unit to establish a supply-demand mapping matrix between the task number and the resource number, and build an initial resource supply relationship graph.

[0008] Optionally, the S2 specifically comprises: S21, collecting construction logs and field sensing data, obtaining current task completion degree of each component in the construction task sequence, resource calling record and construction environment monitoring data, and generating construction progress state original data; S22, according to the component-resource configuration relationship recorded in the initial resource supply relationship graph, combining the current resource calling value of each component in the construction progress state original data, calculating the supply ratio of the component-resource corresponding relationship, constructing the current resource supply state matrix, and the matrix takes the component number and the resource number as the index, recording the ratio of actual supply and configuration; S23, according to the change sequence of the task completion degree of each component in the construction progress state original data with time, calculating the task completion rate according to the component number, and using the sliding time window to calculate the first derivative to form the task completion rate change curve, and generating the construction rhythm change state according to the curve clustering of the component; S24, analyzing the mutation signal in the construction environment monitoring data, combining the keywords and time records about event interference in the construction log, constructing the event interference tuple, including the interference position, time window and influence type; S25, associating and labeling the event interference tuple according to the component number, and obtaining the interference intensity score by weighted superposition, generating the operation interference factor set, each factor in the set contains component number, interference type, interference score and corresponding time window information; S26, structurally unifying the current resource supply state matrix, the construction rhythm change state diagram and the operation interference factor set to form the current actual construction state.

[0009] Optionally, the S3 specifically comprises: S31, obtaining the operation content, dependent task number, construction priority and required resource type corresponding to each task from the construction task sequence, taking each task as a task node, and sequentially numbering to form a task node set; S32, extracting the resource type, number and available quantity of each type of construction resource from the current resource supply state, establishing each type of resource as a resource node to form a resource node set; S33, according to the required resource type and quantity of each task node, establishing a resource connection edge between the task node and the corresponding resource node, and taking the resource requirement quantity as the edge attribute and adding it to the connection edge; S34, according to the dependent task number marked in each task in the construction task sequence, establishing a task connection edge between the task nodes to represent the sequence relationship of the construction tasks; S35, combine the task node set, the resource node set, the resource connection edge and the task connection edge into a heterogeneous graph structure, and append the numerical information of the current resource supply state and the construction rhythm change state in each node attribute, to generate a construction-resource confrontation graph model.

[0010] Optionally, the S4 specifically comprises: S41, based on the construction-resource confrontation graph model, extract the connection relationship between the task nodes in the construction task sequence and the available resource nodes in the resource supply state, take the current actual construction state as the initial root node, and initialize the path search tree structure; S42, according to the construction priority of the task nodes in the construction task sequence, expand the task state corresponding to the current root node, construct an expandable task node set, and generate corresponding child nodes in the path search tree; S43, in the process of generating the child nodes, in combination with the supply amount of each resource number in the current resource supply state, judge the resource matching condition of each task node, and label the resource constraint state tag for the task nodes that cannot be satisfied; S44, input the generated path search tree structure and the task nodes, resource state tags and time step index information therein into the improved heuristic Monte Carlo search algorithm.

[0011] Optionally, the S5 specifically comprises: S51, in the improved heuristic Monte Carlo search algorithm, construct a state space simulation model based on the current path search tree structure, and perform path branch simulation; in each round of simulation, simulate the construction sequence, resource consumption behavior and on-site state response process, and generate a path execution sequence; S52, in the simulation process, introduce a multi-strategy hybrid simulation mechanism, respectively adopt a heuristic scheduling strategy based on experience rules, a local optimization strategy based on reinforcement learning and a resource intervention strategy based on structure weight to dynamically guide the construction path execution sequence, and parallelly evaluate its construction feasibility and resource coordination degree between simulation nodes; S53, in the selection and backtracking phase of the path tree, fuse a construction rhythm adaptive path adjustment mechanism, dynamically adjust the path continuation direction based on the local rate change of each construction node in the construction rhythm change state, and improve the priority of the high-frequency operation area; S54, synchronously fuse a construction environment risk dynamic modeling mechanism, based on the frequency and influence duration of each event disturbance tuple in the operation disturbance factor set in the path branch, generate a path disturbance score and superimpose it into the path backtracking expectation value; S55, according to the execution sequence and simulation result of each path branch, calculate its task delay loss, resource conflict penalty and environmental risk score, respectively give a preset weight, and then weighted sum, to obtain an evaluation score; S56, record all path branches and their corresponding evaluation scores.

[0012] Optionally, the S6 specifically includes: S61, sort the multiple feasible path branches and their corresponding evaluation scores, and select the top m highest scoring paths as a candidate path set; S62, for each path in the candidate path set, sequentially perform path traversal operations to map its node order to the corresponding components and resource units in the construction-resource confrontation graph model, obtaining a mapping result; S63, based on the mapping result, calculate the resource preemption conflict rate between components, construction order feasibility, and interference event hit probability in each path, and combine the construction rhythm change state and the work interference factor to comprehensively score the overall execution stability and execution efficiency of the path; S64, according to the comprehensive scoring result, select the highest scoring path as the optimal construction path under the current state, and output the node order, resource allocation sequence, and key control period corresponding to the path.

[0013] Optionally, the S7 specifically includes: S71, compare the construction node order in the current actual construction state with the node order in the optimal construction path one by one, identify the inconsistent node positions and the degree of offset; S72, compare the current resource supply state with the resource configuration required by the nodes in the optimal construction path, and calculate the resource difference of each node; S73, compare the current construction rhythm change state with the construction time window in the optimal path, and identify the rhythm conflict interval; S74, generate current construction step adjustment suggestions according to the offset degree, resource difference, and rhythm conflict, including node rearrangement order and key task pre-operation instructions; S75, generate resource allocation adjustment instructions according to the resource difference and rhythm conflict, including resource reallocation suggestions and current construction step adjustment suggestions.

[0014] Optionally, the S8 specifically includes: S81, encode the optimal construction path, the construction-resource confrontation graph model state when the path is generated, the construction rhythm change state, and the work interference factor together to form a path state snapshot data record; S82, assign a unique index number to the path state snapshot data record, establish a state snapshot index set, and store the optimal construction path, the corresponding evaluation score, and the state snapshot index in association to form a searchable construction path optimization result database; S83、In the case of subsequent construction state update or path re-planning requirement, the state snapshot index set is called to perform fast search and reuse operation of similar path with the current state.

[0015] The beneficial effects of the present application are: (1) Enhance the dynamic adaptability of path planning: by introducing the construction rhythm self-adaptive path adjustment mechanism, the speed difference and bottleneck position in the construction progress can be identified in real time, the path planning result is dynamically corrected, the congestion and task accumulation in high-frequency construction area are effectively avoided, the continuity and smoothness of path execution are improved, and the adaptability of the planning result to the actual construction state is significantly enhanced.

[0016] (2) Improve the risk perception ability under complex working conditions: integrate the construction environment risk dynamic modeling mechanism, fully consider the space-time distribution and influence intensity of operation interference factors in the path search process, quantify the interference risk of path branches and integrate them into the evaluation system, effectively improve the robustness and fault tolerance of path planning, and ensure that stable and feasible construction paths can be output in complex environments.

[0017] (3) Realize precise coordination and efficient allocation of construction resources: by constructing the construction-resource confrontation graph model, the system integrates the construction task logic and resource supply and demand relationship, simulates and evaluates the resource allocation scheme in the Monte Carlo path search process, and selects the optimal path through the comprehensive score, realizes the collaborative optimization of task order, resource allocation and time window, and greatly reduces the resource conflict rate and project delay risk. BRIEF DESCRIPTION OF DRAWINGS

[0018] The accompanying drawings are included to provide a further understanding of the application, and constitute a part of the specification, together with the embodiments of the application, to explain the application, and do not constitute a limitation on the application. In the drawings:

[0019] Figure 1 A flowchart of a high-rise building construction path dynamic planning method based on artificial intelligence is proposed. DETAILED DESCRIPTION

[0020] The application will now be described in further detail with reference to the drawings. These drawings are simplified schematic diagrams, and only illustrate the basic structure of the application in a schematic manner, and therefore only show the components related to the application.

[0021] Reference Figure 1 A high-rise building construction path dynamic planning method based on artificial intelligence, comprising the following steps: S1, obtain the construction task list and component installation sequence diagram, generate the construction task sequence and initial resource supply relationship diagram; S2, acquire the current construction progress state data, combine the initial resource supply relationship graph to form the current resource supply state, generate the construction rhythm change state according to the task completion rate of each construction node, and combine the field sensing data and the construction log to extract the operation interference factor, and generate the current actual construction state; S3, based on the construction task sequence and the current resource supply state, a construction-resource confrontation graph model is constructed; S4, initialize the path search tree structure based on the construction-resource confrontation graph model, and input it into the improved heuristic Monte Carlo search algorithm; S5, using the improved heuristic Monte Carlo search algorithm, introducing a multi-strategy hybrid simulation mechanism in the simulation stage to execute construction state evolution and resource response simulation, and in the selection and backtracking stage, the construction rhythm adaptive path adjustment mechanism and the construction environment risk dynamic modeling mechanism are fused, the path evaluation value is dynamically corrected, the path sampling and evaluation are completed, and multiple feasible path branches and their corresponding evaluation scores are output; S6, based on the multiple feasible path branches and the evaluation scores, combining the construction-resource confrontation graph model, the construction rhythm change state and the operation interference factor, executing path branch comparison and dynamic optimization, and obtaining the optimal construction path under the current state; S7, comparing the optimal construction path with the current actual construction state in terms of node order, resource allocation and time window, and outputting resource allocation adjustment instructions; S8, record the optimal construction path and the evaluation score to form the construction path optimization result.

[0022] In the embodiment, the S1 specifically includes: S11, acquire the construction task list, and arrange and build a construction task item table according to the task number, operation type, duration and resource demand type recorded in the construction task list in sequence to form a task initial arrangement set according to the task number order; S12, acquire the component installation sequence graph, read the component number, floor elevation and structure connection direction information of each component from the component installation sequence graph, map the component number to the corresponding construction task number, and build the directed connection relationship between tasks according to the connection direction to obtain a construction priority relationship graph; S13, according to the directed edge direction in the construction priority relationship graph, the task initial arrangement set is sorted by using a topological sorting algorithm to generate a construction task sequence that meets the structure connection logic; S14, count the resource demand type of each task in the construction task item table, arrange and generate a resource demand sequence according to the construction task sequence order, and establish a supply-demand mapping matrix between the task number and the resource number in combination with the resource number in the standard construction resource configuration table provided by the construction unit, and build an initial resource supply relationship graph.

[0023] The embodiment constructs a construction task sequence and an initial resource supply relationship graph, adopts topological sorting in the generation process to ensure the logical sequence of construction tasks, and establishes a supply-demand mapping between tasks and standard resource allocation. This method improves the structural integrity of construction task modeling, accurately expresses the resource flow relationship between different trades, floors, and components, and provides accurate input basis for subsequent construction path simulation and resource scheduling, effectively improving the executability of overall path planning, the matching degree of resource allocation, and the coordination of construction rhythm.

[0024] In the embodiment, S2 specifically includes: S21, collect construction logs and on-site sensing data, obtain the current task completion degree of each component in the construction task sequence, resource call records, and construction environment monitoring data, and generate construction progress state original data; S22, according to the component resource allocation relationship recorded in the initial resource supply relationship graph, combined with the current resource call value of each component in the construction progress state original data, calculate the supply ratio of the component-resource corresponding relationship, construct the current resource supply state matrix, and the matrix takes the component number and the resource number as the index, records the ratio of actual supply and allocation; S23, according to the change sequence of the task completion degree of each component in the construction progress state original data with time, calculate the task completion rate according to the component number, and use a sliding time window to calculate the first derivative to form a task completion rate change curve, and generate a construction rhythm change state according to the curve; S24, analyze the mutation signal in the construction environment monitoring data, combine the keywords and time records about event interference in the construction log, construct an event interference tuple, including interference position, time window and influence type; S25, associate and label the event interference tuple according to the component number, and get the interference intensity score by weighted superposition, generate a set of work interference factors, each factor in the set contains component number, interference type, interference score and corresponding time window information; S26, structure unification of the current resource supply state matrix, construction rhythm change state diagram and work interference factor set, form the current actual construction state.

[0025] The embodiment constructs the current resource supply state, construction rhythm change state and work interference factor set, and comprehensively depicts the dynamic evolution process of the construction site. In terms of resource supply, it accurately reflects the actual deviation of resource allocation in construction; in interference modeling, event interference tuples are constructed and interference intensity scores are generated, effectively capturing the influence of construction abnormalities on path planning. This method significantly improves the fineness and real-time performance of construction state modeling, and provides a high-robustness environment perception basis for subsequent path optimization.

[0026] In this embodiment, S3 specifically includes: S31, obtain the work content, dependent task number, construction priority and required resource type corresponding to each task in the construction task sequence, take each task as a task node, sequentially number to form a task node set; S32, extract the resource type, number and available quantity of each type of construction resource from the current resource supply state, establish each type of resource as a resource node to form a resource node set; S33, according to the required resource type and quantity of each task node, establish a resource connection edge between the task node and the corresponding resource node, and add the resource requirement quantity as an edge attribute to the connection edge; S34, according to the dependent task number marked in each task in the construction task sequence, establish a task connection edge between the task nodes to represent the sequence relationship of the construction tasks; S35, combine the task node set, resource node set, resource connection edge and task connection edge into a heterogeneous graph structure, and add the current resource supply state and construction rhythm change state value information in each node attribute to generate a construction-resource confrontation graph model.

[0027] In this embodiment, the construction task sequence and the current resource supply state are structurally integrated to generate a construction-resource confrontation graph model. The model embeds resource supply state and construction rhythm change state information in the node attribute, and has dynamic expression ability for construction process and resource allocation. Compared with traditional construction flow chart or resource arrangement chart, this heterogeneous graph model can describe task dependency relationship and resource conflict structure at the same time, which significantly improves the recognition ability of path planning algorithm for complex resource constraints and construction priority relationship.

[0028] In this embodiment, S4 specifically includes: S41, based on the construction-resource confrontation graph model, extract the connection relationship between the task nodes in the construction task sequence and the available resource nodes in the resource supply state, take the current actual construction state as the initial root node, and initialize the path search tree structure; S42, according to the construction priority of the task nodes in the construction task sequence, expand the task state corresponding to the current root node, construct an expandable task node set, and generate corresponding child nodes in the path search tree; S43, in the process of generating child nodes, combine the supply quantity of each resource number in the current resource supply state to judge the resource matching condition of each task node, and mark the resource constraint state label for the task nodes that cannot be satisfied; S44, input the generated path search tree structure and the task nodes, resource state labels and time step index information in it into the improved heuristic Monte Carlo search algorithm.

[0029] The embodiment realizes dynamic labeling of resource matching conditions by introducing a resource constraint state label in the process of generating a multi-level path search tree structure. The constructed path search tree structure, task nodes, resource states, and time step indexes are collectively input into an improved heuristic Monte Carlo search algorithm, significantly enhancing the algorithm's expression ability for task scheduling rationality and resource coordination in the initialization stage, laying a high-quality data structure foundation for subsequent construction path simulation and evaluation, and improving the accuracy and feasibility of path optimization results.

[0030] In the embodiment, the S5 specifically includes: S51, in the improved heuristic Monte Carlo search algorithm, a state space simulation model based on the current path search tree structure is constructed, and path branch simulation is performed; in each round of simulation, the construction sequence, resource consumption behavior, and site state response process are simulated to generate a path execution sequence; S52, a multi-strategy hybrid simulation mechanism is introduced in the simulation process, and heuristic scheduling strategies based on experience rules, local optimization strategies based on reinforcement learning, and resource intervention strategies based on structure weights are used to dynamically guide the construction path execution sequence, and its construction feasibility and resource coordination degree are evaluated in parallel between simulation nodes; S53, in the path tree selection and backtracking stage, a construction rhythm adaptive path adjustment mechanism is fused, based on the local rate change of each construction node in the construction rhythm change state, the path continuation direction is dynamically adjusted, and the priority of high-frequency operation area is improved; S54, a construction environment risk dynamic modeling mechanism is synchronously fused, based on the frequency and influence time of each event disturbance tuple in the path branch in the set of operation disturbance factors, a path disturbance score is generated and added to the path backtracking expectation value; S55, according to the execution sequence and simulation results of each path branch, its task delay loss, resource conflict penalty, and environmental risk score are calculated, and after being respectively given a preset weight, weighted summation is performed to obtain an evaluation score; S56, all path branches and their corresponding evaluation scores are recorded.

[0031] The embodiment dynamically corrects the path evaluation value in the path search by integrating a multi-strategy hybrid simulation mechanism and two types of adaptive mechanisms, optimizing the path selection process. The multi-strategy simulation mechanism can simultaneously perform heuristic scheduling, reinforcement learning guidance, and structure weight intervention, improving the comprehensiveness of construction path simulation. The construction rhythm adaptive mechanism adjusts the node priority based on the local task rate change, ensuring the continuity of the path. The environmental risk modeling mechanism generates a path disturbance score through operation disturbance factors, making the path evaluation more adaptive to the site. This method significantly improves the accuracy and robustness of construction path planning, providing a more reliable decision basis for dynamic construction scheduling.

[0032] In this embodiment, S6 specifically includes: S61, sort the plurality of feasible path branches and their corresponding evaluation scores, and select the top m highest scoring paths as a candidate path set; S62, for each path in the candidate path set, sequentially perform path traversal operations to map its node order to the corresponding components and resource units in the construction-resource confrontation graph model to obtain a mapping result; S63, based on the mapping result, calculate the resource preemption conflict rate between components, construction order feasibility and interference event hit probability in each path, and combine the construction rhythm change state and the work interference factor to comprehensively score the overall execution stability and execution efficiency of the path; S64, according to the comprehensive scoring result, select the highest scoring path as the optimal construction path under the current state, and output the node order, resource allocation sequence and key control period corresponding to the path.

[0033] The embodiment dynamically selects the optimal construction path under the current state by combining the construction-resource confrontation graph model, the construction rhythm change state and the work interference factor. This method comprehensively considers the resource conflict rate, construction order rationality and environmental interference probability, realizes the comprehensive evaluation of path stability and execution efficiency, effectively avoids the problem of resource waste or process conflict caused by improper path selection, and improves the overall efficiency and reliability of high-rise building construction.

[0034] In this embodiment, S7 specifically includes: S71, compare the construction node order in the current actual construction state with the node order in the optimal construction path one by one, identify the node position and offset degree that are inconsistent in order; S72, compare the current resource supply state with the resource configuration required by the nodes in the optimal construction path, and calculate the resource difference of each node; S73, compare the current construction rhythm change state with the construction time window in the optimal path, and identify the rhythm conflict interval; S74, generate current construction step adjustment suggestions according to the offset degree, resource difference and rhythm conflict, including node rearrangement order and key task pre-operation instructions; S75, generate resource allocation adjustment instructions according to the resource difference and rhythm conflict, including resource reallocation suggestions and current construction step adjustment suggestions.

[0035] The embodiment compares the optimal construction path with the current actual construction state in terms of node sequence, resource allocation and construction time window, etc., identifies the deviated nodes, resource differences and rhythm conflict sections, and generates resource allocation adjustment instructions accordingly. This method can real-time correct the deviation in the construction execution process, timely adjust the key task sequence and resource allocation scheme, ensure the consistency of the construction path and the plan, effectively improve the stability, continuity and execution efficiency of the construction process, and reduce the risk of resource waste and delay.

[0036] In the embodiment, the S8 specifically includes: S81, the optimal construction path is jointly coded with the construction-resource confrontation graph model state, construction rhythm change state and operation interference factor when the path is generated, to form a path state snapshot data record; S82, a unique index number is allocated to the path state snapshot data record, a state snapshot index set is established, and the optimal construction path, the corresponding evaluation score and the state snapshot index are associated and stored to form a searchable construction path optimization result database; S83, when subsequent construction state update or path re-planning demand occurs, the state snapshot index set is called to perform fast search and reuse operation of the similar path with the current state.

[0037] The embodiment realizes the structured storage and efficient management of the optimal construction path and its evaluation state by generating the path state snapshot and establishing the index set. This method supports fast search and reuse of similar paths in the subsequent construction process, avoids repeated calculation, and shortens the planning response time. Compared with the traditional re-evaluation strategy, the intelligent scheduling efficiency and resource adaptation ability of the system in the complex construction environment are significantly improved, and the real-time and stability of the path planning process are enhanced.

[0038] Embodiment 1: In order to verify the feasibility of the application in implementation, the application is applied to a large urban complex project. The construction unit plans to complete the main structure construction task of a 45-story high-rise office building in 20 months. The building structure is complex, the component quantity is large, and the resource allocation demand is high. The traditional construction path planning method mainly relies on the experience of construction personnel to develop scheduling scheme, and it is difficult to maintain the consistency of construction rhythm and resource utilization efficiency in dynamic construction environment. Therefore, the construction unit decides to introduce the dynamic planning method of high-rise building construction path based on artificial intelligence proposed by the application, to comprehensively improve the construction efficiency and intelligent level.

[0039] The project first inputs the construction task list and the component installation sequence diagram into the system, generates the construction task sequence and the initial resource supply relationship diagram by analyzing the task number, work type, resource type and connection logic relationship. Based on this, the topological dependency relationship between construction processes is constructed, and the resource supply and demand mapping matrix is formed in combination with the standard construction resource allocation table. In the construction process, the displacement monitor, vibration sensor, temperature and humidity sensor and RFID material identification system deployed on site collect real-time construction progress state data and resource call records. At the same time, the project management system synchronously accesses the construction log to record information such as interference events, progress delays and material abnormalities.

[0040] The system automatically generates the current resource supply state in the daily scheduling period, generates the construction rhythm change state based on the component task completion rate change, and extracts the work interference factor set through the event interference identification algorithm to form the current actual construction state. Then, the system constructs the construction-resource confrontation graph model according to the construction task sequence and the current supply state, integrates the task logic relationship, resource distribution and interference weight information, and provides basic graph data support for the initialization of the path search tree structure.

[0041] The path search tree structure constructed based on the construction-resource confrontation graph model is input into the improved heuristic Monte Carlo search algorithm. In the path simulation stage, the heuristic scheduling strategy based on experience rules, the reinforcement learning guided strategy and the resource intervention priority strategy are introduced to form a multi-strategy hybrid simulation mechanism; in the selection and backtracking stage, the construction rhythm adaptive path adjustment mechanism and the construction environment risk dynamic modeling mechanism are integrated to realize the fine correction of path evaluation and the dynamic decision of optimal branch, and finally output multiple feasible path branches and their corresponding comprehensive evaluation scores.

[0042] The system selects the top 10 paths with the highest scores as the candidate path set, performs mapping analysis and stability evaluation, and finally outputs the optimal construction path. The optimal construction path is compared with the current actual construction state to generate construction step adjustment suggestions and resource allocation instructions, forming a structured path optimization result. The following is the comparison data statistics of the traditional method and the method of the present application in multiple key indicators in the actual construction process:

[0043] Table 1 Comparison table of construction path planning performance

[0044] From the above table, it can be seen that the method of the application has achieved significant improvement in multiple core construction scheduling indicators. Especially in resource conflict processing, interference event avoidance and path execution efficiency, it shows a high advantage. Taking the core floor (18th to 21st floor) of a construction unit as an example, in the traditional path scheme, the average waiting time of node resource configuration is 41 minutes, the cumulative conflict times reaches 8 times, and the actual construction period is 4 days longer than expected; while the recommended path of the system under the same conditions, the waiting time is shortened to 19 minutes, the conflict times is reduced to 1 time, and the construction period is completed 2 days in advance.

[0045] More importantly, the construction path optimization result database constructed by the method of the application can be quickly reused in the subsequent construction stage, reducing the calculation time of path re-planning and improving the response speed of path adjustment. When material supply anomaly occurs in the subsequent stage project, the system calls the historical state snapshot data to realize the quick matching and scheduling update of similar state path, saving about 58% of path generation time.

[0046] In summary, the application of the application in actual engineering shows that it can effectively support intelligent path planning under the condition of complex process and dynamic construction state of high-rise building; by introducing multi-strategy simulation, rhythm adjustment and risk modeling mechanism, the accuracy and execution stability of path planning are significantly improved; the system has good real-time, expansibility and reusability, providing a feasible technical path for building informatization and intelligent construction, and has a wide application prospect.

[0047] The above describes only the preferred specific embodiments of the application, but the protection scope of the application is not limited thereto, any person skilled in the art can make equivalent replacement or change according to the technical solution and inventive concept of the application within the technical range disclosed by the application, which should be covered within the protection scope of the application.

Claims

1. A dynamic planning method for construction paths of high-rise buildings based on artificial intelligence, characterized in that, Includes the following steps: S1. Obtain the construction task order and component installation sequence diagram, and generate a diagram showing the relationship between the construction task sequence and the initial resource supply. S2. Obtain the current construction progress status data, combine it with the initial resource supply relationship diagram to form the current resource supply status, generate the construction rhythm change status according to the task completion rate of each construction node, and extract the operation interference factors by combining the on-site sensor data and construction logs to generate the current actual construction status. S3. Based on the construction task sequence and the current resource supply status, construct a construction-resource confrontation graph model; S4. Initialize the path search tree structure based on the construction-resource adversarial graph model and input it into the improved heuristic Monte Carlo search algorithm; S5. Using an improved heuristic Monte Carlo search algorithm, a multi-strategy hybrid simulation mechanism is introduced in the simulation phase to perform construction state evolution and resource response simulation. In the selection and backtracking phase, the construction rhythm adaptive path adjustment mechanism and the construction environment risk dynamic modeling mechanism are integrated to dynamically correct the path evaluation value, complete path sampling and evaluation, and output multiple feasible path branches and their corresponding evaluation scores. S6. Based on multiple feasible path branches and evaluation scores, combined with the construction-resource confrontation diagram model, construction rhythm change status and operation interference factors, perform path branch comparison and dynamic optimization to obtain the optimal construction path under the current state. S7. Compare the optimal construction path with the current actual construction status in terms of node sequence, resource allocation and time window, and output resource allocation adjustment instructions. S8. Record the optimal construction path and evaluation score to form the construction path optimization result.

2. The method for dynamic planning of construction paths for high-rise buildings based on artificial intelligence according to claim 1, characterized in that, S1 specifically includes: S11. Obtain the construction task sheet. Based on the task number, job type, duration and resource requirement type recorded in the construction task sheet, arrange and construct the construction task item table in sequence, and form the initial task arrangement set according to the task number order. S12. Obtain the component installation sequence diagram. Read the component number, floor elevation and structural connection direction information of each component from the component installation sequence diagram. Map the component number to the corresponding construction task number and construct the directed connection relationship between tasks according to the connection direction to obtain the construction priority relationship diagram. S13. Based on the direction of the directed edges in the construction priority relationship diagram, the initial set of tasks is sorted using the topological sorting algorithm to generate a construction task sequence that conforms to the structural connection logic. S14. Statistically analyze the resource requirement types of each task in the construction task item table, arrange them in the order of the construction task sequence to generate a resource requirement sequence, and combine them with the resource numbers in the standard construction resource allocation table provided by the construction unit to establish a supply and demand mapping matrix between task numbers and resource numbers, and construct an initial resource supply relationship diagram.

3. The method for dynamic planning of construction paths for high-rise buildings based on artificial intelligence according to claim 1, characterized in that, S2 specifically includes: S21. Collect construction logs and on-site sensor data, obtain the current task completion status, resource call records and construction environment monitoring data of each component in the construction task sequence, and generate raw data of construction progress status. S22. Based on the component resource allocation relationship recorded in the initial resource supply relationship diagram, and combined with the current resource call value of each component in the original data of construction progress status, calculate the supply ratio of the component-resource correspondence, and construct the current resource supply status matrix. The matrix records the ratio of actual supply to allocation using component number and resource number as indexes. S23. Based on the time sequence of the task completion rate of each component in the original data of construction progress status, calculate the task completion rate according to the component number, and use the sliding time window to calculate its first derivative to form a task completion rate change curve. Based on the curve, group and cluster the components to generate the construction rhythm change status. S24. Analyze the abrupt change signals in the construction environment monitoring data, and combine the keywords and time records of event interference in the construction log to construct event interference tuples, including interference location, time window and impact type; S25. The event interference tuples are associated and labeled according to the component number, and the interference intensity score is obtained by weighted superposition. A set of operation interference factors is generated, and each factor in the set contains the component number, interference type, interference score and corresponding time window information. S26. Unify the current resource supply status matrix, construction rhythm change status diagram and operation interference factor set to form the current actual construction status.

4. The method for dynamic planning of construction paths for high-rise buildings based on artificial intelligence according to claim 1, characterized in that, S3 specifically includes: S31. Obtain the work content, dependent task number, construction priority and required resource type corresponding to each task from the construction task sequence, and number each task as a task node to form a task node set. S32. Extract the resource type, number and available quantity of various construction resources from the current resource supply status, establish each type of resource as a resource node, and form a set of resource nodes; S33. Based on the type and quantity of resources required by each task node, establish resource connection edges between the task node and the corresponding resource node, and attach the resource requirement quantity as an edge attribute to the connection edge. S34. Based on the dependent task number marked for each task in the construction task sequence, establish task connection edges between task nodes to represent the sequential relationship of construction tasks. S35. Combine the task node set, resource node set, resource connection edge and task connection edge into a heterogeneous graph structure, and add numerical information of the current resource supply status and construction rhythm change status to the attributes of each node to generate a construction-resource confrontation graph model.

5. The method for dynamic planning of construction paths for high-rise buildings based on artificial intelligence according to claim 1, characterized in that, S4 specifically includes: S41. Based on the construction-resource adversarial graph model, extract the connection relationship between task nodes in the construction task sequence and available resource nodes in the resource supply status, and initialize the path search tree structure with the current actual construction status as the initial root node. S42. Based on the construction priority of the task nodes in the construction task sequence, expand the task status corresponding to the current root node, construct an expandable task node set, and generate corresponding child nodes in the path search tree. S43. During the generation of child nodes, the resource matching status of each task node is determined by combining the supply of each resource number in the current resource supply status, and the resource constraint status label is marked for the task nodes that cannot be satisfied. S44. Input the generated path search tree structure and its task nodes, resource status labels and time step index information into the improved heuristic Monte Carlo search algorithm.

6. The method for dynamic planning of construction paths for high-rise buildings based on artificial intelligence according to claim 1, characterized in that, S5 specifically includes: S51. In the improved heuristic Monte Carlo search algorithm, a state-space simulation model based on the current path search tree structure is constructed, and path branch simulation is performed; in each round of simulation, the construction sequence of components, resource consumption behavior and on-site state response process are simulated to generate a path execution sequence. S52. In the simulation process, a multi-strategy hybrid simulation mechanism is introduced. The heuristic scheduling strategy based on empirical rules, the local optimization strategy based on reinforcement learning, and the resource intervention strategy based on structural weights are used to dynamically guide the execution sequence of the construction path. The construction feasibility and resource coordination degree are evaluated in parallel among the simulation nodes. S53. In the path tree selection and backtracking stage, an adaptive path adjustment mechanism for construction rhythm is integrated. Based on the local rate changes of each construction node in the state of construction rhythm change, the path continuation direction is dynamically adjusted to improve the priority of high-frequency operation areas. S54. Synchronous integration of construction environment risk dynamic modeling mechanism: Based on the frequency and duration of each event interference tuple in the set of operation interference factors in the path branch, generate path interference score and superimpose it on the path backtracking expectation value. S55. Based on the execution sequence and simulation results of each path branch, calculate its task delay loss, resource conflict penalty and environmental risk score, assign preset weights to each, and then sum them up to obtain the evaluation score. S56. Record all path branches and their corresponding evaluation scores.

7. The method for dynamic planning of construction paths for high-rise buildings based on artificial intelligence according to claim 1, characterized in that, S6 specifically includes: S61. Sort the multiple feasible path branches and their corresponding evaluation scores, and select the top m paths with the highest scores as a candidate path set. S62. For each path in the candidate path set, perform path traversal operation in sequence, and map its nodes sequentially to the corresponding components and resource units in the construction-resource confrontation graph model to obtain the mapping result; S63. Based on the mapping results, calculate the resource contention conflict rate, construction sequence feasibility and interference event hit probability among components in each path, and combine the construction rhythm change status and operation interference factors to comprehensively score the overall execution stability and execution efficiency of the path. S64. Based on the comprehensive scoring results, select the path with the highest score as the optimal construction path in the current state, and output the node sequence, resource allocation sequence and key control period corresponding to the path.

8. The method for dynamic planning of construction paths for high-rise buildings based on artificial intelligence according to claim 1, characterized in that, Specifically, S7 includes: S71. Compare the order of construction nodes in the current actual construction state with the order of nodes in the optimal construction path one by one, and identify the position and degree of offset of nodes with inconsistent order. S72. Compare the current resource supply status with the resource allocation required for nodes in the optimal construction path, and calculate the resource differences for each node. S73. Compare the current construction rhythm changes with the construction time window in the optimal path, and identify the rhythm conflict intervals. S74. Based on three indicators—degree of deviation, resource differences, and rhythm conflicts—generate suggestions for adjusting the current construction steps, including the reordering of nodes and the pre-operation instructions for key tasks. S75. Based on resource differences and rhythm conflicts, generate resource allocation adjustment instructions, including resource reallocation suggestions and current construction step adjustment suggestions.

9. The method for dynamic planning of construction paths for high-rise buildings based on artificial intelligence according to claim 1, characterized in that, S8 specifically includes: S81. Encode the optimal construction path together with the construction-resource confrontation graph model state when the path was generated, the construction rhythm change state, and the operation interference factor to form a path state snapshot data record. S82. Assign a unique index number to the path status snapshot data record, establish a status snapshot index set, and associate and store the optimal construction path, the corresponding evaluation score and the status snapshot index to form a searchable construction path optimization result database. S83. When subsequent construction status updates or path replanning requirements occur, call the status snapshot index set to perform a fast search and reuse operation for paths similar to the current status.

Images