Intelligent construction management method for intelligent beam field

Abstract

The invention relates to an intelligent construction management method for an intelligent beam field, and the method comprises the steps: collecting multi-source data of a field beam field control system in a preset target period, and carrying out the preprocessing of the multi-source data into standardized input data; substituting the standardized input data into a productivity prediction model trained according to historical multi-source data, and generating a corresponding productivity interval; on the basis of a constrained optimization algorithm, generating a corresponding production scheduling result according to a pre-acquired production scheduling plan and a production capacity interval, and driving a field beam field control system to execute corresponding scheduling operation according to the production scheduling result; and when interference factors are detected, a new production scheduling result is dynamically adjusted based on the productivity prediction model, so that beam manufacturing, beam storage and beam discharging form a closed-loop collaborative management chain, thereby improving the precision and flexibility of a production plan, reducing resource waste and construction conflicts, and improving the overall operation efficiency and scheduling response capability.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent beam yard construction management, and in particular to an intelligent beam yard intelligent construction management method. BACKGROUND

[0002] At present, the intelligent beam yard management system has realized the informatization management of the prefabricated beam production process, covering task planning, progress display, process execution and other links, but there are still obvious deficiencies in resource scheduling and capacity prediction.

[0003] On the one hand, the existing system mostly uses static parameters or manual rules for capacity evaluation, which cannot reflect the dynamic changes of site pedestal occupation, personnel scheduling, raw material supply and the like in real time, resulting in inaccurate capacity prediction and difficulty in effectively guiding production scheduling strategy, on the other hand, the existing production scheduling system processes beam manufacturing, beam storage and beam delivery as independent modules, without establishing a unified scheduling optimization mechanism, which is prone to problems such as uneven utilization of pedestals, inability to timely store beams or limited transportation, affecting the overall construction efficiency. SUMMARY

[0004] In order to solve the problems of inaccurate capacity evaluation and disconnection of production scheduling in the existing beam yard system, the present application provides an intelligent beam yard intelligent construction management method.

[0005] An intelligent beam yard intelligent construction management method, the intelligent beam yard intelligent construction management method comprises: Collecting multi-source data of a site beam yard control system within a preset target period, the multi-source data comprising resource state data and external condition data, and preprocessing the multi-source data into standardized input data; Substituting the standardized input data into a capacity prediction model trained according to historical multi-source data to generate a corresponding production capacity interval; Based on a constraint optimization algorithm, generating a corresponding production scheduling result according to a pre-acquired production scheduling plan and the production capacity interval, the production scheduling result being used for collaborative optimization of one or more scheduling objects participating in resource coordination task arrangement; According to the production scheduling result, driving the site beam yard control system to perform corresponding scheduling operations; Real-time monitoring of interference factors occurring in the scheduling operations, when the interference factors are detected, dynamically adjusting a new production scheduling result based on the capacity prediction model, and real-time pushing the update to the site beam yard control system.

[0006] By adopting the technical scheme, the multi-source data of the field control system is collected and standardized, the production capacity interval is generated by combining the trained production capacity prediction model, the feasible production scheduling result is generated by the constraint optimization algorithm, the scheduling scheme can be automatically adjusted when the interference factor occurs, the closed-loop intelligent control of the beam field construction management from data collection, capacity evaluation, production scheduling execution to abnormal response is realized, and the resource coordination efficiency and the field scheduling flexibility are significantly improved.

[0007] Preferably, the field beam field control system collects multi-source data in a preset target period, and the multi-source data includes resource state data and external condition data. In the step of preprocessing the multi-source data into standardized input data, the following steps are included: The resource state data of the field beam field control system in the preset target period is collected, and the resource state data includes historical beam production record data, pedestal occupation data, field personnel scheduling data and raw material inventory data. The external condition data of the field beam field control system in the preset target period is collected, and the external condition data includes weather forecast data and transportation scheduling plan data. The resource state data and external condition data are classified and managed respectively, and are structured and packaged according to the classification results to generate corresponding standardized input data.

[0008] By adopting the technical scheme, the resource state data and external condition data collected are classified and managed and structured and packaged, so that the original multi-source information is converted into standardized input data in a unified format, which lays a high-quality data foundation for accurate calculation of the subsequent production capacity prediction model, thereby improving the accuracy of the prediction result and the response ability of the system as a whole.

[0009] Preferably, in the step of substituting the standardized input data into the production capacity prediction model to generate the production capacity interval in the corresponding preset target period, the following steps are included: The corresponding identification field is extracted from the standardized input data. According to the identification field, the corresponding function sub-model is matched in the production capacity prediction model. Based on the function sub-model, the prediction result corresponding to the standardized input data is generated. According to each prediction result, the production capacity interval in the corresponding preset target period is generated.

[0010] By adopting the technical scheme, the identification field is extracted from the standardized input data and matched with the corresponding function sub-model in the capacity prediction model, ensuring that the input data is accurately connected with the model structure, effectively improving the pertinence and efficiency of the prediction, and enabling the model to adapt to prediction tasks under multiple data dimensions, thereby enhancing the flexibility and adaptability of the prediction system.

[0011] Preferably, in the step of generating a production capacity interval corresponding to a preset target period according to each prediction result, the step comprises: structurally processing each prediction result to extract a capacity value set in multiple time periods; determining the maximum value in each capacity value set as an upper limit parameter, and determining the minimum value in the capacity value in each time period as a lower limit parameter; generating a corresponding production capacity interval according to the upper limit parameter and the lower limit parameter.

[0012] By adopting the technical scheme, the maximum and minimum values of the capacity in each time period are extracted by structurally processing multiple prediction results, and a production capacity interval with boundary constraints is constructed, thereby providing a clear and quantitative capacity boundary basis for subsequent production scheduling, and ensuring that the scheduling strategy does not exceed the actual carrying capacity.

[0013] Preferably, in the step of structurally processing each prediction result to extract a capacity value set in multiple time periods, the step comprises: dividing the preset target period into multiple time periods; structurally processing each prediction result to extract a capacity value with a timestamp identifier; generating a capacity value set in multiple time periods based on each capacity value according to the inclusion relationship between the timestamp identifier and the multiple time periods.

[0014] By adopting the technical scheme, the target period is divided into multiple time periods, and the prediction result is mapped to the capacity set in the corresponding time period according to the timestamp identifier, so that the capacity state of different time slices can be finely presented, thereby improving the timeliness and granularity of the capacity boundary construction, and helping to improve the time sequence adaptability of the production scheduling scheme.

[0015] Preferably, in the step of generating a corresponding production scheduling result based on a constraint optimization algorithm according to a pre-acquired production scheduling plan and the production capacity interval, the production scheduling result is used for collaborative optimization of one or more scheduling objects participating in resource coordination task arrangement, the step comprises: calling a pre-determined production scheduling plan, the production scheduling plan at least comprising an optimization target and a constraint condition; According to the optimization target, at least one objective function is set; According to the constraint condition and the objective function, a constraint optimization algorithm is initialized and configured; The production capacity interval is substituted into the configured constraint optimization algorithm to generate a corresponding production scheduling result.

[0016] By adopting the above technical solution, the initialized configuration of the optimization algorithm is realized by calling the set scheduling plan and constructing the objective function and the constraint condition, and the production scheduling result is solved in combination with the actual capacity boundary condition, so that the production scheduling strategy has the executability and constraint adaptability, and the scientificity and on-site adaptation ability of the overall scheduling result are improved.

[0017] Preferably, in the step of initializing and configuring the constraint optimization algorithm according to the constraint condition and the objective function, the step includes: Defining a variable set for describing the production scheduling state of the scheduling object, the variable set including a task arrangement variable and a resource allocation variable associated with a time period; According to the constraint condition, a value range limiting parameter corresponding to the variable set is set, the value range limiting parameter including a boundary parameter and a resource occupation rule parameter; The objective function is converted into a corresponding function expression; The variable set, the value range limiting parameter and the objective function expression are input into the constraint optimization algorithm to complete the initialization and configuration of the constraint optimization algorithm.

[0018] By adopting the above technical solution, the scheduling variable set associated with the time period is constructed, the variable value range is set in combination with the resource boundary and the occupation rule, the abstract objective function is converted into a solvable expression and input into the optimization algorithm, so that the scheduling task is efficiently converged to the optimal result within the constraint condition, and the stability and precision of the scheduling strategy are enhanced.

[0019] Preferably, in the step of substituting the production capacity interval into the configured constraint optimization algorithm to generate a corresponding production scheduling result, the step specifically includes: The production capacity interval is taken as a value boundary condition in the constraint optimization algorithm to limit the maximum and minimum values of the production scheduling variable in each time period; In the process of executing the constraint optimization algorithm, the variable is solved according to the objective function within the range of the production capacity interval, and a production scheduling result satisfying the constraint condition is output.

[0020] By adopting the technical scheme, the production capacity interval is taken as the boundary limit of the production scheduling variable, the maximum and minimum values of task arrangement in different time periods are controlled, and in the execution process of the optimization algorithm, it is ensured that the generated production scheduling result meets the production capacity constraint condition, and the synchronization balance of production scheduling feasibility and system load is realized.

[0021] Preferably, in the step of dynamically adjusting a new production scheduling result based on the production capacity prediction model when the interference factor is detected, the step comprises: When the interference factor is detected, it is identified whether the interference factor is determined as an effective abnormal event. If not, the corresponding scheduling operation is continued to be performed; If yes, new multi-source data in a preset target period at the current time is re-acquired, and a new production scheduling result is generated according to the new multi-source data.

[0022] By adopting the technical scheme, the abnormal event identification and judgment of the interference factor are performed, if it is an effective abnormal event, the multi-source data in the current state is re-acquired and a new production scheduling scheme is dynamically generated, so that the system has the self-adaptive ability to sudden changes, thereby ensuring that the construction task can still be steadily promoted and the efficient response ability is maintained in the dynamic environment.

[0023] In summary, the present application includes at least one of the following beneficial technical effects: The present application realizes dynamic perception and intelligent decision of the beam field resource state and external construction conditions by introducing a production capacity prediction model based on machine learning combined with a constraint optimization algorithm. Specifically, it takes multi-source data as input, including historical production records, pedestal usage, personnel allocation, raw material inventory, weather and transportation plan and other key factors, constructs a production capacity prediction mechanism that can be updated in real time with the changes in the field, and can accurately reflect the available production capacity of the beam field in a certain period in the future; on this basis, the prediction result is taken as the boundary input of the scheduling algorithm, combined with the pre-set construction target and resource constraint, and an optimization model is used to generate a reasonable production scheduling scheme, ensuring that the task arrangement is within the production capacity bearing range, and can maximize the matching of the current resource condition and plan demand. It effectively breaks through the problem of relying on static configuration and lacking of linkage scheduling in traditional beam field management, so that beam manufacturing, beam storage and beam delivery form a closed-loop collaborative management chain, thereby improving the precision and flexibility of production planning, reducing resource waste and construction conflicts, and improving the overall operation efficiency and scheduling response ability. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 is a flowchart of a smart beam field intelligent construction management method in an embodiment of the present application. DETAILED DESCRIPTION

[0025] The application will be further described in detail below with reference to the accompanying drawings.

[0026] In an embodiment, as shown in Figure 1 The application discloses a smart beam field intelligent construction management method. The smart beam field refers to a prefabricated beam production station system with data perception, model prediction and intelligent decision-making capabilities. Its core capability lies in realizing automatic, optimized and responsive management of the whole beam production process through the fusion of multi-source data and algorithm support. The smart beam field intelligent construction management method comprises the following steps: S10, collecting multi-source data of a field beam field control system within a preset target period, the multi-source data including resource state data and external condition data, and preprocessing the multi-source data into standardized input data; the field beam field control system is a control and execution system deployed in the physical field of the beam field, including components such as pedestals, templates, equipment, sensors and controllers, and has the capabilities of data collection, state feedback and execution scheduling. The preset target period refers to the time range set by the user for prediction and production scheduling, usually with a granularity of days, weeks or continuous working shifts, and is a unified time period for coordinating data collection and scheduling strategies. The multi-source data refers to structured or unstructured information collected from different data sources, which mainly includes resource state data and external condition data in this method. The resource state data refers to the running and configuration state of the available production resources in the current beam field, covering historical beam production records, pedestal occupation, personnel scheduling information of each work team and raw material inventory level. The external condition data refers to external environmental factors that are not controlled by the beam field system but have a significant impact on its production rhythm and capacity, such as weather changes, transportation arrangements, etc. The standardized input data is a unified input format obtained by classifying, formatting and structuring the original multi-source data, which is convenient for subsequent model processing.

[0027] S20, substituting the standardized input data into a production capacity prediction model trained based on historical multi-source data to generate a corresponding production capacity interval; the production capacity prediction model is a calculation model for predicting the change range of production capacity in a future period based on historical multi-source data and constructed through machine learning methods, which can output 0 production capacity values for each time period. The production capacity interval refers to the upper and lower limit values of the production capacity corresponding to the time period calculated based on the model prediction results within the preset target period, which is used to constrain the boundary conditions of the production scheduling scheme.

[0028] S30, generating a corresponding production scheduling result based on a constraint optimization algorithm according to the pre-acquired production scheduling plan and the production capacity interval, the production scheduling result being used for collaborative optimization of one or more scheduling objects participating in resource coordination task arrangement; the constraint optimization algorithm is an algorithm tool for seeking an optimal production scheduling solution under a series of resource and capacity constraints, and the purpose is to realize task collaboration and reasonable resource allocation among actual scheduling objects. The production scheduling plan is a construction task arrangement scheme generated by manual or system in advance, which determines the production scheduling target, execution sequence and resource utilization strategy. The production scheduling result is the final production scheduling scheme generated by executing the optimization algorithm under the constraint of the production capacity interval, and specifically includes the matching of tasks and resources, the allocation of benches and personnel, etc. The scheduling object refers to a unit body that needs to be configured with tasks or allocated with resources in actual production scheduling, which can be a bench, equipment, etc.

[0029] S40, driving the on-site beam field control system to execute corresponding scheduling operations according to the production scheduling result; S50, real-time monitoring of interference factors occurring in the scheduling operation, when the interference factors are detected, generating a new production scheduling result based on the capacity prediction model and real-time pushing the update to the on-site beam field control system; the interference factors refer to events deviating from the original plan occurring in the scheduling operation process, such as equipment failure, personnel absence, material delay, etc., which will affect the current execution progress. Dynamic adjustment refers to the process of regenerating a scheduling scheme based on the current actual situation when the system detects effective interference factors, to ensure that the system scheduling remains real-time and effective.

[0030] Further, collecting multi-source data of the on-site beam field control system within a preset target period, the multi-source data including resource state data and external condition data, in the step of pre-processing the multi-source data into standardized input data, comprising: Collecting resource state data of the on-site beam field control system within a preset target period, the resource state data including historical beam production record data, bench occupation data, on-site personnel scheduling data and raw material inventory data; the historical beam production record data describes the actual production quantity, time distribution and bench usage of various beams in the past period of the beam field, which helps the model to capture the production rhythm and bench usage efficiency; the bench occupation data refers to the tasks already arranged, idle period and space capacity state of each beam bench within the target period, which is a key indicator for measuring the current production scheduling carrying capacity; the on-site personnel scheduling data refers to the shift distribution, job structure and scheduling strategy of the beam field operators within the target period, reflecting the availability of personnel and labor density; the raw material inventory data describes the current inventory quantity, available time and replenishment plan of key materials such as reinforcing steel, concrete and prestressed materials required by the beam, which is used to evaluate the supply guarantee capacity at the material level.

[0031] The field acquisition beam yard control system collects external condition data within a preset target period, including weather forecast data and transportation scheduling plan data. The external condition data reflects the dynamic factors of the beam yard operation affected by the external environment, including weather forecast data and transportation scheduling plan data. The former is used to reflect the possible interference of weather changes on construction operations, such as the impact of rainfall on the pouring process. The latter describes the transportation vehicle arrangement and frequency plan required for the beam to enter and exit the yard, which has a direct impact on the beam storage capacity and beam release frequency.

[0032] The resource status data and external condition data are classified and managed respectively, and structured packaging is performed according to the classification results to generate corresponding standardized input data. Classification management refers to logically dividing and labeling the collected resource status data and external condition data according to their data types, usage scenarios, or corresponding model input structures, so as to facilitate subsequent structured packaging. Structured packaging refers to organizing data content according to the required format and fields of the model, including field reorganization, missing value completion, timestamp alignment, etc. Finally, standardized input data with unified semantic expression and direct input into the model for calculation are generated.

[0033] Further, in the step of substituting the standardized input data into the production capacity prediction model to generate the production capacity interval within the corresponding preset target period, the following steps are included: Extracting the corresponding identification field from the standardized input data; Matching the corresponding function sub-model in the production capacity prediction model according to the identification field; the identification field refers to a feature field embedded in the standardized input data, which is used to indicate the type, source or usage intention of the data. It is usually used to guide the subsequent model calling logic, such as distinguishing whether it is used to predict the production capacity of concrete beams or steel members, or whether it is used for a certain time window data segment.

[0034] Generating a prediction result corresponding to the standardized input data based on the function sub-model; the function sub-model is a plurality of sub-model units divided in the production capacity prediction model according to different data structures, prediction objects or process flows. Each sub-model is parameterized in the training stage according to its task target and specific data dimension, and automatically matches and calls the corresponding sub-model in the inference stage to output the prediction result. The prediction result is the production capacity value or production capacity trend information obtained by the function sub-model based on the received standardized input data through the forward inference process. It is usually expressed as the production capacity estimate value in each time period within the target period. These results can provide data basis for generating the production capacity interval.

[0035] Generating a production capacity interval within a corresponding preset target period according to each prediction result.

[0036] Further, in the step of generating a production capacity interval corresponding to the preset target period according to each prediction result, the step comprises: The prediction results are structured and processed to extract a production capacity value set in each time period. The time period refers to a continuous time interval artificially divided in the entire prediction target period for the purpose of granular processing of production capacity data. The time period usually has a fixed time length, such as one hour, four hours, or one day, and is used to collect and analyze prediction production capacity data in different time slices. The production capacity value set refers to a set composed of production capacity values corresponding to multiple prediction results in the same time period, which is used to reflect the production capacity fluctuation range that may occur in the time period.

[0037] The maximum value in each production capacity value set is determined as an upper limit parameter, and the minimum value in each time period production capacity value is determined as a lower limit parameter. The upper limit parameter and the lower limit parameter respectively refer to the maximum value and the minimum value taken in the set, which are used to define the boundary range of production capacity in the time period. The production capacity interval is the numerical range formed by the upper limit parameter and the lower limit parameter corresponding to each time period, which represents the production capacity fluctuation range that can be tolerated when scheduling resources in the time period, and is input as a constraint condition into the subsequent scheduling algorithm.

[0038] The production capacity interval corresponding to the upper limit parameter and the lower limit parameter is generated.

[0039] Specifically, if a certain beam field system performs 5 rounds of production capacity prediction for the future 72 hours using different models, and each 8 hours is divided as a time period, then the 5 production capacity prediction values collected in the first time period (i.e. 1-8 hours) are 12, 14, 15, 13, and 16. The production capacity value set in this time period is {12, 13, 14, 15, 16}, the maximum value 16 is the upper limit parameter, the minimum value 12 is the lower limit parameter, and the generated production capacity interval is [12, 16]. This interval will be used as a feasible boundary for constraint scheduling to participate in subsequent task and resource matching optimization.

[0040] Further, in the step of structuring and processing each prediction result to extract a production capacity value set in each time period, the step comprises: The preset target period is divided into multiple time periods. The prediction results are structured and processed to extract production capacity values with timestamp identifiers. The timestamp identifier is a time mark attached to each production capacity prediction data, which records the specific time point or time period corresponding to the data, and is used for subsequent alignment and grouping processing of prediction results in the time dimension.

[0041] Based on the time stamp and the inclusion relationship of the plurality of time periods, the production capacity value set under the plurality of time periods is generated; the plurality of time periods refers to dividing the entire preset target period into a plurality of continuous and non-overlapping subintervals, so as to refine the production capacity analysis and scheduling planning in each time slice, and the time periods can be scheduling windows with hours or custom lengths. The inclusion relationship refers to the mapping relationship between the divided time periods and each time stamp, that is, judging whether the time stamp of a piece of prediction data falls within the start and end intervals of a certain time period, so as to classify the data into the corresponding production capacity value set of the time period.

[0042] Specifically, under the condition that the set preset target period is 48 hours in the future, the system divides it into 6 time periods, each of which is 8 hours, and each production capacity prediction value is accompanied by a time stamp, such as "2025 June 12 08:00:00", which has an inclusion relationship with the first time period "June 12 00:00-08:00", so the prediction value will be classified into the production capacity value set of the first time period, and participate in the production capacity upper and lower limit calculation and capacity interval construction of the time period.

[0043] Further, based on the constraint optimization algorithm, the corresponding production scheduling result is generated according to the pre-acquired production scheduling plan and production capacity interval, and the production scheduling result is used in the step of collaborative optimization of one or more scheduling objects participating in the resource coordination task arrangement, including: The pre-determined production scheduling plan is called, and the production scheduling plan at least includes an optimization target and a constraint condition; the production scheduling plan refers to the operation arrangement scheme in the beam field construction process, which covers the time arrangement of the construction task, the resource allocation strategy, the phased target and the corresponding execution logic, and the plan is usually generated by the project management system based on the engineering progress, the construction period constraint and the resource supply situation. The optimization target is the key performance indicator expected to be achieved in the scheduling plan, such as maximizing the production capacity utilization rate, minimizing the pedestal vacancy rate or minimizing the scheduling conflict, which is the core driving criterion for the execution of the scheduling algorithm. The constraint condition refers to the rule limit that needs to be strictly followed in the production scheduling process, for example, the maximum number of operations per day for each pedestal, the continuous working time of a specific team should not exceed the specified upper limit, the same type of beam should not be repeatedly arranged in the resource conflict area, etc. These conditions determine the feasibility boundary of the scheduling result.

[0044] According to the optimization target, at least one objective function is set; the objective function is a calculation model that expresses the optimization target in mathematical form, which is used to quantify the advantages and disadvantages of the scheduling scheme, and its constituent form usually includes weight items, variable combination items and penalty items, and it is one of the inputs for algorithm solving.

[0045] According to the constraint condition and the objective function, the constraint optimization algorithm is initialized and configured; the constraint optimization algorithm can be constructed by using an integer linear programming method, first, a corresponding objective function is defined based on the optimization objective in the production scheduling plan, for example, a linear expression is constructed in the optimization direction of maximizing the pedestal utilization rate or minimizing the production delay; then, the time period scheduling boundary value formed by the production capacity interval is combined as the upper and lower limit constraints of the variable, and the actual constraint conditions (such as resource capacity limitation, pedestal conflict rules, personnel shift work requirements, etc.) that need to be met in the construction process are integrated and uniformly converted into a set of inequality constraint conditions; finally, the integer linear model constructed is iterated and calculated by calling a solver (such as CPLEX, Gurobi, etc.), and the scheduling and dispatching result that meets all the constraint conditions and makes the objective function take the optimal value is obtained, so as to realize the reasonable arrangement of the scheduling object in the production capacity interval, and guarantee the efficiency and feasibility of the construction execution process.

[0046] The production capacity interval is substituted into the configured constraint optimization algorithm to generate a corresponding production scheduling result.

[0047] Specifically, in the beam yard construction process, the production scheduling plan requires that 6 precast beams are poured every 48 hours, and specifies that the maximum utilization rate of the pedestal is the optimization objective, and at the same time, it is stipulated that the continuous work of each shift personnel does not exceed 10 hours as a constraint condition. At this time, the "maximum utilization rate" is constructed as an objective function, and the upper and lower limit parameters of the production capacity interval and the personnel scheduling restriction are input into the constraint optimization algorithm, and the output production scheduling result is the most reasonable job allocation scheme under the premise of meeting the resource boundary constraints and time rules.

[0048] Further, in the step of initializing and configuring the constraint optimization algorithm according to the constraint condition and the objective function, the step includes: A variable set for describing the scheduling object production state is defined, and the variable set includes a task arrangement variable and a resource allocation variable associated with a time period; the variable set According to the constraint condition, set the value range limitation parameter corresponding to the variable set, the value range limitation parameter includes boundary parameter and resource occupation rule parameter; refers to a set of formalized variables set when constructing the production scheduling model, which is used to express the task arrangement and resource allocation of the scheduling object in different time periods. These variables constitute the solution space of the optimization model, wherein the task arrangement variable is used to describe whether the scheduling object undertakes a specific task in a certain time period, and the resource allocation variable is used to describe whether the scheduling object occupies a specific amount or type of resource in a certain task; the value range limitation parameter corresponding to the variable set refers to the numerical range, boundary or logical limit given to each variable of the above variable set in order to ensure the feasibility of the scheduling result, wherein the boundary parameter defines the upper and lower limit values of the variable (for example, a maximum of 2 tasks is arranged in a certain time period), and the resource occupation rule parameter is used to express the logical relationship between multiple resources, such as exclusivity, capacity upper limit or concurrent configuration (for example, a template resource cannot be occupied by multiple beams at the same time); The objective function is converted into a corresponding function expression; the function expression refers to formalizing the optimization target (such as maximizing production capacity, minimizing delay, balancing resource load, etc.) in the production scheduling plan into a mathematical expression as a solution basis in the optimization algorithm, which usually includes linear weighted items or penalty item combinations The variable set, the value range limitation parameter and the objective function expression are input into the constraint optimization algorithm to complete the initialization configuration of the constraint optimization algorithm. The process of inputting the variable set, the value range limitation parameter and the objective function expression into the constraint optimization algorithm is to load the above structured modeling information as input conditions into the solving engine, so that the algorithm can output the optimal or near-optimal solution under the premise of meeting all the limitation conditions when solving the production scheduling scheme. Taking bench production scheduling as an example, the task arrangement variable can represent "whether bench 1 is used for beam A on the 3rd day", the resource allocation variable can represent "whether template B is allocated to beam A for use", the boundary parameter limits a maximum of 3 beams per day, the resource occupation rule limits that one template is used for one beam per day, and the objective function expression is defined as "maximum total output". By modeling and inputting these elements into the optimization algorithm, the systematic arrangement of the construction plan can be realized.

[0049] Further, in the step of substituting the production capacity interval into the configured constraint optimization algorithm to generate the corresponding production scheduling result, specifically: The production capacity interval is taken as the value boundary condition in the constraint optimization algorithm to limit the maximum and minimum values of the production scheduling variables in each time period; the value boundary condition refers to taking the production capacity interval as the legal upper and lower limits of the production scheduling variables in each time period. This boundary condition not only limits the solution space of the production scheduling variables, but also serves as a basic feasibility constraint in the algorithm solving process. The production scheduling variable refers to the specific production capacity value arranged in each time period in the scheduling scheme in the optimization model, which must be taken within the production capacity interval of the corresponding time period to ensure the executability of the production scheduling scheme.

[0050] In the process of executing the constraint optimization algorithm, the variable is solved according to the target function within the range of the production capacity interval, and the production scheduling result meeting the constraint condition is output. The production scheduling variable refers to the specific production capacity value arranged in each time period in the scheduling scheme in the optimization model, which must be taken within the production capacity interval of the corresponding time period to ensure the executability of the production scheduling scheme.

[0051] Further, when the interference factor is detected, the step of dynamically adjusting a new production scheduling result based on the production capacity prediction model includes: When the interference factor is detected, it is identified whether the interference factor is determined as an effective abnormal event; the abnormal event refers to the interference factor that is identified by the system preset judgment mechanism and has a significant impact on the current production scheduling plan. The system will identify whether the production scheduling strategy needs to be adjusted based on the set sensitivity, influence range or duration, etc. judgment standard.

[0052] If not, continue to perform the corresponding scheduling operation; If yes, new multi-source data in the preset target period at the current time is reacquired, and a new production scheduling result is generated according to the new multi-source data.

[0053] Specifically, in a certain beam field, two pedestals are originally planned to produce T-beams in parallel from 8:00 to 12:00 on a certain day, but at 8:30, the concrete tank truck is delayed, the system detects the interference factor, determines it as an effective abnormal event, immediately collects the new data of raw material inventory, transportation plan, weather condition, etc. at this time, generates the adjusted production capacity interval based on the existing production capacity prediction model, and calls the constraint optimization algorithm to generate a new production scheduling result, delays or transfers part of the task to the idle pedestal, and ensures the sustainable progress of the construction plan.

[0054] The above-described embodiments are only used to illustrate the technical solutions of the present application, but not limit them; although the present application is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1.A method for intelligent construction management of a smart beam yard, characterized in that, The intelligent beam field intelligent construction management method comprises: Collecting multi-source data of a field beam field control system within a preset target period, the multi-source data comprising resource state data and external condition data, preprocessing the multi-source data into standardized input data; Substitute the standardized input data into the capacity prediction model trained according to historical multi-source data to generate a corresponding production capacity interval; Based on the constraint optimization algorithm, according to the pre-acquired production scheduling plan and the production capacity interval, a corresponding production scheduling result is generated, which is used to cooperatively optimize the scheduling object of one or more actual participating resource coordination task arrangements; According to the production scheduling result, drive the field beam field control system to execute the corresponding scheduling operation; Real-time monitoring of interference factors occurring in the scheduling operation, when the interference factors are detected, a new production scheduling result is dynamically adjusted based on the capacity prediction model, and is real-time pushed to the field beam field control system. 2.The intelligent construction management method of the intelligent beam field according to claim 1, characterized in that, In the step of collecting multi-source data of the field beam field control system within a preset target period, the multi-source data comprising resource state data and external condition data, preprocessing the multi-source data into standardized input data, comprising: Collecting resource state data of the field beam field control system within a preset target period, the resource state data comprising historical beam production record data, pedestal occupation data, field personnel scheduling data and raw material inventory data; Collecting external condition data of the field beam field control system within a preset target period, the external condition data comprising weather forecast data and transportation scheduling plan data; Classify and manage the resource state data and external condition data respectively, and structure package according to the classification result to generate corresponding standardized input data. 3.The intelligent construction management method of the intelligent beam field according to claim 1, characterized in that, In the step of substituting the standardized input data into the capacity prediction model to generate a production capacity interval within a corresponding preset target period, comprising: Extract the corresponding identification field from the standardized input data; According to the identification field, match the corresponding function sub-model in the capacity prediction model; Based on the function sub-model, generate a prediction result corresponding to the standardized input data; Generate a production capacity interval within a corresponding preset target period according to each prediction result. 4.The intelligent construction management method of the intelligent beam field according to claim 3, characterized in that, In the step of generating a production capacity interval within a corresponding preset target period according to each prediction result, comprising: Structurally process each prediction result to extract a set of capacity values in multiple time periods; Determine the maximum value in each set of capacity values as the upper limit parameter, and determine the minimum value in each time period capacity value as the lower limit parameter; According to the upper limit parameter and the lower limit parameter, generate a corresponding production capacity interval. 5.The intelligent construction management method of the intelligent beam field according to claim 4, characterized in that, In the step of structurally processing each prediction result to extract a set of capacity values in multiple time periods, comprising: Divide the preset target period into multiple time periods; Structurally process each prediction result to extract a capacity value with a timestamp identifier; According to the time stamp and the inclusion relationship of the plurality of time periods, a plurality of sets of capacity values in the plurality of time periods are generated. 6.The intelligent construction management method of the intelligent beam field according to claim 1, characterized in that, The constraint optimization algorithm is based on the pre-acquired production scheduling plan and the production capacity interval to generate a corresponding production scheduling result, which is used in the step of collaborative optimization of one or more scheduling objects participating in the coordination task arrangement of the actual resource. The production scheduling plan is determined in advance, and the production scheduling plan at least includes an optimization target and a constraint condition; According to the optimization target, at least one objective function is set; According to the constraint condition and the objective function, the constraint optimization algorithm is initialized and configured; The production capacity interval is substituted into the configured constraint optimization algorithm to generate a corresponding production scheduling result. 7.The intelligent construction management method of the intelligent beam field according to claim 6, characterized in that, The step of initializing and configuring the constraint optimization algorithm according to the constraint condition and the objective function includes: Define a variable set for describing the scheduling state of the scheduling object, which includes a task arrangement variable and a resource allocation variable associated with a time period; According to the constraint condition, set the value range limiting parameter corresponding to the variable set, which includes boundary parameter and resource occupation rule parameter; The objective function is converted into a corresponding function expression; The variable set, the value range limiting parameter and the objective function expression are input into the constraint optimization algorithm to complete the initialization and configuration of the constraint optimization algorithm. 8.The intelligent construction management method of the intelligent beam field according to claim 6, characterized in that, The step of substituting the production capacity interval into the configured constraint optimization algorithm to generate a corresponding production scheduling result includes: The production capacity interval is used as the value boundary condition in the constraint optimization algorithm to limit the maximum and minimum values of the scheduling variables in each time period; In the process of executing the constraint optimization algorithm, the variable is solved according to the objective function in the range of the production capacity interval, and the production scheduling result satisfying the constraint condition is output. 9.The intelligent construction management method of the intelligent beam field according to claim 1, characterized in that, The step of dynamically adjusting a new production scheduling result based on the capacity prediction model when the interference factor is detected includes: When the interference factor is detected, it is identified whether the interference factor is determined as an effective abnormal event; If not, continue to execute the corresponding scheduling operation; If yes, new multi-source data in a preset target period at the current time is reacquired, and a new production scheduling result is generated according to the new multi-source data.

Images