Construction site precast beam steel bar production management system

Abstract

The invention discloses a construction site precast beam reinforcement production management system, and relates to the technical field of steel production management, a prediction module predicts an assembly rhythm state through a prediction model, and a model establishment module takes a real-time reinforcement processing plan and an assembly rhythm state prediction result as adaptive floating input variables; steel bar machining is used as a dynamic controlled process, a floating influence model is established, and the floating influence model comprises the real-time adjustment amount of a steel bar machining plan, assembly rhythm state change and dynamic adjustment of the steel bar machining process when the steel bar machining plan and the assembly rhythm state change at the same time. The management system establishes the floating influence model under the conditions of the real-time adjustment amount of the reinforcing steel bar machining plan, the change of the assembly rhythm and the simultaneous change of the real-time adjustment amount and the assembly rhythm, quantifies the specific influence of floating on the reinforcing steel bar machining process, dynamically adjusts the machining process, and improves the machining efficiency. And the self-adaptive capacity of the reinforcing steel bar machining process to the actual demand change on site is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of steel production management, and particularly relates to a steel bar production management system for precast beams at a construction site. Background Art

[0002] In modern infrastructure construction projects, especially during the construction of bridges, viaducts, railways, etc., precast beams, as important structural components, play a key role in transmitting loads and maintaining structural stability. In order to ensure the quality and construction progress of precast beams, the standardized and scientific management of the steel bar processing and production link is particularly important.

[0003] The existing technologies have the following defects:

[0004] The existing management methods regard the steel bar processing process as an independent static link, and fail to establish a dynamic adjustment model based on the two-way floating changes of the steel bar processing plan and the on-site construction rhythm. It is impossible to realize the real-time dynamic regulation of the steel bar processing scheduling and dispatching process under the real-time adjustment amount of the steel bar processing plan, the change of the construction rhythm, and the combined change of the two, which severely restricts the responsiveness and adaptability of the steel bar processing process to the actual construction needs.

[0005] Based on this, the present invention proposes a steel bar production management system for precast beams at a construction site, establishes a floating influence model for three situations: the real-time adjustment amount of the steel bar processing plan, the change of the assembly rhythm, and the simultaneous change of the two, quantifies the specific influence of the floating on the steel bar processing process, dynamically adjusts the processing process, and effectively improves the self-adaptive ability of the steel bar processing process to the changes in the actual on-site needs. Summary of the Invention

[0006] The purpose of the present invention is to provide a steel bar production management system for precast beams at a construction site to solve the deficiencies in the background art.

[0007] To achieve the above purpose, the present invention provides the following technical solution: A steel bar production management system for precast beams at a construction site, including a collection module, a prediction module, a model establishment module, and a dynamic adjustment module;

[0008] Collection module: Real-time collect the steel bar processing plan information and the assembly rhythm status;

[0009] Prediction module: Predict the assembly rhythm status through a prediction model;

[0010] Model establishment module: Take the real-time steel bar processing plan and the prediction result of the assembly rhythm status as self-adaptive floating input variables, and establish a floating influence model with the steel bar processing as a dynamically controlled process;

[0011] Dynamic adjustment module: Dynamically adjust the steel bar processing process according to the floating influence model.

[0012] Preferably, the model establishment module collects the real-time adjustment amount of the steel bar processing plan and the prediction deviation value of the assembly rhythm state as the input layer of the BP network;

[0013] The BP neural network structure is: two inputs → four hidden layer nodes → one output. After the BP neural network training is completed, the network weights and biases are output;

[0014] The calculation of the neural network is first to obtain the activation value of each hidden layer node through the weighted sum from the input layer to the hidden layer;

[0015] Calculate the final output through the activation value of the hidden layer and the weights of the output layer.

[0016] Preferably, for each hidden layer neuron , the calculation formula is: , where is the weight input to the hidden layer node, is the input variable, is the bias term of the hidden layer node;

[0017] The activation value of the hidden layer in the neural network is calculated through the Sigmoid function. The formula of the Sigmoid function is: , where h is the hidden layer neuron. Calculate the final output through the activation value of the hidden layer and the weights of the output layer. The calculation formula of the output layer is:

[0018] , where is the output value, is the activation value of each hidden layer, is the output layer bias, is the weight from the hidden layer to the output layer.

[0019] Preferably, the prediction module collects data and selects a multi-variable linear regression model to predict the assembly rhythm state to predict the completion rate of the assembly rhythm state of girder i at time t;

[0020] Substitute the current time steel bar plan adjustment amount, historical rhythm benchmark value, and design change urgency intensity into the multi-variable linear regression model to calculate the completion rate of the assembly rhythm state of each girder at a future time.

[0021] Preferably, the prediction module collects data, including the actual progress data of the mold preparation, binding process, and concrete pouring of the current girder, the construction historical rhythm data of the last N similar girders, and the current steel bar processing plan adjustment, urgency, and change situation.

[0022] Preferably, the acquisition module obtains in real time the steel bar processing plan information released in the construction management platform, the project scheduling system, and the design change notice;

[0023] Through on-site Internet of Things devices, the progress management platform, and the manual input system, the assembly rhythm status of the beam construction site is collected in real time;

[0024] Automatically compare and confirm the status of the collected steel bar processing plan information and the assembly rhythm status. For missing, abnormal, or conflicting data, automatically push a warning to notify the management personnel;

[0025] Classify and code the confirmed steel bar processing plan information and the assembly rhythm status according to the preset data format and coding specifications.

[0026] Preferably, in the floating influence model, when there is a real-time adjustment amount of the steel bar processing plan, a change in the assembly rhythm status, and a simultaneous change in the steel bar processing plan and the assembly rhythm status, the steel bar processing process is dynamically adjusted.

[0027] Preferably, the steel bar processing plan information includes plan adjustments, on-site urgent requirements, and design changes, and the assembly rhythm status includes the preparation of beam molds, the binding process, and the progress nodes of concrete pouring.

[0028] In the above technical solution, the technical effects and advantages provided by the present invention:

[0029] In the present invention, the acquisition module collects in real time the steel bar processing plan information and the assembly rhythm status. The prediction module predicts the assembly rhythm status through the prediction model. The model establishment module uses the real-time steel bar processing plan and the prediction result of the assembly rhythm status as adaptive floating input variables. The steel bar processing is used as a dynamically controlled process to establish a floating influence model. In the floating influence model, when there is a real-time adjustment amount of the steel bar processing plan, a change in the assembly rhythm status, and a simultaneous change in the steel bar processing plan and the assembly rhythm status, the steel bar processing process is dynamically adjusted. The dynamic adjustment module dynamically adjusts the steel bar processing process according to the floating influence model. This management system constructs a floating influence dynamic regulation model to achieve dynamic response in multiple situations. The model establishment module uses the steel bar processing plan and the prediction result of the construction rhythm as adaptive floating input variables, and uses the steel bar processing operation as a dynamically controlled process to establish a floating influence model in three situations: the real-time adjustment amount of the steel bar processing plan, the change in the assembly rhythm, and the simultaneous change of both. This model can quantify the specific impact of floating on the steel bar processing process, dynamically adjust the processing scheduling order, batches, processing quantities, and distribution strategies, and effectively improve the adaptive ability of the steel bar processing process to changes in on-site actual requirements. Description of the Drawings

[0030] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the accompanying drawings required in the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0031] Figure 1 It is the system architecture diagram of the present invention.

[0032] Figure 2 It is the system timing diagram of the present invention.

[0033] Figure 3 It is the system mind map of the present invention. Specific embodiments

[0034] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0035] Embodiment 1: Please refer to Figures 1 - 3 As shown, a precast beam steel bar production management system at the construction site in this embodiment includes a collection module, a prediction module, a model establishment module, and a dynamic adjustment module;

[0036] Collection module: Real-time collect steel bar processing plan information and assembly rhythm status. The steel bar processing plan information includes plan adjustment, on-site urgent needs, and design changes. The assembly rhythm status includes progress nodes such as beam mold preparation, binding process, and concrete pouring. The steel bar processing plan information is sent to the model establishment module, and the assembly rhythm status is sent to the prediction module;

[0037] Prediction module: Predict the assembly rhythm status through a prediction model, and send the prediction result of the assembly rhythm status to the model establishment module;

[0038] Model establishment module: Use the real-time steel bar processing plan and the prediction result of the assembly rhythm status as adaptive floating input variables. The steel bar processing is a dynamically controlled process, and a floating influence model is established. The floating influence model includes the real-time adjustment amount of the steel bar processing plan, the change in the assembly rhythm status, and when the steel bar processing plan and the assembly rhythm status change simultaneously, dynamically adjust the steel bar processing process. The floating influence model is sent to the dynamic adjustment module;

[0039] Dynamic adjustment module: Dynamically adjust the steel bar processing process according to the floating influence model.

[0040] This application collects the steel bar processing plan information and the assembly rhythm status in real time through a collection module. The prediction module predicts the assembly rhythm status through a prediction model. The model establishment module uses the real-time steel bar processing plan and the prediction result of the assembly rhythm status as adaptive floating input variables. Since the steel bar processing is a dynamic controlled process, a floating influence model is established. The floating influence model includes the real-time adjustment amount of the steel bar processing plan, the change in the assembly rhythm status, and when both the steel bar processing plan and the assembly rhythm status change simultaneously, the steel bar processing process is dynamically adjusted. The dynamic adjustment module dynamically adjusts the steel bar processing process according to the floating influence model. This management system constructs a floating influence dynamic regulation model to achieve dynamic response in multiple scenarios. The model establishment module uses the steel bar processing plan and the construction rhythm prediction result as adaptive floating input variables, takes the steel bar processing operation as a dynamic controlled process, and establishes a floating influence model in three scenarios: the real-time adjustment amount of the steel bar processing plan, the change in the assembly rhythm, and the simultaneous change of both. This model can quantify the specific influence of floating on the steel bar processing process, dynamically adjust the processing scheduling order, batch, processing quantity, and distribution strategy, and effectively improve the adaptive ability of the steel bar processing process to the changes in the actual on-site requirements.

[0041] The specific working process of the management system is as follows:

[0042] The collection end collects the steel bar processing plan information and the assembly rhythm status in real time. The steel bar processing plan information includes plan adjustments, on-site urgent requirements, and design changes. The assembly rhythm status includes progress nodes such as beam formwork preparation, binding process, and concrete pouring. The assembly rhythm status is predicted through a prediction model. The real-time steel bar processing plan and the prediction result of the assembly rhythm status are used as adaptive floating input variables. Since the steel bar processing is a dynamic controlled process, a floating influence model is established. The floating influence model includes the real-time adjustment amount of the steel bar processing plan, the change in the assembly rhythm status, and when both the steel bar processing plan and the assembly rhythm status change simultaneously, the steel bar processing process is dynamically adjusted. The steel bar processing process is dynamically adjusted according to the floating influence model.

[0043] Embodiment 2: The collection module collects the steel bar processing plan information and the assembly rhythm status in real time. The steel bar processing plan information includes plan adjustments, on-site urgent requirements, and design changes. The assembly rhythm status includes progress nodes such as beam formwork preparation, binding process, and concrete pouring.

[0044] The collection module is used to collect the steel bar processing plan information and the assembly rhythm status at the construction site in real time to ensure the synchronization and matching of steel bar production and construction progress. This collection process includes the following steps:

[0045] Obtain the latest steel bar processing plan information released in the construction management platform, project scheduling system, and design change notice in real time. The content includes:

[0046] Plan adjustment information: Collect information such as new orders, cancelled orders, changed work orders, adjusted processing quantities, and changed processing priorities in the steel bar processing plan.

[0047] On-site urgent requirements: Collect on-site immediate information such as urgent processing, temporary replenishment of materials, and processing requirements for special specification steel bars temporarily proposed by the construction site or the supervision unit.

[0048] Design change information: Collect construction drawing modification notices issued by the design institute or the owner, including design change content involving adjustments to steel bar models, quantities, specifications, and structural forms.

[0049] After all the steel bar processing plan information is summarized in real time, it is automatically sent to the model establishment module as an important basis for regulating the steel bar processing process.

[0050] Through on-site Internet of Things devices, progress management platforms, and manual input systems, the assembly rhythm status of the beam body construction site is collected in real time, specifically including the following progress nodes:

[0051] Status of beam body mold preparation: Collect the completion status of mold processing, transportation, assembly, and inspection, and confirm whether the mold status meets the conditions for steel bar binding.

[0052] Progress of the steel bar binding process: Collect the completion status of process nodes such as steel bar feeding, positioning, binding completion, and acceptance, and mark the current steel bar binding progress of each beam body.

[0053] Concrete pouring progress: Collect the status of process nodes such as beam body steel bar acceptance, formwork closure, concrete pouring, and curing completion, and determine the construction rhythm of the beam body and the time sequence of processing requirements.

[0054] After all the assembly rhythm status information is summarized in real time, it is automatically sent to the prediction module for predicting the construction rhythm trend and as a basis for subsequent processing scheduling decisions.

[0055] Automatically compare and confirm the status of the collected steel bar processing plan information and the assembly rhythm status to ensure the integrity, accuracy, and timeliness of the information. For missing, abnormal, or conflicting data, automatically push warning notifications to the management personnel to ensure timely data synchronization and reliable real-time status. Classify and code the confirmed steel bar processing plan information and assembly rhythm status according to the preset data format and coding specifications for easy subsequent model invocation;

[0056] To ensure data standardization and compatibility, standardized coding structures are developed for the two types of data respectively, as shown in the following examples: Coding structure for steel bar processing plan information: GJ-YYYYMMDD-XXXX-TT-NN, where GJ: identification of steel bar processing plan, YYYYMMDD: plan release date, XXXX: beam body number, TT: plan type (01 = adjustment, 02 = urgent, 03 = change), NN: sequence number (two digits).

[0057] Coding structure for assembly rhythm status: AZ-YYYYMMDD-XXXX-SS-NN, where AZ: identification of assembly rhythm status, YYYYMMDD: status collection date, XXXX: beam body number, SS: process node (01 = mold preparation, 02 = tying up completed, 03 = concrete pouring), NN: status sequence number (two digits).

[0058] For the confirmed steel bar processing plan information, each item is coded according to the coding structure as shown in Table 1:

[0059] Table 1

[0060] Plan type Release date Beam body number Coding result Adjustment 20250421 L001 GJ-20250421-L001-01-01 Urgent 20250421 L002 GJ-20250421-L002-02-01 Design change 20250421 L003 GJ-20250421-L003-03-01

[0061] For the confirmed assembly rhythm status, each item is coded according to the coding structure as shown in Table 2:

[0062] Table 2

[0063] Process node Collection date Beam body number Coding result Mold preparation completed 20250421 L001 AZ-20250421-L001-01-01 Steel bar binding completed 20250421 L002 AZ-20250421-L002-02-01 Concrete pouring completed 20250421 L003 AZ-20250421-L003-03-01

[0064] Perform uniqueness verification on the generated coding information to prevent duplication or omission. After the verification is correct, the coded data is archived and stored in the system database for use by the model establishment module and the prediction module.

[0065] Through the above collection steps, the following can be achieved: real-time perception of changes in steel bar processing plans, on-site temporary requirements, and design adjustments; full-process dynamic monitoring of the assembly rhythm nodes of beam bodies at the construction site; ensuring data linkage and collaborative control between steel bar processing scheduling and construction progress, providing reliable and comprehensive real-time data support for subsequent prediction analysis and dynamic regulation.

[0066] The prediction module predicts the assembly rhythm status through a prediction model.

[0067] The prediction module is used to predict the progress trend of each assembly rhythm status during the beam body construction process based on the existing construction data and historical progress characteristics, ensuring the coordination and matching between steel bar processing scheduling and construction rhythm. The execution steps of the prediction module are as follows:

[0068] Collect and organize the following data: the actual progress data of the current beam body's mold preparation, binding process, and concrete pouring, the construction historical rhythm data of the most recent N similar beam bodies, the adjustment, urgency, and change situation of the current steel bar processing plan. Select a multi-variable linear regression model to predict the assembly rhythm state, and the prediction formula is as follows:

[0069] , where is the completion rate of the assembly rhythm state of beam body i at prediction time t, is the model constant term, and , is the influence coefficient of the adjustment of the steel bar processing plan on the assembly rhythm, is the quantity of the adjustment of the steel bar processing plan, is the influence coefficient of the historical rhythm reference value of the beam body on the current prediction, is the historical average rhythm completion rate of beam body i under similar conditions, is the influence coefficient of the current design change, is the quantity of the current design change.

[0070] In this application:

[0071] Method for obtaining the model constant term:

[0072] The model constant term represents the basic level of the assembly rhythm completion rate when all input factors (such as the adjustment of the steel bar processing plan, the historical rhythm reference value, the design change, etc.) are zero. The obtaining method: it is obtained by regression modeling and fitting. In the regression analysis, substitute the historical construction data into the model, use the least squares method to solve the regression equation, and the automatically calculated intercept value is the constant term.

[0073] Method for obtaining the influence coefficient of the adjustment of the steel bar processing plan on the assembly rhythm:

[0074] This coefficient represents the influence intensity of the adjustment quantity of the steel bar processing plan on the change of the assembly rhythm completion rate. The obtaining method: based on historical construction cases, collect the data of different adjustment quantities of the steel bar processing plan (such as the quantity of plan change, the frequency of temporary urgency) and the corresponding actual assembly rhythm completion rate data. Through regression analysis and fitting, statistically obtain the change value of the assembly rhythm completion rate when this factor changes by one unit, and use it as the coefficient. The more frequent and larger the plan adjustment, the larger the coefficient value, indicating that its interference with the rhythm is more obvious.

[0075] Method for obtaining the influence coefficient of the historical rhythm reference value of the beam body on the current prediction:

[0076] This coefficient represents the reference weight of the historical assembly rhythm benchmark of the beam body under similar construction conditions for the current predicted progress. Acquisition method: Summarize the historical actual assembly rhythm completion rates of the same type of beam bodies under similar conditions (season, construction team, equipment, cooperating unit), calculate the average value, and use it as the rhythm benchmark value. During the regression modeling process, based on the corresponding relationship between these historical benchmark values and the current assembly rhythm completion rate, the linear influence coefficient of this historical value on the current progress prediction is obtained by fitting.

[0077] The acquisition method of the influence coefficient of the current design change:

[0078] This coefficient represents the degree of interference of the number of design changes, the complexity of change content, the emergency level, etc. involved in the current construction process on the assembly rhythm progress. Acquisition method: Statistically analyze the number of design changes, change levels (ordinary, important, urgent) and the changes in the assembly rhythm completion rate occurring in different construction stages. Quantify the change quantity and change level into a comprehensive change intensity index, and then establish a regression relationship with the progress change amount. The influence coefficient generated by each unit change in the design change intensity on the assembly rhythm completion rate is obtained by fitting.

[0079] Substitute the current moment's steel bar plan adjustment amount, historical rhythm benchmark value, and design change urgency intensity into the formula to calculate the assembly rhythm state completion rate of each beam body at the future moment.

[0080] In this application, assume:

[0081]

[0082] Substituting into the formula gives: 10 + 0.5×20 + 0.8×75 - 0.3×5 = 10 + 10 + 60 - 1.5 = 78.5 = 10 + 10 + 60 - 1.5 = 78.5 = 10 + 10 + 60 - 1.5 = 78.5, that is, the predicted assembly rhythm state completion rate of beam body i at time t is 78.5%.

[0083] Send the predicted completion rate results of all beam bodies to the model establishment module as adaptive floating input variables to participate in the dynamic process adjustment of steel bar processing.

[0084] In this application, scikit-learn is used for modeling, and the code is as follows:

[0085] Multivariable linear regression to predict the assembly rhythm state:

[0086] import - numpy as np

[0087] Import - pandas as pd

[0088] From sklearn.linear_model import LinearRegression

[0089] # Simulated data: Each row corresponds to a construction record sample

[0090] # G: Adjustment amount of steel bar plan at the current moment

[0091] # H: Historical rhythm benchmark value

[0092] # D: Intensity of urgent design changes

[0093] # P: Actual completion rate (target value)

[0094] data = {

[0095] 'G': [20, 15, 30, 25, 10],

[0096] 'H': [75, 70, 80, 78, 68],

[0097] 'D': [5, 3, 8, 6, 2],

[0098] 'P': [78.5, 74.0, 85.0, 82.5, 70.0]

[0099] }

[0100] # Convert to DataFrame

[0101] df = pd.DataFrame(data)

[0102] # Features and target value

[0103] X = df[['G', 'H', 'D']] # Input variables

[0104] y = df['P'] # Target variable

[0105] # Create a regression model

[0106] model = LinearRegression()

[0107] # Fit the model

[0108] model.fit(X, y)

[0109] # Output regression coefficients

[0110] print("Regression coefficients (alpha1, alpha2, alpha3):", model.coef_)

[0111] print("Constant term (alpha0):", model.intercept_)

[0112] # Use the model for prediction (such as new input data)

[0113] new_data = np.array([[22, 76, 4]]) # G = 22, H = 76, D = 4

[0114] predicted_P = model.predict(new_data)

[0115] print("Predicted assembly rhythm completion rate:", predicted_P[0], "%")

[0116] Code description:

[0117] Input variables: G - Adjustment amount of steel bar plan at the current moment; H - Historical rhythm benchmark value; D - Intensity of urgent design changes.

[0118] Target variable: Actual completion rate;

[0119] After fitting, we can get:

[0120] (Regression coefficients) 、 、 ;

[0121] (Constant term) ;

[0122] New G, H, and D values can be input at any time to predict the new assembly rhythm completion rate. Suppose after fitting, we get: Regression coefficients (alpha1, alpha2, alpha3): [0.5 0.8 -0.3]; Constant term (alpha0): 10.0; Predicted assembly rhythm completion rate: 80.9%.

[0123] Among them, the adjustment amount of the steel bar plan can be obtained by real-time collecting the adjustment data in the steel bar processing plan; the historical rhythm benchmark value can be obtained by reviewing historical construction data or setting benchmark data; the intensity of design changes can be calculated and obtained through the urgent part or the change amount in the design change notice.

[0124] This application quantifies the dynamic impact of different factors on the construction rhythm, predicts the progress completion rate of the beam body at a specific time point, predicts the trend of rhythm advancement or lag based on the adjustment of the steel bar plan, historical rhythm benchmark, and design change situation, facilitates on-site scheduling decisions, and uses the prediction results to guide the adjustment of the steel bar processing plan to match the rhythm dynamically, preventing processing accumulation or waiting for materials in construction, converting real-time data into quantifiable and predictable progress control indicators, and improving the intelligent level of on-site production scheduling.

[0125] The model building module takes the real-time rebar processing plan and the assembly rhythm status prediction results as adaptive floating input variables, and the rebar processing is taken as a dynamically controlled process to establish a floating influence model. The floating influence model includes the real-time adjustment amount of the rebar processing plan, the change of the assembly rhythm status, and the dynamic adjustment of the rebar processing process when the rebar processing plan and the assembly rhythm status change at the same time.

[0126] In this application, steel bar processing is a dynamically controlled process: the steel bar processing process is a process that is dynamically adjusted according to real-time plan adjustments and construction progress. For example, the processing speed, quantity, processing sequence, etc. of the steel bars will be dynamically adjusted according to the changes in the on-site construction progress and plan. This process can be automatically adjusted through a control system or a feedback mechanism based on a predictive model.

[0127] Suppose that in a bridge construction project, the steel bar processing needs to be dynamically adjusted according to the construction progress on site. If the progress of steel bar processing lags behind the scheduled plan, or there is a deviation in the construction progress, the system will make adjustments based on real-time data. The following is an example of how each module works:

[0128] Assume that at a specific moment: the real-time steel bar processing plan shows that the steel bar processing volume in the current plan is 100 tons. The assembly rhythm state prediction result is based on historical data and the current construction status, and predicts that the assembly rhythm state completion rate at this moment is 80% (that is, the expected progress is 80%).

[0129] Rebar processing is a dynamically controlled process, which means that its progress and adjustments are in real-time response to changes on the construction site. For example, if the construction schedule is behind schedule, the rebar processing plan needs to be dynamically adjusted to increase the processing speed or add processing personnel. If there are design changes, the rebar processing plan may also need to be modified, which will affect the entire production process.

[0130] The core of the floating impact model is to consider how multiple factors interact to affect the progress of rebar processing. The model requires the following variables as input:

[0131] Rebar processing plan changes (#G): For example, an increase or decrease in the planned number of rebars, or accelerated production due to urgent needs.

[0132] Assembly rhythm status change (#S): For example, delays in mold preparation, steel bar tying process, or concrete pouring progress during construction.

[0133] Coupling effect of rebar processing plan and assembly rhythm status (#G * #S): The mutual influence between the adjustment of rebar processing and the construction rhythm. For example, the adjustment of rebar processing speed may affect the overall assembly rhythm, and vice versa.

[0134] Suppose at this time, the steel bar processing plan increases by 5 tons (#G = 5), and at the same time, the assembly rhythm state lags behind by 10% due to the delay in mold preparation (#S = -10%). The coupling effect of the two may cause the processing speed to increase by 10.

[0135] According to the floating influence model, the system will use a neural network (such as a BP neural network) to dynamically adjust the processing process. For example:

[0136] Neural network input: The adjustment amount of the steel bar processing plan (#G = 5), the deviation of the assembly rhythm state (#S = -10%), and the coupling effect (#G * #S = 5 * -10% = -0.5) are used as inputs.

[0137] Neural network calculation: The network compares these inputs with historical data and learns the optimal adjustment strategy. Suppose the output of the neural network is an adjustment coefficient (#R), which determines the change in the steel bar processing speed. For example, the output result of the neural network is an adjustment coefficient of 1.2, indicating that the processing speed of the steel bar needs to increase by 20%.

[0138] In the above process, the standard speed of steel bar processing is 100 tons per hour. According to the calculation result of the floating influence model, the new processing speed will become: Adjusted steel bar processing speed = 100 × 1.2 = 120 tons per hour. In this way, the processing speed of the steel bar is adjusted in real time to cope with the changes in the on-site progress and ensure that the overall construction progress is synchronized with the steel bar processing plan.

[0139] On the construction site, the steel bar processing progress is always affected by the real-time adjustment amount of the steel bar processing plan and the fluctuation of the assembly rhythm. When the two change simultaneously, the impact on the processing process is greater. By using a BP neural network and taking multi-dimensional floating information as inputs, the steel bar processing rhythm is adaptively adjusted to achieve intelligent adjustment under multi-factor dynamic coupling, so as to construct a floating influence model.

[0140] The model establishment module collects the real-time adjustment amount of the steel bar processing plan and the predicted deviation value of the assembly rhythm state. These two are used as input variables, which float in real time with time and serve as the input layer of the BP network.

[0141] The predicted deviation value of the assembly rhythm state is obtained by comparing the difference between the historical rhythm completion rate and the predicted rhythm completion rate. Specifically, the assembly rhythm state at the current moment can be predicted through a multi-variable regression model, and then the actual progress is compared with the predicted result to obtain the deviation value, that is, the predicted deviation value of the assembly rhythm state is obtained by subtracting the predicted rhythm completion rate from the historical rhythm completion rate. The larger the predicted deviation value of the assembly rhythm state, the more the predicted result deviates from the historical reference value.

[0142] The structure of the BP neural network is: two inputs → four hidden layer nodes → one output. After the BP neural network training is completed, the network weights and biases are output. The calculation of the neural network first obtains the activation value of each hidden layer node through the weighted sum from the input layer to the hidden layer. For each hidden layer neuron , its calculation formula is: , where is the weight input to the hidden layer node, is the input variable (i.e., the real-time adjustment amount of the steel bar processing plan, the assembly rhythm state deviation, and the coupling term), is the bias term of the hidden layer node;

[0143] The activation value of the hidden layer in the neural network is calculated through the Sigmoid function. The formula of the Sigmoid function is: , where h is the hidden layer neuron. Through the activation value of the hidden layer and the weights of the output layer, the final output is calculated. The calculation formula of the output layer is:

[0144] , where is the output value, is the activation value of each hidden layer, is the output layer bias, and is the weight from the hidden layer to the output layer. Assuming the standard speed of steel bar processing is 100 pieces per hour, according to the output result of the neural network, we apply the processing rhythm adjustment coefficient R(t) to the processing rate, that is, the adjusted speed = 100×R(t)=100×0.572 = 57.2 pieces per hour. The result shows that after the dynamic adjustment of the neural network, the steel bar processing speed should be significantly reduced, and the adjusted speed is 57.2 pieces per hour.

[0145] Through the calculation of the BP neural network model, we can dynamically adjust the steel bar processing rhythm based on the changes in the steel bar processing plan, the predicted deviation of the assembly rhythm state, and the coupling effect between the two. This can automatically optimize the production schedule and ensure the smooth progress of the processing process when facing on-site changes.

[0146] Key points:

[0147] The real-time adjustment amount of the steel bar processing plan and the assembly rhythm state deviation value are the key inputs affecting the processing rhythm.

[0148] The coupling term is also considered in the calculation. The coupling term is the interaction effect between the real-time adjustment amount of the steel bar processing plan and the assembly rhythm state deviation value. Through historical data analysis, the influence relationship between the changes in the steel bar processing plan and the assembly rhythm can be modeled and then used in the BP neural network.

[0149] The steps to obtain the coupling term are as follows:

[0150] Collect the past steel bar processing plans and the assembly rhythm data at the construction site. The data should include relevant information such as the adjustment amount of the steel bar plan, the completion rate of the assembly rhythm, the historical rhythm benchmark value, and design changes. Each historical record should contain the adjustment amount of the steel bar processing plan (#G), the actual assembly rhythm status (#S), and the deviation between the two at that moment.

[0151] Pair the adjustment amount of the steel bar processing plan (#G) in the historical data with the deviation value of the assembly rhythm status (#S), and observe the relationship between the two. For example, when the adjustment amount of the steel bar processing plan increases, whether there will be a corresponding change in the assembly rhythm status, and vice versa.

[0152] Calculate the interaction effect between the change in the steel bar processing plan and the change in the assembly rhythm status. The expression is: the coupling term is obtained by multiplying the adjustment amount of the steel bar processing plan by the deviation value of the assembly rhythm status and then multiplying by the interaction coefficient. The interaction coefficient is obtained through historical data analysis and usually takes a value of 0.8 - 1.0.

[0153] The neural network model performs complex non - linear mapping through the weights and biases of the hidden layer, providing an adaptive rhythm adjustment scheme.

[0154] According to the real - time data change situation, it is divided into the following three categories: the steel bar processing plan changes alone, the assembly rhythm status changes alone, and both change simultaneously.

[0155] The following is a code example based on the BP neural network, showing how to dynamically adjust the steel bar processing process according to the real - time adjustment amount of the steel bar processing plan, the change in the assembly rhythm status, and the coupling effect between the two. We will use Python and the TensorFlow / Keras libraries to implement this neural network.

[0156] Environment preparation: pip_install_tensorflow_numpy;

[0157] Python code example:

[0158] import_numpy_as_np

[0159] import_tensorflow_as_tf

[0160] from_tensorflow.keras.models_import_Sequential

[0161] from_tensorflow.keras.layers_import_Dense

[0162] from tensorflow.keras.optimizers import Adam

[0163] # Data preparation

[0164] # Assume we have the following input data (real-time adjustment amount of steel bar processing plan, assembly rhythm state deviation, coupling term)

[0165] X = np.array([[0.08, -0.04, -0.0032], # Input sample 1

[0166] [0.1, -0.03, -0.002], # Input sample 2

[0167] [0.05, -0.05, -0.0025], # Input sample 3

[0168] [0.07, -0.06, -0.003], # Input sample 4

[0169] [0.09, -0.02, -0.0034]]) # Input sample 5

[0170] # Corresponding target output data (steel bar processing rhythm adjustment coefficient)

[0171] y = np.array([[0.572], # Output sample 1

[0172] [0.6], # Output sample 2

[0173] [0.55], # Output sample 3

[0174] [0.65], # Output sample 4

[0175] [0.58]]) # Output sample 5

[0176] # Create a neural network model

[0177] model = Sequential()

[0178] # Input layer to hidden layer

[0179] model.add(Dense(4, input_dim = 3, activation='relu')) # 3 input features, 4 hidden nodes

[0180] # Hidden layer to output layer

[0181] model.add(Dense(1, activation='linear')) # Output a value: steel bar processing rhythm adjustment coefficient

[0182] # Compiled model

[0183] model.compile(loss='mean_squared_error', optimizer=Adam(learning_rate=0.001))

[0184] # Train the model

[0185] model.fit(X, y, epochs=200, batch_size=1)

[0186] # Predict the new adjustment coefficient for steel bar processing rhythm

[0187] new_input = np.array([[0.08, -0.04, -0.0032]]) # New input sample

[0188] predicted_adjustment = model.predict(new_input)

[0189] print(f"Predicted adjustment coefficient for steel bar processing rhythm: {predicted_adjustment[0][0]}")

[0190] # Assume the standard speed of steel bar processing is 100 pieces per hour

[0191] standard_speed = 100

[0192] adjusted_speed = standard_speed * predicted_adjustment[0][0]

[0193] print(f"Adjusted speed of steel bar processing: {adjusted_speed} pieces per hour")

[0194] X is the input data matrix, which contains the real-time adjustment amount of the steel bar processing plan, the deviation of the assembly rhythm state, and the coupling term. y is the target output data, representing the corresponding adjustment coefficient of the steel bar processing rhythm.

[0195] Neural network architecture: Create a Sequential model using Keras. The input layer has 3 features (real-time adjustment amount of the steel bar processing plan, deviation of the assembly rhythm state, coupling term). The hidden layer has 4 neurons and uses the ReLU activation function. The output layer has 1 neuron, representing the adjustment coefficient of the steel bar processing rhythm, and uses the linear activation function.

[0196] The model is compiled using the Adam optimizer and the mean_squared_error loss function. When training the model, the input data X and the output data y are used for 200 iterations. After training, the model can predict the coefficient for adjusting the steel bar processing rhythm based on new inputs. Based on the predicted adjustment coefficient, the adjusted steel bar processing speed is calculated.

[0197] The dynamic adjustment module dynamically adjusts the steel bar processing process according to the floating influence model.

[0198] The dynamic adjustment module first needs to receive the floating influence model generated by the model establishment module. This model includes the real-time adjustment amount of the steel bar processing plan, the change in the assembly rhythm state, and the influence when they change together. The model contains various factors affecting the steel bar processing process and their relationships.

[0199] Obtain the latest steel bar processing plan information (including plan adjustments, on-site urgent requirements, and design changes) and assembly rhythm state information (such as progress nodes for beam mold preparation, tying operations, concrete pouring, etc.) from the acquisition module.

[0200] According to the current steel bar processing plan and assembly rhythm state, dynamically calculate the floating influence between them. The floating influence can be obtained through weighted calculation or neural networks, etc. Common floating factors include:

[0201] Real-time adjustment amount of the steel bar processing plan (ΔP): such as plan adjustments, urgent requirements, etc.

[0202] Change in the assembly rhythm state (ΔS): such as delays in mold preparation progress, lags in tying operation progress, etc.

[0203] Simultaneous change in the steel bar processing plan and the assembly rhythm state (ΔP*ΔS): the synergistic effect of the two, which may lead to a greater floating influence.

[0204] Execute specific adjustments according to the adjustment strategy through an automated control system or manual intervention. For example, if it is calculated that the processing speed needs to be increased, the equipment can be instructed to speed up; if it is calculated that there is a process delay, more labor may need to be scheduled or the equipment may need to be replaced.

[0205] The dynamic adjustment module should have a real-time monitoring and feedback mechanism. By monitoring the real-time progress, the dynamic adjustment module can evaluate the effect of the adjustment strategy and further optimize the steel bar processing process based on the feedback. This step can be achieved through continuous data collection and analysis.

[0206] The dynamic adjustment module affects the model through floating and can adjust the steel bar processing process in real time to ensure the optimal coordination of production progress and resources. In the above example, the changes in the steel bar processing plan and the assembly rhythm status directly affected the adjustment of the processing speed, thus optimizing the resource utilization and time arrangement in the production process.

[0207] In the description of this specification, the descriptions referring to terms such as "one embodiment", "example", "specific example", etc. mean that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0208] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the present invention to the specific embodiments described. Obviously, according to the content of this specification, many modifications and changes can be made. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can well understand and utilize the present invention. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A precast beam steel bar production management system for construction sites, characterized in that: It includes a collection module, a prediction module, a model establishment module, and a dynamic adjustment module; Collection module: Collect the steel bar processing plan information and the assembly rhythm status in real time; Prediction module: Predict the assembly rhythm status through a prediction model. Substitute the current moment's steel bar plan adjustment amount, historical rhythm reference value, and design change urgency intensity into the multivariable linear regression model to calculate the completion rate of the assembly rhythm status of each beam body at a future moment; Model establishment module: Take the real-time steel bar processing plan and the prediction result of the assembly rhythm status as adaptive floating input variables. The steel bar processing is a dynamically controlled process, and the dynamically controlled process is a process that is dynamically adjusted according to the real-time plan adjustment and construction progress, and establish a floating influence model; Dynamic adjustment module: Dynamically adjust the steel bar processing process according to the floating influence model.

2. The precast beam steel bar production management system at the construction site according to claim 1, wherein: The model establishment module collects the real-time adjustment amount of the steel bar processing plan and the prediction deviation value of the assembly rhythm status as the input layer of the BP network; The BP neural network structure is: two inputs → four hidden layer nodes → one output. After the BP neural network training is completed, the network weights and biases are output; The calculation of the neural network first obtains the activation value of each hidden layer node through the weighted sum from the input layer to the hidden layer; Calculate the final output through the activation value of the hidden layer and the weights of the output layer.

3. The prefabricated beam steel bar production management system at the construction site according to claim 2, characterized in that: For each hidden layer neuron , the calculation formula is: , where are the weights input to the hidden layer node, are the input variables, is the bias term of the hidden layer node; Activation values of the hidden layer in the neural network are calculated through the Sigmoid function. The formula of the Sigmoid function is: , where h is the neuron in the hidden layer. The final output is calculated through the activation value of the hidden layer and the weights of the output layer. The calculation formula of the output layer is: , wherein, is the output value, is the activation value of each hidden layer, is the output layer bias, is the weight from the hidden layer to the output layer.

4. The precast beam steel bar production management system at the construction site according to claim 3, characterized in that: The prediction module collects data and selects a multivariable linear regression model to predict the assembly rhythm status to predict the completion rate of the assembly rhythm status of beam body i at prediction moment t.

5. The prefabricated beam steel bar production management system at the construction site according to claim 4, characterized in that: The prediction module collects data, including the actual progress data of the mold preparation, binding process, and concrete pouring of the current beam body, the construction historical rhythm data of the recent N similar beam bodies, and the current steel bar processing plan adjustment, urgency, and change situation.

6. The precast beam steel bar production management system at the construction site according to claim 5, characterized in that: The collection module obtains the steel bar processing plan information released in the construction management platform, project scheduling system, and design change notice in real time; Through on-site Internet of Things devices, progress management platforms, and manual input systems, collect the assembly rhythm status of the beam body construction site in real time; Automatically compare and confirm the status of the collected steel bar processing plan information and the assembly rhythm status. For missing, abnormal, or conflicting data, automatically push a warning notice to the management personnel; Classify and code the confirmed steel bar processing plan information and assembly rhythm status according to the preset data format and coding specification.

7. The precast beam steel bar production management system at the construction site according to claim 6, characterized in that: In the floating influence model, it includes the real-time adjustment amount of the steel bar processing plan, the change of the assembly rhythm status, and when the steel bar processing plan and the assembly rhythm status change simultaneously, dynamically adjust the steel bar processing process.

8. The precast beam steel bar production management system at the construction site according to claim 7, characterized in that: The steel bar processing plan information includes plan adjustment, on-site urgency requirements, and design changes. The assembly rhythm status includes the progress nodes of beam body mold preparation, binding process, and concrete pouring.

Images