Method for outputting list of rebars from artificial intelligence-based architectural drawing and controlling equipment of rebar processing plant

Abstract

The present invention relates to a method for outputting a list of rebar from an artificial intelligence-based architectural drawing and controlling equipment of a rebar processing plant. The method may be characterized by comprising the steps of: a customer terminal inputting an architectural drawing into a main server; the main server converting the architectural drawing into a shop drawing using artificial intelligence; the main server outputting a list of rebar from the shop drawing using artificial intelligence; an equipment control terminal receiving a list of rebar from the main server; the equipment control terminal outputting an equipment control command from the list of rebar using artificial intelligence; and a plurality of equipment operating according to the equipment control command of the equipment control terminal.

Description

Technology Field

[0001] The present invention relates to an artificial intelligence-based method for extracting rebar processing information from floor plans and shop drawings of buildings in the field of construction, particularly, and for automatically controlling equipment (cutting machines, bending machines, etc.) of a rebar processing plant. Background Technology

[0002] The conventional rebar processing process involved human personnel interpreting drawings to manually create rebar bar lists and inputting commands into processing equipment based on that information. This method carries a high risk of human error, is time-consuming, and has limitations in terms of automation and application at large-scale construction sites. Despite recent advancements in Building Information Modeling (BIM) and AI technologies, a system that integrates the entire process—from drawing interpretation to automated factory control—is currently lacking. The problem to be solved

[0003] The problem that the present invention aims to solve is to provide a method for outputting a list of rebars from an AI-based architectural drawing and controlling the equipment of a rebar processing plant.

[0004] The problems that the present invention aims to solve are not limited to those mentioned above, and other unmentioned problems will be clearly understood by a person skilled in the art from the description below. means of solving the problem

[0005] The method of the present invention for solving the above-mentioned problem may be characterized by comprising the steps of: a customer terminal inputting an architectural drawing to a main server; the main server converting the architectural drawing into a shop drawing using artificial intelligence; the main server outputting a list of rebar from the shop drawing using artificial intelligence; an equipment control terminal receiving a list of rebar from the main server; the equipment control terminal outputting an equipment control command from the list of rebar using artificial intelligence; and a plurality of equipment operating according to the equipment control command of the equipment control terminal.

[0006] The step of converting the above architectural drawing into a shop drawing may be characterized by being performed using a deep learning-based artificial intelligence that object-recognizes structural elements of the architectural floor plan and converts them into a rebar placement drawing according to structural analysis standards.

[0007] The step of outputting a rebar list from the above shop drawing may be characterized by extracting rebar type, thickness, length, quantity, and location information listed in the shop drawing using OCR and object recognition algorithms, and automatically converting it into a standardized bar list form.

[0008] The step of the above equipment control terminal outputting equipment control commands from a rebar list may be characterized by deriving an optimal work order through a reinforcement learning-based scheduling algorithm by considering the work speed, availability time, priority, etc. of the plurality of equipment, and automatically generating processing commands for each of the plurality of equipment according to the result.

[0009] The step of the plurality of equipment operating according to an equipment control command may be characterized by receiving feedback information such as working distance, temperature, and cutting pressure from sensors attached to the plurality of equipment, and a time-series based LSTM artificial intelligence model analyzing the information to correct real-time control errors. Effects of the invention

[0010] According to the method of the present invention, by automating the entire process from drawing interpretation to equipment control using artificial intelligence, there are advantages such as reduced labor costs, minimized errors, and improved work efficiency.

[0011] In addition, according to the method of the present invention, the speed is improved by more than five times compared to conventional manual work by automating the bar list using artificial intelligence.

[0012] Furthermore, according to the method of the present invention, there is an advantage of realizing factory automation by linking with processing equipment using artificial intelligence and providing a foundation for implementing a smart factory.

[0013] The effects of the present invention are not limited to those mentioned above, and other unmentioned effects will be clearly understood by a person skilled in the art from the description below. Brief explanation of the drawing

[0014] Figure 1 is a conceptual diagram illustrating a method of outputting a list of reinforcing bars from an artificial intelligence-based architectural drawing and controlling the equipment of a reinforcing bar processing plant according to the present invention. Figure 2 is a flowchart illustrating a method for outputting a list of rebar from an AI-based architectural drawing and controlling the equipment of a rebar processing plant. Specific details for implementing the invention

[0015] The advantages and features of the present invention and the methods for achieving them will become clear by referring to the embodiments described below in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below but may be implemented in various different forms. These embodiments are provided merely to ensure that the disclosure of the present invention is complete and to fully inform those skilled in the art of the scope of the present invention, and the present invention is defined only by the scope of the claims.

[0016] The terms used in this specification are for describing embodiments and are not intended to limit the invention. In this specification, the singular form includes the plural form unless specifically stated otherwise in the text. The terms "comprises" and / or "comprising" used in this specification do not exclude the presence or addition of one or more other components in addition to the components mentioned. Throughout the specification, the same reference numerals refer to the same components, and "and / or" includes each of the mentioned components and all combinations of one or more. Although terms such as "first," "second," etc., are used to describe various components, these components are not limited by these terms. These terms are used merely to distinguish one component from another. Therefore, the first component mentioned below may be the second component within the technical scope of the invention.

[0017] Unless otherwise defined, all terms used herein (including technical and scientific terms) may be used in a meaning commonly understood by those skilled in the art to which the present invention pertains. Additionally, terms defined in commonly used dictionaries are not to be interpreted ideally or excessively unless explicitly and specifically defined otherwise.

[0019] Hereinafter, a method (1000) for outputting a list of reinforcing bars in an AI-based architectural drawing and controlling the equipment of a reinforcing bar processing plant is described with reference to the drawings. FIG. 1 is a conceptual diagram showing the method of outputting a list of reinforcing bars in an AI-based architectural drawing and controlling the equipment of a reinforcing bar processing plant according to the present invention, and FIG. 2 is a flowchart showing the method of outputting a list of reinforcing bars in an AI-based architectural drawing and controlling the equipment of a reinforcing bar processing plant.

[0021] The method (1000) of the present invention can be performed within a system comprising a customer terminal (10), an equipment control terminal (20), a plurality of equipment (30), a network (40), and a main server (50).

[0022] The customer terminal (10) and the equipment control terminal (20) may be terminals provided with a web page, an app page (app list), a program, or an application related to the method (1000) of the present invention.

[0023] Multiple steel bars (30) may be equipment of a steel bar processing plant controlled by an equipment control terminal (20).

[0024] In this case, the customer terminal (10) and the equipment control terminal (20) can be implemented as computers that can be connected to and accessed via a network (40) a main server (50) at a remote location.

[0025] In detail, the customer terminal (10) and the equipment control terminal (20) are wireless communication devices and may include all types of handheld-based wireless communication devices such as navigation, PCS (Personal Communication System), GSM (Global System for Mobile communications), PDC (Personal Digital Cellular), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), Wibro (Wireless Broadband Internet) terminals, smartphones, smartpads, tablet PCs, etc., as well as wired communication devices such as general desktop PCs, but are not limited thereto.

[0026] A network (40) refers to a connection structure capable of exchanging information between each node, such as a customer terminal (10) and an equipment control terminal (20). Examples of such networks include, but are not limited to, RF, 3GPP (3rd Generation Partnership Project) network, LTE (Long Term Evolution) network, 5GPP (5th Generation Partnership Project) network, WIMAX (World Interoperability for Microwave Access) network, Internet, LAN (Local Area Network), Wireless LAN (Wireless Local Area Network), WAN (Wide Area Network), PAN (Personal Area Network), Bluetooth network, NFC network, satellite broadcasting network, analog broadcasting network, DMB (Digital Multimedia Broadcasting) network.

[0027] The main server (50) is the main server of an operating platform operated by a platform manager, and may be a server that provides a web page, app page, program, or application related to the method (1000) of the present invention.

[0029] The method (1000) of the present invention may include the steps of: a customer terminal (10) inputting an architectural drawing into a main server (50) (100); the main server (50) converting the architectural drawing into a shop drawing using artificial intelligence (200); the main server (50) outputting a list of rebar from the shop drawing using artificial intelligence (300); the equipment control terminal (20) receiving the list of rebar from the main server (50) (400); the equipment control terminal (20) outputting an equipment control command from the list of rebar using artificial intelligence (500); and a plurality of equipment (30) operating according to the equipment control command of the equipment control terminal (20) (600).

[0030] Artificial intelligence (AI) used in the method (1000) of the present invention is a field of computer engineering and information technology that realizes human learning ability, reasoning ability, perception ability, and natural language understanding ability as computer programs. In particular, supervised learning, which is utilized in the most fields among artificial intelligence, is a method of predicting future values ​​using training data that includes correct answers (labels).

[0031] For reference, machine learning can be defined as a methodology that enables computers to learn autonomously without being explicitly programmed; it is a method in which a program learns patterns from data on its own. Machine learning is classified into supervised learning and unsupervised learning depending on whether correct answers are specified in the data required for learning. Depending on the purpose of use, it is further categorized into methodologies such as classification, which divides data into a finite number of categories; regression, which maps data to continuous values; clustering, which groups similar data; and dimensionality reduction, which maps multidimensional data to a representative lower dimension.

[0032] Deep learning is a machine learning technique that has dramatically improved the performance of machine learning, which had been stagnant for some time. Deep learning is a methodology based on Artificial Neural Network (ANN) algorithms that mimic the superposition of synapses in the structure of the human brain. Deep learning structures can include Deep Neural Networks (DNNs), which have multiple hidden layers between the input layer and the output layer; Convolutional Neural Networks (CNNs), which place filters necessary for factor extraction in front of the hidden layers and learn the filters together; and Recurrent Neural Networks (RNNs), which can process time-series data by stacking artificial neural networks at each time step. Here, the high performance of deep learning models is explained in two ways. First, artificial neural networks are universal approximators capable of approximating all types of functions through the superposition of weighted sums of functions in each layer; therefore, if sufficiently general data is provided, they can simulate data with high accuracy. Second, in order to effectively classify data, it is important to appropriately extract representative factors; this can be achieved by utilizing spiral neural networks and learning filters to extract optimal factors. Furthermore, deep learning is an advanced form of the artificial intelligence model known as the neural network, characterized by a structure in which hidden layers within the hierarchical neural network are arranged in multiple stages. Recent deep learning models have increased the number of hidden layers, with the number of weights (representing connection strength) connecting nodes reaching up to billions.

[0033] The step (100) in which the client terminal (10) inputs the architectural drawing to the main server (50) may be a step in which a client, such as a construction company requesting processed rebar, inputs the architectural drawing for calculating the processed rebar into the platform.

[0034] The step (200) in which the main server (50) uses artificial intelligence to change the architectural drawing into a shop drawing may be a step of changing the architectural drawing, which is a 2D floor plan or 3D drawing without reinforcing bars indicated on the platform, into a shop drawing with reinforcing bars.

[0035] In this case, the artificial intelligence may be characterized by being performed using a deep learning-based artificial intelligence that object-recognizes structural elements of an architectural floor plan and converts them into a rebar placement drawing (shop drawing) according to structural analysis standards.

[0036] The artificial intelligence system used in the present invention may include the following components to automatically generate a shop drawing from an architectural drawing.

[0037] In this case, the input data for the system can be broadly divided into three categories. First, design drawings such as 2D architectural floor plans or 2D BIM models (IFC format) may be input; these drawings include the overall structure and layout of the building and are typically provided in DWG, PDF, or IFC formats. Second, information on structural members included in each drawing is input. Structural members refer to load-bearing elements such as beams, columns, walls, and slabs, and their location, orientation, and cross-sectional shape may be included as data. Third, material specification tables and design standard information used as design and construction standards may be provided. For example, this may include rebar material specifications such as SD400 and SD500, design codes such as KS or ACI, and standards regarding reinforcement spacing and cover thickness.

[0038] The core of the present invention consists of the following three artificial intelligence modules. First, the object recognition module (Structure Detector) utilizes the latest deep learning-based object detection algorithms such as YOLOv8 or Vision Transformer (ViT). It analyzes structural member elements (e.g., columns, beams, walls, openings, etc.) present on the drawing using a semantic segmentation method and automatically extracts information regarding the location, orientation, length, width, and material of each element. This process includes converting pixel data from the drawing into vector data and processing it into structural information that a machine can understand.

[0039] Next, the structural analysis-based rebar placement judgment module (Rebar Pattern AI) is an artificial intelligence that learns rebar placement rules according to the structural role of each member and automatically determines them based on structural member information extracted in the object recognition step. For example, in a beam, tensile reinforcement is placed at the top and bottom, and stirrups to fix it may also be included; in a column, main bars and ties (spirals) may be regularly placed; and in a slab, bidirectional main bars (up, down, left, and right directions) and on-top reinforcement (reinforcement) may be set according to the required locations. This rebar placement judgment applies reinforcement learning (Policy Gradient) according to structural design standards (KDS, ACI, etc.) to derive an optimized rebar placement that considers the connection relationships between members, load distribution, and the influence of openings.

[0040] Next, the shape generation module (Rebar Drawing Generator) may be a module that ultimately generates rebar shape data that can be displayed on an actual drawing based on the rebar placement judgment results. It automatically encodes the shape, length, diameter, location coordinates, and quantity of each rebar and outputs them in a shop drawing format, and the results can be exported in vector formats such as DWG, PDF, and SVG. This process is performed by a Generative Transformer-based vector rendering algorithm and provides a level of precision that allows the drawing to be used directly for factory delivery without human intervention.

[0041] This artificial intelligence system is trained with the following deep learning-based structure and possesses precise judgment capabilities applicable to actual processes.

[0042] The training dataset includes thousands of architectural floor plans or 3D BIM drawings, along with corresponding actual shop drawings and bar list data. This is built based on drawing data produced at actual construction sites or design offices.

[0043] For object recognition, YOLOv8 or Vision Transformer (ViT) series algorithms are used to enable high-precision object detection, and for reinforcement judgment, a Graph Convolutional Network (GCN)-based graph neural network and a reinforcement learning algorithm (Policy Gradient) are combined to perform judgments that reflect the relationships between members and the load structure.

[0044] Drawing (output) establishes a consistent automated process up to drawing output by utilizing a Transformer-based vector generation algorithm.

[0045] The step (300) in which the main server (50) outputs a list of rebar from a shop drawing using artificial intelligence may be a step of outputting a list of rebar for factory automation work on a platform.

[0046] In this case, the step of outputting a rebar list from the shop drawing may be characterized by extracting rebar type, thickness, length, quantity, and location information listed in the shop drawing using OCR and object recognition algorithms, and automatically converting it into a standardized bar list form.

[0047] First, users can input shop drawing files into the system. Drawings are typically provided in formats such as PDF or DWG, and these files are processed into semantic images or vectors and converted into a structure that can be analyzed by artificial intelligence.

[0048] Subsequently, the latest object recognition algorithms, such as YOLOv8 or Vision Transformer, are applied to automatically detect and classify rebar-related elements (e.g., rebar symbols, line lengths, dimensions or symbols included in annotations, etc.) expressed on the drawing.

[0049] For detected rebar objects, the type of rebar (e.g., SD400, SD500), thickness (e.g., D10, D13, etc.), length (unit: mm), quantity (unit of count), and coordinate information of the placement location on the drawing (X, Y) are automatically extracted through OCR (Optical Character Recognition) and object tracking algorithms.

[0050] In this process, artificial intelligence improves the accuracy of symbol interpretation by comprehensively considering visual characteristics on the drawing (e.g., layer separation, line color, text spacing, line thickness, etc.).

[0051] These individual pieces of information regarding reinforcing bars are automatically converted into a "standardized bar list" in accordance with a defined data format (e.g., Excel, XML, JSON, etc.). For example, items such as reinforcing bar item number, material, thickness, length, quantity, bend shape, and placement location coordinates are organized into a list consisting of rows.

[0052] The finally generated bar list is exported to enable integration with MES systems, ERP, or PLC-based equipment control systems, and can be immediately put into production through linkage with factory automation equipment (cutters, benders, hoists, etc.).

[0053] The step (400) in which the equipment control terminal (20) receives a rebar list from the main server (50) may be a step in which a controller for equipment automation receives a rebar list (bar list).

[0054] The step (500) in which the equipment control terminal (20) outputs an equipment control command from a rebar list using artificial intelligence may be a step of deriving an optimal work order through a reinforcement learning-based scheduling algorithm by considering the work speed, available time, priority, etc. of multiple equipment (30), and automatically generating a processing command for each of the multiple equipment (30) according to the result.

[0055] First, the equipment control terminal (20) receives a "Bar List" transmitted from the main server (50). This list includes the type, length, thickness, processing shape (whether it is cut / bent, etc.), quantity, etc. of each bar.

[0056] Additionally, the equipment control terminal (20) collects operational data in real time, such as the current availability status, work speed, remaining work time, past work history, failure status, and energy consumption of multiple pieces of equipment (30; e.g., cutters, benders, hoists, etc.) within the factory. This information is provided through a sensor network, MES integration, or feedback received from a direct controller.

[0057] Based on this information, the equipment control terminal (20) establishes an optimal work distribution plan using a reinforcement learning-based scheduling artificial intelligence installed in the equipment control terminal (20).

[0058] This artificial intelligence is based on the Markov Decision Process (MDP), where "state" refers to the current equipment status and the rebar list, "action" refers to which equipment to assign each rebar item and in what order, and "reward" can be calculated based on total work time, energy efficiency, work balance between equipment, equipment priority, reduction of waiting time, etc.

[0059] [State Formula 1]

[0060]

[0061] [Action Formula 2]

[0062]

[0063] [Compensation Formula 3]

[0064]

[0065] [Policy Formula 4]

[0066]

[0068] Artificial intelligence learns work allocation policies that maximize rewards by repeating numerous trial-and-error processes, and through this, derives the most efficient work sequence and allocation plan for each facility.

[0069] For example, if a specific cutter is highly efficient at cutting long rebar while another is suitable for short, repetitive cuts, the AI ​​analyzes the length and shape of each rebar item and distributes it to the appropriate equipment. At the same time, it adjusts the work sequence to minimize work delays by considering the remaining operating time of the equipment, recent failure frequency, and work peak times.

[0070] Once scheduling is complete, the artificial intelligence automatically generates a customized set of control commands for multiple pieces of equipment (30). These commands include the type, length, shape, and quantity of rebar to be processed, the working mode of the cutter or bender (cutting length, angle, number of repetitions, etc.), priority, start / end time, waiting time, and whether simultaneous processing is possible. These commands are transmitted to the equipment in real time via an MES or PLC system, and the operator can proceed with the rebar processing work without separate manual input.

[0071] In this case, the artificial intelligence algorithm structure can use reinforcement learning series such as DQN, PPO (Proximal Policy Optimization), or A3C, and a Custom GNN Scheduler + RL Hybrid model is also possible (when processing equipment-rebar connection graphs).

[0072] In addition, the state encoding of the artificial intelligence can be integrated into a Joint Embedding by double-encoding the equipment state and rebar items, and the state history can be inferred using LSTM or GCN.

[0073] In this case, the learning objective of the artificial intelligence can be reduced to optimizing the reward function and verifying convergence based on hundreds of thousands of episodes.

[0074] In particular, the artificial intelligence of the present invention is not limited to static scheduling but includes a function to receive real-time sensor feedback and "dynamically reschedule" the work sequence in order to respond to exceptional situations that may occur during processing, such as equipment errors, work delays, and material deviations.

[0075] For example, if cutter A completes a task faster than expected or a delay occurs at bender B, the AI ​​analyzes the current work queue and equipment availability in real time to immediately readjust the next order. This maximizes equipment utilization and minimizes bottlenecks in the entire process.

[0076] In summary, the artificial intelligence installed in the equipment control terminal (20) of the present invention establishes an optimal work distribution plan for multiple pieces of equipment using a reinforcement learning method based on a rebar list and equipment status data, and generates and transmits automatic control commands for the processing equipment based on the results, thereby dramatically improving the efficiency, accuracy, and flexibility of the rebar processing process. This can function as a core technology for realizing smart construction and intelligent factories, going beyond simple rebar processing automation.

[0077] The step (600) in which a plurality of equipment (30) operates according to an equipment control command of an equipment control terminal (20) may be a step of receiving feedback information such as working distance, temperature, and cutting pressure from sensors attached to the plurality of equipment (30), and a time-series based LSTM artificial intelligence model analyzing the information to correct real-time control errors.

[0078] In this case, the following sensors are installed in each facility, and artificial intelligence at regular time intervals ( The following feedback information can be collected for every t.

[0079] [Feedback Information]

[0080]

[0081] In addition, error prediction and correction based on LSTM can be performed, and the collected time series data is input into a Long Short-Term Memory (LSTM) network to predict the feedback value at the next time point.

[0082] [Feedback Value Formula 4]

[0083]

[0084] In this case, the error vector can be defined by the following formula.

[0085] [Error Vector Formula 5]

[0086]

[0087] In addition, error correction can be achieved by adjusting specific parameters of the equipment control command (e.g., processing speed, cutting pressure) as follows.

[0088] [Parameter Formula 6]

[0089]

[0090] For example, if a command is issued to cut rebar to 1800mm from a specific cutter, but the actual working distance measured by the feedback sensor is received as 1770mm and the LSTM model predicts 1772mm, it can be corrected as follows.

[0091] [Example Formula 7]

[0092]

[0093] This correction can significantly contribute to reducing repetitive errors in the actual cutting length and realizing high-precision machining.

[0095] The method (1000) of the present invention described above may be implemented as a program (or application) to be executed in combination with a server, which is hardware, and stored on a medium.

[0096] The aforementioned program may include code encoded in a computer language such as C, C++, JAVA, or machine language, which can be read by the computer's processor (CPU) through the computer's device interface, in order for the computer to read the program and execute the methods implemented in the program. Such code may include functional code related to functions that define the necessary functions for executing the methods, and may include control code related to execution procedures necessary for the computer's processor to execute the functions according to a predetermined procedure. Additionally, such code may further include memory reference code regarding where (address) additional information or media necessary for the computer's processor to execute the functions should be referenced in the computer's internal or external memory. In addition, if the processor of the computer needs to communicate with any other computer or server located remotely in order to execute the above functions, the code may further include communication-related code regarding how to communicate with any other computer or server located remotely using the communication module of the computer, and what information or media to transmit or receive during communication.

[0097] The above-mentioned storage medium refers to a medium that stores data semi-permanently and is readable by a device, rather than a medium that stores data for a short period of time, such as a register, cache, or memory. Specifically, examples of the above-mentioned storage medium include, but are not limited to, ROM, RAM, CD-ROM, magnetic tape, floppy disk, and optical data storage device. That is, the above-mentioned program may be stored on various recording media on various servers that the computer can access, or on various recording media on the user's computer. Additionally, the above-mentioned medium may be distributed across networked computer systems, and computer-readable code may be stored in a distributed manner.

[0098] The steps of the method or algorithm described in connection with embodiments of the present invention may be implemented directly in hardware, implemented as a software module executed by hardware, or implemented by a combination thereof. The software module may reside in RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), Flash Memory, a hard disk, a removable disk, a CD-ROM, or any form of computer-readable recording medium well known in the art to which the present invention belongs.

[0099] Although embodiments of the present invention have been described above with reference to the attached drawings, those skilled in the art will understand that the present invention may be implemented in other specific forms without altering its technical concept or essential features. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive.

Claims

Claim 1 A step in which a client terminal inputs architectural drawings into a main server; a step in which the main server converts the architectural drawings into shop drawings using artificial intelligence; a step in which the main server outputs a rebar list from the shop drawings using artificial intelligence; a step in which an equipment control terminal receives the rebar list from the main server; a step in which the equipment control terminal outputs equipment control commands from the rebar list using artificial intelligence; The method includes a step in which a plurality of facilities operate according to facility control commands of the facility control terminal, and the step of converting the architectural drawing into a shop drawing is performed by the object recognition module of the main server object-recognizing structural members included in the architectural drawing using a semantic segmentation method to generate vector data of the structural members including location, direction, length, width, and material information of the structural members; then the rebar placement judgment module of the main server learns rebar placement rules according to the structural role of the structural members based on the vector data of the structural members to derive a rebar placement judgment result; and the shape generation module of the main server outputs rebar shape data, in which the shape, length, thickness, location, and quantity of the rebar are coded, as a shop drawing in vector format based on the rebar placement judgment result; the rebar list includes item information regarding the type, length, thickness, processing shape, and quantity of the rebar; and the artificial intelligence used by the facility control terminal is in state S at time t according to a policy formula. t Action A of allocating specific reinforcing bars to specific equipment according to t Determine and,[policy formula] State S in the policy formula t is determined by the state formula below, and compensation R t is determined by the compensation formula below, where γ represents the discount rate, [State Formula] M in the state formula t is a state vector of multiple facilities regarding remaining work time, operation status, temperature, and energy consumption, and R t is the item information for the pending rebar, [Compensation Formula] T in the compensation formula makespan is the total task completion time, and E total is the total energy consumption, and W imbalance A method for controlling equipment, wherein the work load imbalance between equipment is a work load imbalance, α, β, and γ are weighting coefficients, and the step of operating multiple pieces of equipment according to an equipment control command of the equipment control terminal comprises receiving feedback information regarding working distance, temperature, and cutting pressure from sensors attached to the multiple pieces of equipment, an artificial intelligence mounted on the equipment control terminal collecting feedback information at regular time intervals, inputting the collected time series data into a Long Short-Term Memory (LSTM) network to predict the feedback value at the next time point, and correcting the error by adjusting parameters regarding the processing speed and cutting pressure of the equipment control command based on the error regarding the feedback information.

Images