An IT equipment infrastructure automated design, construction, and related solution delivery method, apparatus, and system that considers security and resource efficiency factors in industry characteristics

Abstract

The application provides an automatic solution method for designing and constructing IT equipment infrastructure according to industry characteristics considering safety and resource efficiency. The equipment first receives a first factory plan of a user terminal, and if it is confirmed that the factory produces a first product, the equipment required for production and the IT equipment required for operation are classified as necessary production equipment and necessary IT equipment respectively. The equipment matches the first plan with the necessary production equipment, encodes to generate a first input signal, inputs a first trained AI model, obtains a first output signal indicating the placement position of the equipment, marks to create a second plan. Then match the second plan with the necessary IT equipment, encode to generate a second input signal, input a second AI model, get a second output signal, based on this, display the necessary IT equipment on the second plan to create a third plan, and transmit it to the user terminal.

Description

Technical Field

[0001] The technologies involved in the following embodiments are designed to fully consider security and resource efficiency, and to tailor automated solutions for the design and construction of IT equipment infrastructure for various industries. Background Technology

[0002] When building a factory based on the characteristics of the industry, it is crucial to conduct structural design in advance and rationally plan the placement and location of the equipment required for factory operation.

[0003] Currently, structural design is often implemented solely by reviewing factory floor plans before actual equipment installation. This approach is not only inefficient and lacks accuracy, but also makes it extremely difficult to select the optimal placement location for equipment within a large factory space.

[0004] As the requirements for the rationality of equipment layout in factory construction and operation continue to increase, the market demand for technologies that can provide the best spatial location information for equipment is becoming more and more urgent, thus requiring in-depth research on related technologies.

[0005] Previous documents

[0006] [Patent Documents]

[0007] (Patent Document 1) Korean Patent Registration No. 10-2739197

[0008] (Patent Document 2) Korean Patent Registration No. 10-2524126

[0009] (Patent Document 3) Korean Patent Registration No. 10-2398562

[0010] (Patent Document 4) Korean Patent Registration No. 10-2340834 Summary of the Invention

[0011] One embodiment of the present invention aims to provide a method, apparatus and system that can fully balance security and resource efficiency, and tailor-make automated solutions for the design and construction of IT equipment infrastructure in various industries.

[0012] Of course, the objectives of this invention are not limited to the aspects mentioned above, and other objectives not elaborated in detail can be understood from the following specific description.

[0013] The present invention solves the technical problem by adopting the following technical solution:

[0014] According to one embodiment, in a method for providing automated solutions for designing and building IT equipment infrastructure based on industry characteristics, considering security and resource efficiency, the process is executed by the equipment. First, a first floor plan, i.e., a floor plan of a first factory, is received from a user terminal. If it is confirmed that the first factory is a factory producing a first product, the production equipment required to produce the first product should be classified as necessary production equipment, and the IT equipment required to operate the first factory should be classified as necessary IT equipment. The first floor plan and the required production equipment are matched to generate a primary matching result. The first matching result is encoded to generate a first input signal. The first input signal is input into a first AI model, which has been trained and can selectively... The optimal location for placing production equipment on a floor plan is determined. When the location for the necessary production equipment on the first floor plan is selected via a first input signal, a first output signal indicating the required location is obtained from a first artificial intelligence model. Based on the first output signal, the required production equipment is marked on the first floor plan, generating a second floor plan. The second floor plan is matched with the required IT equipment to generate a second matching result. The second matching result is encoded to generate a second input signal. The second input signal is input into a trained second artificial intelligence model to select the optimal location for placing the IT equipment on the floor plan. The second input signal is then placed on the second floor plan. Once a location is selected, a second output signal is obtained from the second artificial intelligence model, indicating the required location for the IT equipment. Based on the second output signal, the required IT equipment is marked on the second floor plan, creating a third floor plan. Including the step of transmitting the third-floor floor plan to a user terminal, this provides a method for designing and building automated solutions for IT equipment infrastructure in various industries, while considering security and resource efficiency.

[0015] The first AI model selects the location of production equipment to minimize the length of the production line connecting to the equipment in the order of the production process. This earns the first reward. A second reward is given for placing the equipment closer to the entrance as the production process moves faster, and a third reward is given for placing it closer to the exit as the production process moves slower. Its characteristics can be represented by an AI model trained using reinforcement learning. A fourth reward is given if the equipment is placed in a larger area as its footprint increases, and a fifth reward is given if it is placed in a smaller area due to its smaller footprint.

[0016] The second AI model awards a sixth reward if the IT device is placed on communication cabling, a seventh reward for placing it in each area, and an eighth reward for placing it near production equipment. The higher the security level of the IT device, the farther it is from entrances and exits. If the location of the IT device results in a higher ninth reward, and the lower the security level of the IT device, the closer it is to entrances and exits, this can be described as an AI model trained through reinforcement learning, awarding many up to ten rewards.

[0017] The present invention has the following beneficial effects:

[0018] According to one embodiment, by selecting the optimal location for production equipment and IT equipment while taking into account security and resource efficiency, the effect is to provide an optimized deployment status of equipment while increasing user convenience through the automation of IT equipment infrastructure design and construction.

[0019] On the other hand, the effects of the embodiments are not limited to those described above, and other unmentioned effects can be clearly understood by those skilled in the art from the following description. Attached Figure Description

[0020] Figure 1 A diagram providing an overview of the configuration of a diaphragm-based system;

[0021] Figure 2 This is a flowchart illustrating the process of providing automated design and construction solutions for IT equipment infrastructure across various industries, taking into account both security and resource efficiency.

[0022] Figure 3 This is a flowchart illustrating the process of color-coding areas according to their congestion levels, based on a standard implementation method.

[0023] Figure 4 This is a flowchart illustrating the process of marking region boundary lines according to a single embodiment;

[0024] Figure 5 This is a preliminary schematic diagram of the device configuration based on an embodiment. Detailed Implementation

[0025] The embodiments are described in detail below with reference to the accompanying drawings. However, various modifications can be made to the embodiments, and therefore the scope of the patent application is not limited to or restricted by these embodiments. Any changes, equivalents, or substitutions to the embodiments should be understood to be included within the scope of the claims.

[0026] The specific structural or functional descriptions of the embodiments are provided for illustrative purposes only and may be modified and implemented in various forms. Therefore, the embodiments are not limited to a particular form of disclosure, and the scope of this specification includes changes, uniformities, or substitutions incorporated into the descriptive concepts.

[0027] Terms such as "first" or "second" can be used to describe various components, but the interpretation of these terms should only be used to distinguish one component from another. For example, the first component can be named the second component, and similarly, the second component can be named the first component.

[0028] When a component is said to be "connected" to another component, it should be understood that it may be directly connected to or connected to another component, but there may be another component between them.

[0029] The terminology used in the embodiments is for illustrative purposes only and should not be construed as restrictive. Singular expressions include plural expressions unless the context clearly implies otherwise. In this specification, the terms "comprising" or "having" should be understood to mean the presence of the functions, numbers, steps, actions, components, parts, or combinations thereof described herein, and should not exclude the presence or addition of one or more other functions or numbers, steps, actions, components, parts, or combinations thereof.

[0030] Unless otherwise defined, all terms used herein, including technical or scientific terms, shall have the same meaning as commonly understood by one of ordinary skill in the art to which the embodiments pertain. Terms such as those defined in common dictionaries shall be interpreted as having the meaning consistent with their meaning in the relevant descriptive context and shall not be interpreted in an idealistic or overly formal sense unless expressly defined in this application.

[0031] Furthermore, when describing the accompanying drawings, regardless of the drawing code, the same reference numerals should be assigned to the same elements, and identical repetitive descriptions should be omitted. When describing embodiments, detailed descriptions should be omitted if it is determined that a specific description of the relevant technical notifications may unnecessarily obscure the essential points of the embodiment.

[0032] The embodiments can be implemented in various types of products, including personal computers, laptops, tablets, smartphones, televisions, smart home appliances, smart cars, kiosks, and wearable devices.

[0033] In this embodiment, the artificial intelligence (AI) system is a computer system that achieves human-level intelligence. Unlike existing rule-based intelligent systems, it is a system in which machines learn and make judgments independently. As AI systems improve their recognition rates and more accurately understand sellers' preferences, existing rule-based intelligent systems are gradually being replaced by deep learning-based AI systems.

[0034] Artificial intelligence technology consists of machine learning and element technologies that use machine learning. Machine learning is an algorithmic technique that classifies / learns the features of input data on its own. Element technologies are techniques that use machine learning algorithms such as deep learning to simulate the cognitive and judgmental functions of the human brain, and consist of technical fields such as language understanding, visual understanding, reasoning / prediction, knowledge representation, and motion control.

[0035] The various fields where artificial intelligence technology is applied are as follows: Language understanding is a technology for recognizing, adapting to, and processing human language / text, including natural language processing, machine translation, dialogue systems, question answering, and speech recognition / synthesis. Visual understanding is a technology for recognizing and processing objects like human vision, including object recognition, object tracking, image search, human recognition, scene understanding, spatial understanding, and image improvement. Reasoning and prediction is a technology for making logical inferences and predictions by judging information, including knowledge / probability-based reasoning, optimization prediction, preference-based planning, and recommendation. Knowledge representation is a technology for automatically processing human experience information into knowledge data, including knowledge construction (data generation / classification) and knowledge management (data utilization). Motion control is a technology for controlling the autonomous driving of vehicles and the movement of robots, including motion control (navigation, collision, driving) and action control (behavioral control).

[0036] Generally, to apply machine learning algorithms to real-world applications, training is conducted through trial and error due to the nature of basic machine learning methods. Deep learning, in particular, requires hundreds of thousands of iterations. Since this is impossible to achieve in a real physical environment, the actual physical environment is virtualized on a computer, and learning is conducted through simulation.

[0037] Figure 1 It is a diagram that outlines the system configuration based on the diaphragm.

[0038] Reference Figure 1 According to one embodiment, the system may include a user terminal 100 and a device 200.

[0039] First, user terminal 100 and device 200 can be connected through a communication network, and the communication network can be configured without considering wired or wireless communication methods. Communication between servers and between servers and terminals can be implemented in various forms.

[0040] User terminal 100 can be implemented as a computing device with communication functions, such as a mobile phone, desktop computer, laptop computer, tablet computer, smartphone, etc., but not limited to, and can be implemented as various types of communication devices that can connect to external servers.

[0041] User terminal 100 can be configured to perform all or part of the computing, storage / reference, input / output, and control functions of a conventional computer, and can be configured to communicate with device 200 via wired or wireless communication.

[0042] User terminal 100 can access web pages created by individuals or groups providing services through device 200, or can install applications developed and distributed by individuals or groups providing services through device 200. For this purpose, user terminal 100 can link to device 200 via web pages or applications.

[0043] User terminal 100 can access device 200 through a webpage or application provided by device 200.

[0044] Device 200 may be its own server, owned by an individual or organization using services provided by device 200, or it may be a cloud server, or it may be a peer-to-peer (P2P) collection of distributed nodes. Device 200 may be configured to perform all or part of the computing, storage / reference, input / output, and control functions of a conventional computer. Device 200 may be equipped with at least one artificial intelligence model that performs inference functions.

[0045] The device 200 can be configured to communicate with the user terminal 100 via wired or wireless communication, and can control the operation of the user terminal 100 and control what information is displayed on the screen of the user terminal 100.

[0046] Device 200 is implemented as a server, providing automated solutions for the design and construction of IT equipment infrastructure based on industry characteristics, taking into account security and resource efficiency, and can also provide a platform for related services.

[0047] On the other hand, for ease of explanation, Figure 1 Only user terminal 100 is shown, but the number of terminals may vary depending on the embodiment. There is no particular limitation on the number of terminals as long as the processing power of device 200 allows it.

[0048] According to one embodiment, device 200 can select the optimal location on the floor plan based on artificial intelligence, as well as the optimal location for placing IT equipment on the floor plan, which will be described in detail later. Figure 2 Please provide an explanation.

[0049] In this invention, artificial intelligence (AI) refers to the technology of mimicking human learning, reasoning, and perception abilities and implementing them in computers, and may include concepts such as machine learning and symbolic logic. Machine learning (ML) is an algorithmic technique that can classify or learn features of input data on its own. Artificial intelligence technology is a machine learning algorithm that analyzes input data, learns from the analysis results, and can make judgments or predictions based on the learning results. Furthermore, the technology of using machine learning algorithms to mimic the cognitive and judgmental functions of the human brain can also be understood as falling within the scope of artificial intelligence. For example, it may include technical fields such as language understanding, visual understanding, reasoning / prediction, knowledge representation, and motion control.

[0050] Machine learning can refer to the process of training neural network models using experience in processing data. Machine learning can mean that computer software improves its ability to process data on its own. Neural network models are built by modeling the correlations between data, which can be represented by multiple parameters. A neural network model extracts features from given data, analyzes it, and derives the correlations between the data. Machine learning can be said to repeat this process to optimize the parameters of the neural network model. For example, a neural network model can learn the mapping (correlation) between the inputs and outputs of data given in the form of input / output pairs. Alternatively, a neural network model can deduce regularities between given datasets and learn relationships, even if only the input data is given.

[0051] Artificial intelligence learning models, or neural network models, can be designed to replicate the structure of the human brain on a computer and can include multiple network nodes that simulate and weight neurons in a human neural network. These multiple network nodes can simulate the synaptic activity of neurons sending and receiving signals through synapses and are interconnected. In an AI learning model, multiple network nodes can reside in layers of different depths and send and receive data based on their convolutional connections. For example, an AI learning model can be an artificial neural network, a convolutional neural network (CNN), etc. As an example, the AI ​​learning model can perform machine learning using methods such as supervised learning, unsupervised learning, and reinforcement learning. Machine learning algorithms used to perform machine learning can include decision trees, Bayesian networks, support vector machines, artificial neural networks, Ada-boost, perceptrons, genetic programming, and clustering.

[0052] A CNN (Neural Network Array) is a multilayer perceptron designed to use minimal preprocessing. A CNN consists of one or more convolutional layers and typical neural network layers above them, with additional weights and pooling layers. Due to this structure, CNNs can fully utilize input data from two-dimensional structures. Compared to other deep learning architectures, CNNs perform well in both video and audio domains. CNNs can also be trained using standard backpropagation. CNNs are easier to train than other feedforward neural network techniques and have the advantage of using fewer parameters.

[0053] Convolutional networks are neural networks consisting of a set of nodes with bound parameters. Increased availability of training data and computational power, coupled with advancements in algorithms such as discriminative linear units and randomized deactivation training, have significantly improved many computer vision tasks. In large datasets (such as those available for many tasks today), overfitting is not critical, and increasing network size can improve test accuracy. Optimal use of computational resources is a limiting factor. To address this, decentralized, scalable implementations of deep neural networks can be used.

[0054] Figure 2 It is a flowchart illustrating the process of providing automated solutions for the design and construction of IT equipment infrastructure for each industry, taking into account security and resource efficiency, according to one embodiment.

[0055] Reference Figure 2 First, in step S201, device 200 can receive a first floor plan, i.e., a floor plan of the first factory, from user terminal 100. Here, the first floor plan is a drawing showing the design of the first factory's functions, structure, scale, layout, etc., and may include information about the products produced by the first factory, information about the production equipment installed in the first factory, information about the IT equipment installed in the first factory, etc.

[0056] In step S202, if it is determined that the first factory is a factory that produces the first product, then equipment 200 can classify the production equipment required to produce the first product as necessary production equipment and the IT equipment required to operate the first factory as necessary IT equipment. Here, production equipment can refer to industrial machinery and equipment required to produce the product, and IT equipment can refer to communication and electronic equipment required to operate the factory.

[0057] Specifically, by examining the products produced by the first factory according to the first floor plan, it can be determined that equipment 200 is the factory producing the first product, and it can be confirmed that the first factory is the factory producing the first product. Among numerous pre-registered production equipment, the production equipment identified as essential for producing the first product can be classified as necessary production equipment, and among numerous pre-registered IT equipment, the IT equipment identified as essential for the operation of the first factory can be classified as necessary IT equipment. Therefore, information on individual production equipment and individual IT equipment can be stored in a database. Necessary production equipment can be classified using the comparison results between the first floor plan and the information of each production equipment, and necessary IT equipment can be classified using the comparison results between the first floor plan and the information of each IT equipment. At this point, one or more production equipment may be classified as necessary production equipment, and one or more IT equipment may be classified as necessary IT equipment.

[0058] For example, if an analysis based on a first floor plan and first production equipment information determines whether the first production equipment is necessary to produce the first product, and thus determines that the first production equipment is essential for producing the first product, and if the first production equipment is not required for producing the first product, then equipment 200 can classify the first production equipment as necessary production equipment. For this purpose, the production equipment information may include information indicating which categories of goods need to be produced.

[0059] Furthermore, if, based on the first floor plan and the first IT equipment information, it is determined that the first IT equipment is necessary for operating the first factory, and if it is determined that the first IT equipment is not necessary for operating the first factory, then equipment 200 can classify the first IT equipment as necessary IT equipment. For this purpose, the IT equipment information may include information indicating which type of factory needs to be operated.

[0060] In step S203, the device 200 can match the first plan view with the required production equipment to generate a first matching result.

[0061] In step S204, device 200 may encode the first matching result to generate a first input signal.

[0062] Specifically, the device 200 can generate a first input signal by performing normal information processing on the first matching result, so as to use the first matching result as input to the first artificial intelligence model.

[0063] In step S205, device 200 can input a first input signal to a first AI model, which has been trained to select the optimal location for the production equipment to be placed on the floor plan. Here, the first AI model may have already been trained to select the optimal location for the production equipment on the floor plan.

[0064] According to one embodiment, the first AI model may encode the matching result of the floor plan and the production equipment, receive the input signal, select the best position for the production equipment on the floor plan based on the input signal, and output an output signal indicating the selected position.

[0065] In other words, the first AI model can consider the floor plan and production equipment, analyze where the production equipment should be placed on the floor plan, select the placement location of the production equipment based on the analysis results, and output an output signal indicating the selected location.

[0066] In step S206, once the location of the required production equipment is placed on the first plan view via the first input signal, the device 200 can obtain a first output signal from the first AI model indicating where the required production equipment will be placed.

[0067] For example, if the first production equipment is classified as a necessary production equipment, then device 200 inputs a first input signal to the first artificial intelligence model. If the location of the first production equipment is selected as the coordinates (5,13) on the first plan view using the first input signal, then a first output signal representing the coordinates (5,13) can be obtained from the first artificial intelligence model. In this case, if there are multiple necessary production equipment, the first output signal may contain coordinate information of the location to be placed on the first plan view of each production machine.

[0068] In other words, the first AI model can examine the floor plan and production equipment using input signals, analyze the correct placement of the production equipment on the floor plan, select the appropriate placement based on the analysis results, and output a signal indicating the selected location. To this end, the first AI model can be pre-trained to select the optimal location for placing the production equipment on the floor plan based on information stored in a database.

[0069] The learning device for training the first AI model can be the same device 200 that selects the production equipment location using the learned first AI model, or it can be a separate device. The process of training the first AI model is described below.

[0070] First, the learning device can generate input based on the matching results of the matching plan and the production equipment.

[0071] Specifically, the learning device can perform a preprocessing process on the matching results by matching floor plans with production equipment. The preprocessed matching results can be used as input to the first AI model, or they can be generated as input through normal processing to remove unnecessary information.

[0072] Next, the learning device can apply the input to the first AI model. This first AI model may be an artificial neural network trained using reinforcement learning. It can be a Q-network, deep Q-network (DQN), or relational network (RN) structure, suitable for outputting abstract reasoning through reinforcement learning.

[0073] The first AI model was trained using reinforcement learning and can be updated and optimized by reflecting evaluations in various rewards.

[0074] For example, the first reward could be the location of the production equipment so that the length of the production line connecting the equipment according to the production process sequence is minimized. The second reward could be the location of the production equipment so that the faster the production process sequence, the closer it is to the entrance. The third reward could be the location of the production equipment so that the slower the production process sequence, the closer it is to the exit. If the location of the production equipment is chosen such that placing it in a larger area increases the reward value, the fifth reward might be higher if the production equipment is placed in a smaller area because the area occupied by the production equipment is smaller.

[0075] Next, the learning device can acquire the output of the first AI model. In this case, the output of the first AI model might be information indicating the placement of production equipment on the floor plan. In other words, the first AI model can select the placement of production equipment on the floor plan by analyzing the matching results between the floor plan and the production equipment, and output information about the selected location. At this point, the first AI model can select the optimal location of the production equipment by considering it as the shortest possible production line, connecting it in the order of the production process. The faster the production process, the closer it is to the entrance; the slower the production process, the closer it is to the exit. The larger the area occupied by the production equipment, the smaller the area occupied, and the more of it can be placed in a smaller area. Information about the optimal location of the production equipment can be printed.

[0076] Next, the learning device can evaluate the output of the first AI model and pay a reward. At this point, the evaluation of the output can be divided into five rewards, from the first to the fifth, with different payments made based on the importance of the reward.

[0077] For example, if the learning device chooses a location for the production equipment such that the length of the production line connecting to the equipment according to the production process sequence is as short as possible, a first reward can be awarded. If the production equipment is located closer to the entrance, a second reward can be awarded. The faster the production process sequence, the closer to the exit. If you choose to place the production equipment in a larger area with a large number of third rewards, and the smaller the area occupied by the production equipment, the larger the fourth reward can be awarded. The smaller the area occupied by the production equipment, the more fifth rewards can be obtained.

[0078] Next, the learning device can update the first AI model based on the evaluation.

[0079] Specifically, the learning device can update the first AI model by placing the production equipment on a plan view in the environment of the first AI model, thereby optimizing the strategy for determining the actions to be taken in certain states, thereby maximizing the consensus expectation of the reward.

[0080] For example, if the learning device determines that there is no anomaly in the selection of the production equipment location as the first coordinate based on the first matching result, it generates a first learning signal indicating that the selection result of the first production equipment location is not abnormal. This first learning signal is then applied to the first artificial intelligence model, so that when the input value is similar to the first matching result, the production equipment location is selected to a position similar to the first coordinate. The first AI model can be updated through the process of training it.

[0081] On the other hand, the process of optimizing the policy can be accomplished by estimating the maximum value of the consensus expectation of the reward, or the maximum value of the Q-function, or the minimum value of the loss function of the Q-function. The minimum value of the loss function can be achieved using stochastic gradient descent (SGD). The process of optimizing the policy is not limited to these; various optimization algorithms used in reinforcement learning can be employed.

[0082] By repeating the learning process of the first AI model described above, the learning device can gradually update the first AI model. In this way, the learning device can train the first AI model to select and output the location of production equipment on a floor plan.

[0083] When selecting the location of production equipment on the floor plan, the learning device can train the first AI model by adjusting the analysis criteria to reflect reinforcement learning, through the first or fifth reward.

[0084] In other words, the first AI model is based on the sequence of the production process. If the location of the production equipment minimizes the length of the production line, a first reward is awarded. If the production equipment is located closer to the entrance, the production process is faster, and a second reward is awarded. If the production equipment is located closer to the exit, the production process is slower, and a third reward is awarded. If the production equipment is placed in a larger area, the larger the area it occupies, the greater the fourth reward. Conversely, if the production equipment occupies a smaller area and is placed in a smaller area, the fifth reward may be higher, possibly based on the AI ​​model trained using reinforcement learning.

[0085] In step S207, the device 200 can generate a second plan view, marking the required production equipment on the first plan view based on the first output signal.

[0086] For example, if it is classified as a first production device, the device 200 can mark the first production device in the portion corresponding to the coordinates (5, 13) on the first floor plan based on the first output signal. If the placement of the first production device is determined by the coordinates (5, 13) and in the same way as the first production device, you can create a second floor plan by marking each production device classified as a necessary production device on the first floor plan.

[0087] In step S208, device 200 can match the second floor plan with the required IT equipment to generate a second matching result.

[0088] In step S209, device 200 may encode the second matching result to generate a second input signal.

[0089] Specifically, the device 200 can generate a second input signal by performing normal information processing on the second matching result, so as to input the second matching result into the second artificial intelligence model.

[0090] In step S210, device 200 can input the second input signal into a trained second AI model to select the optimal location for placing the IT device on the floor plan. Here, the second AI model can be trained to select the optimal location for the IT device on the floor plan.

[0091] According to one embodiment, the second artificial intelligence model may encode the matching result of the floor plan and the IT equipment, receive input signals, select the best position of the IT equipment to be placed on the floor plan through the input signals, and output an output signal indicating the selected position.

[0092] In other words, the second artificial intelligence model can analyze where IT equipment should be placed on the floor plan, select the location of the IT equipment based on the analysis results, and output an output signal indicating the selected location.

[0093] In step S211, once the location of the required IT equipment is placed on the second floor plan via the second input signal, the device 200 can obtain a second output signal from the second AI model, indicating where the required IT equipment will be placed.

[0094] For example, if the first IT device is classified as a necessary IT device, device 200 inputs a second input signal to the second AI model. If the second input signal selects the position of the first IT device on the second plan view as the coordinates (5, 13), a second output signal representing the coordinates (5, 13) can be obtained from the second AI model. In this case, if there are multiple necessary IT devices, the second output signal may contain the coordinate information of the position to be configured on the second plan view for each IT device.

[0095] In other words, the second AI model can view the floor plan and IT equipment through input signals, analyze the appropriate placement of the IT equipment on the floor plan, select the location of the IT equipment based on the analysis results, and output a signal indicating the selected location. To this end, the second AI model can be pre-trained to select the optimal location for placing the IT equipment on the floor plan based on information stored in a database.

[0096] The learning device that learns the second AI model can be the same device 200 that uses the learned second AI model to select the location of the IT device, or it can be a separate device. The process of training the second AI model is described below.

[0097] First, the learning device can generate input based on the matching results of the matching floor plan and the IT equipment.

[0098] Specifically, the learning device can perform a preprocessing process on the matching results by matching floor plans and IT equipment. The preprocessed matching results can be used as input for a second AI model, or they can be used to generate input through normal processing that removes unnecessary information.

[0099] Next, the learning device can apply its input to a second AI model. This second AI model may be an artificial neural network trained using reinforcement learning. It can be a Q-network, DQN (Deep Q-network), or relational network (RN) structure, suitable for outputting abstract reasoning through reinforcement learning.

[0100] The second AI model is trained using reinforcement learning and can be updated and optimized by reflecting evaluations of various rewards.

[0101] For example, if the location of the IT equipment allows it to be placed on communication cabling, the sixth reward might increase the reward value; if the location allows the IT equipment to be placed in every area, the seventh reward might increase the reward value; if the IT equipment is placed adjacent to production equipment, the eighth reward might increase the reward value; the ninth reward is the location of the IT equipment, so the higher the security level of the IT equipment and the farther it is from the entrance and exit, the higher the reward value may be. If the IT equipment is located closer to the entrance and exit, the tenth reward may be higher because the security level of the IT equipment is lower.

[0102] Next, the learning device can acquire the output from the second AI model. At this point, the output of the second AI model might be information indicating the placement of the IT equipment on the floor plan. In other words, the second AI model can select the placement location of the IT equipment on the floor plan by analyzing the matching results between the floor plan and the IT equipment, and output information about the selected location. The second AI model then processes the IT equipment to be placed on the communication lines, processing it and placing it in each area accordingly. For example, it processes IT equipment placed near production equipment; for IT equipment with higher security levels, it places it closer to entrances and exits; and for IT equipment with lower security levels, it places it closer to entrances and exits. This process allows it to select the optimal location for the IT equipment and output information about the optimal location.

[0103] Next, the learning device can evaluate the output of the second AI model and reward it. At this point, the evaluation of the output can be divided into 6th to 10th rewards, which may be paid differently depending on their importance.

[0104] For example, if you choose the location of the IT equipment to place it on the communication cabling, you can receive a large reward 6. If you choose the location of the IT equipment to place it in each area, you can receive a large reward 7. If you choose the location of the IT equipment to place it near the production equipment, you can receive a large reward 8. If you choose the location of the IT equipment, the higher the security level of the IT equipment and the farther it is from the entrance and exit, the more rewards you can receive. The lower the security level of the IT equipment and the closer it is to the entrance and exit, the more rewards you can get in Reward 10, depending on its location.

[0105] Next, the learning device can update the second AI model based on the evaluation.

[0106] Specifically, the learning device can update the second AI model in an environment where the second AI model selects the location of IT equipment on a floor plan, thereby maximizing the consensus expectation of the reward by optimizing the process of determining the strategy to take action in certain states.

[0107] For example, if the learning device determines that there is no anomaly in selecting the location of the IT device as the second coordinate based on the second matching result, it generates a second learning signal indicating that there is no anomaly in selecting the location of the first IT device. This second learning signal is then applied to the second artificial intelligence model, so that when the input value is similar to the second matching result, the location of the IT device is selected as a location similar to the second coordinate. The second AI model can be updated through the process of training it.

[0108] On the other hand, the process of optimizing the policy can be accomplished by estimating the maximum value of the consensus expectation of the reward, or the maximum value of the Q-function, or the minimum value of the loss function of the Q-function. The minimum value of the loss function can be achieved using stochastic gradient descent (SGD). The process of optimizing the policy is not limited to these; various optimization algorithms used in reinforcement learning can be employed.

[0109] The learning device can gradually update the second AI model by repeating the learning process described above. In this way, the learning device can train a second AI model that selects the location of IT equipment on a floor plan and outputs the result.

[0110] When selecting the location of IT equipment on a floor plan, the learning device can train a second AI model by adjusting the analysis criteria to reflect reinforcement learning until the 6th or 10th reward.

[0111] In other words, the second AI model works as follows: If you choose the location of the IT equipment so that it is placed on the communication cabling, you will receive the 6th reward. If the IT equipment is placed in each area, you will receive the 7th reward. If the IT equipment is placed near the production equipment, you will receive the 8th reward. The higher the security level of the IT equipment, the farther it will be placed from the entrance and exit. If the location of the IT equipment is closer to the entrance and exit, instead of the 9th reward of higher security level, and the IT equipment has a lower security level, it may be rewarded with many 10 rewards based on the AI ​​model trained by reinforcement learning.

[0112] In step S212, device 200 can generate a third plan view based on the second output signal by displaying the required IT equipment on the second plan view.

[0113] For example, if it is classified as the first IT device, and if the location of the first IT device is determined based on the second output signal as coordinates (5,13), then device 200 can display the first IT device in the part corresponding to the coordinates (5,13) on the second floor plan, and display each IT device classified as necessary IT device on the second floor plan in the same way as the first IT device. You can create a third floor plan.

[0114] In step S213, device 200 can send the third floor plan to user terminal 100.

[0115] In other words, device 200 can take security and resource efficiency into account, design the infrastructure of IT equipment in each industry, create a third floor plan based on the first floor plan received from user terminal 100, and provide the third floor plan to user terminal 100.

[0116] Figure 3 This is a flowchart illustrating, according to a routine implementation, the process of color-coding an area based on its congestion level.

[0117] According to one embodiment, Figure 3 Each step shown can be performed between steps S212 and S213.

[0118] refer to Figure 3 First, in step S301, device 200 can create a first virtual space, representing the first factory as a digital twin, based on the third floor plan, by placing the required production equipment and necessary IT equipment in a virtual space corresponding to the physical space of the first factory. Here, the first virtual space is a virtual model implemented in the network in the same way as the first factory and can simulate the operation of the first factory. At this time, device 200 can receive further information such as 3D images, layouts, and drawings of the first factory from user terminal 100, and create the first virtual space based on the received information and the third floor plan.

[0119] In step S302, device 200 simulates the operation of factory 1 implemented in virtual space 1, confirming that personnel passing through area 1 during the first time period can be identified as passers-throughs of area 1, and personnel confirmed to have stayed in area 1 for more than 1 hour during the first time period can be classified as workers in area 1. Here, area 1 is any one of multiple areas; therefore, factory 1 can be divided into multiple areas. Furthermore, the first period can be set differently depending on the embodiment, for example, within the most recent 24 hours. Additionally, the first hour can be set differently depending on the embodiment, for example, 30 minutes.

[0120] In other words, device 200 can simulate the operation of a first factory in a first virtual space, and can simulate the production of a first product through the operation of the first factory in the first virtual space. As a result of simulating the operation of the first factory, all people confirmed to have passed through the first area during the first period can be classified as passers-throughs of the first area, and all people confirmed to have stayed in the first area for more than the first hour during the first period can be classified as workers of the first area.

[0121] For the simulation of the first factory's operation, additional production-related conditions such as personnel information, equipment information, and material information can be input through the user terminal 100. Based on the information input into the user terminal 100, the equipment 200 can set the working status of the personnel working in the factory, the operating status of the production equipment, and the input and working status of the materials put into production. After setting the operating status and input status respectively, the simulation can be carried out so that the first factory can be run and the first product can be produced in the first virtual space.

[0122] In step S303, the device 200 can identify the number of people identified as passers-by in area 1 as the number of people classified as people in area 1, and the number of people classified as people in the first area as the number of people classified as people in the first area.

[0123] In step S304, the device 200 can set the number of digits 1 to a higher value within the first reference range as the number of digits 1 increases. Here, the first reference range can be set differently depending on the embodiment, for example, within the range of 0 to 100.

[0124] For example, if the first number is identified as 1, the device 200 can set the first value to 10, and if the second number is identified as 2, the first value can be set to 20.

[0125] In step S305, the device 200 can set the second value to a higher value within the second reference range as the number of second people increases. Here, the second reference range can be set to a wider range than the first reference range; for example, it can be set to a range from 0 to 200.

[0126] For example, if the second number is identified as 1, the device 200 can set the second value to 20, and if the second number is identified as 2, the second value can be set to 40.

[0127] In step S306, device 200 is able to calculate the congestion of the first region by the sum of the first and second values.

[0128] The device 200 can calculate the congestion of each of the multiple regions in the same way as it calculates the congestion of the first part.

[0129] In step S307, if congestion for each of the multiple regions has been calculated, the device 200 may calculate the average of the congestion levels calculated for each of the multiple regions as a first average.

[0130] In step S308, device 200 can generate a first differential value because the congestion in the first region is subtracted from the first average value.

[0131] In step S309, device 200 can determine whether the first difference is higher than the first reference value. Here, the first reference value can be set differently depending on the embodiment, for example, 10.

[0132] If, in step S309, the first difference is found to be higher than the first reference value, in step S310, the device 200 can be configured to display the first area on the third floor plan in a first color to indicate congestion. At this time, a first color can be preset to indicate the congestion situation; for example, it can be set to red.

[0133] In other words, if device 200 determines that the first difference is higher than the first reference value, it can process the third plan view so that the background color of the first area is displayed as the first color.

[0134] If it is determined that the first difference in step S309 is not higher than the first reference value, then in step S311, the device 200 can check whether the first difference is higher than the second reference value. Here, the second reference value can be set to a value lower than the first reference value, for example, it can be set to -10.

[0135] If the first difference in step S311 is found to be higher than the reference value in the second reference value, then in step S312, the device 200 can be configured to display the first area on the third floor plan in a second color representing the normal state. At this time, the second color can be preset to represent the normal state; for example, it can be set to yellow.

[0136] In other words, if device 200 determines that the first difference is not higher than the first reference value but is higher than the second reference value, it can process the third plan view and display the background color of the first area as the second color.

[0137] If it is determined that the first difference in step S311 is not higher than the second reference value, then in step S313, the device 200 can be configured to display the first area representing the gap in a third color on the third plan view. At this time, a third color can be preset to represent the free color; for example, it can be set to blue.

[0138] In other words, if device 200 determines that the first difference is not higher than the second reference value, it can process the third floor plan and display the background color of the first area as the third color.

[0139] Figure 4 It is a flowchart illustrating the process of marking the boundary lines of a region according to a single embodiment.

[0140] According to one embodiment, Figure 4 Each step shown can be executed. Figure 3 Perform each step shown.

[0141] refer to Figure 4 First, in step S401, the number of basic production equipment deployed in the first area is determined as the number of first equipment based on the third plan view, and the number of basic IT equipment deployed in the first area is determined as the number of second equipment.

[0142] In step S402, as the number of first devices increases, device 200 can set the third value to a higher value within the third reference range. Here, depending on the embodiment, the third reference range can be set differently, for example, within the range of 0 to 100.

[0143] For example, if the quantity of the first device is determined to be 1, then device 200 can set the third value to 10; if the quantity of the first device is determined to be 2, then the third value can be set to 20.

[0144] In step S403, as the number of second devices increases, the device 200 can set the fourth value to a higher value within the fourth reference range. Here, the fourth reference range can be set differently depending on the embodiment; for example, it can be set in the range of 0 to 100.

[0145] For example, if the quantity of the second device is determined to be 1, then device 200 can set the fourth value to 10; if the quantity of the second device is determined to be 2, then the fourth value can be set to 20.

[0146] In step S404, device 200 can calculate the importance of region 1 by adding the third and fourth values.

[0147] Device 200 can calculate the importance of each complex region in the same way as it calculates the importance of the first sector.

[0148] In step S405, if the importance of each of the multiple regions has been calculated, the device 200 can calculate the average of the calculated importance of each complex region as a second average.

[0149] In step S406, device 200 can generate a second difference because the importance of the first region is subtracted from the second average value.

[0150] In step S407, the device 200 can set the first thickness to a thicker thickness within a thickness range having a higher second difference value. Here, the thickness range is a range of linewidths, which can be set differently depending on the embodiment, for example, within a range of 0 mm to 10 mm.

[0151] For example, if the second difference is identified as -10, the device 200 can set the first thickness to 4mm; if the second difference is determined to be 0, the first thickness can be set to 5mm; and if the second difference is determined to be 10, the first thickness can be set to 6mm.

[0152] In step S408, the device 200 can determine the thickness of the first line as the thickness of the first line.

[0153] In step S409, device 200 can determine the complementary color of the display color of the first area on the third floor plan as the color of the first line.

[0154] Specifically, if the display color of the first cross section is identified as the first color on the third plan view, the device 200 can determine the complementary color of the first color as the color of the first line through step S310. Through step S312, if the display color of the first cross section is identified as the second color on the third plan view, the complementary color of the second color can be determined as the color of the first line. Through step S313, if the display color of the first cross section is identified as the third color on the third plan view, the complementary color of the third color can be determined as the color of the first row.

[0155] For example, if the complementary color of blue is yellow, then if the display color of the first area is identified as blue on the third plan view, then device 200 can determine yellow as the color of the first row.

[0156] In step S410, device 200 can be configured to display the first line as the boundary line of the first part on the third floor plan.

[0157] In other words, when the thickness and color of the first line are determined through steps S408 to S409, the device 200 can be configured to display the first line by adding the first line to the boundary line of the first part on the third floor plan.

[0158] Figure 5 This is a preliminary schematic diagram of the device configuration according to an embodiment.

[0159] An apparatus 200 according to one embodiment includes a processor 210 and a memory 220. The processor 210 may include reference... Figures 1 to 4 At least one of the above-mentioned devices, or referring to Figure 4 Perform at least one of the above methods. An individual or group using device 200 may provide information as referenced. Figures 1 to 4 Services related to some or all of the methods mentioned above.

[0160] The memory 220 may store information related to the methods described above, or it may store programs that implement the methods described below. The memory 220 may be volatile or non-volatile memory.

[0161] Processor 210 can execute programs and control device 200. The code of the program executed by processor 210 can be stored in memory 220. Device 200 is connected to external devices (e.g., personal computers or networks) via input / output devices (not shown) and can exchange data via wired and wireless communications.

[0162] Device 200 can be used to train an AI model or to use a trained AI model. Memory 220 may contain a trained AI model or a pre-trained AI model. Processor 210 can train or run algorithms for the AI ​​model stored in memory 220. The learning device for training the AI ​​model and device 200 for using the trained AI model can be the same or separate.

[0163] The above embodiments can be implemented as hardware components, software components, and / or combinations of hardware and software components. For example, the devices, methods, and components described in the embodiments can be implemented using one or more general-purpose or special-purpose computers, such as processors, controllers, arithmetic logic units (ALUs), digital signal processors, microcomputers, field-programmable gate arrays (FPGAs), programmable logic units (PLUs), microprocessors, or any other device capable of executing and responding to instructions. The processing unit can execute an operating system (OS) and one or more software applications executed on the operating system. The processing unit can also access, store, process, and generate data in response to software execution. For ease of understanding, a processing unit can be described as a single processing unit, but those skilled in the art will recognize that a processing unit may comprise multiple processing elements and / or various types of processing elements. For example, a processing unit may include multiple processors or a processor and a controller. Furthermore, other processing configurations, such as parallel processors, may be used.

[0164] The method according to this embodiment can be implemented in the form of program instructions, which can be executed and recorded on a computer-readable medium by various computer means. The computer-readable medium can contain program instructions, data files, data structures, etc., alone or in combination. The program commands recorded on the medium can be specifically designed and configured for the embodiment, or they can be known and available to a computer software craftsman. Examples of computer-readable recording media include magnetic media (such as hard disks, floppy disks, and magnetic tapes), optical media (such as CD-ROMs and DVDs), magneto-optical disk media (such as floppy disks), and hardware devices specifically configured to store and execute program commands (such as ROMs, RAMs, flash memory, etc.). Examples of program instructions include machine code (e.g., code generated by a compiler) and high-level language code (e.g., code executable by a computer using an interpreter). The hardware device can be configured to run as one or more software modules to perform the operation of the embodiment, and vice versa.

[0165] This software may include computer programs, code, instructions, or one or more combinations thereof, and may configure processing units to operate as intended, or may individually or collectively command processing units. Software and / or data may be permanently or temporarily contained in any type of machine, component, physical device, virtual device, computer storage medium, or apparatus, or in transmitted signal waves, so that processing units may interpret them or provide instructions or data to processing units. This software is distributed on networked computer systems and may be stored or executed in a distributed manner. Software and data may be stored on one or more computer-readable recording media.

[0166] Although the above embodiments have been described with limited drawings, those skilled in the art can make various technical modifications and alterations based on the above description. For example, appropriate results may be obtained if the described techniques are performed in a different order than the described methods, and / or if the components of the described systems, structures, devices, circuits, etc., are combined or combined in a different manner than the described methods, or are replaced or substituted by other components or equivalents.

[0167] Therefore, other embodiments, other embodiments, and those equivalent to the patent claims also fall within the scope of the claims described below.

Claims

1. A method, device, and system for the automated design, construction, and related solution delivery of IT equipment infrastructure that considers security and resource efficiency factors in industry characteristics, characterized in that... The system receives a first floor plan from the user terminal, which serves as a floor plan of the first factory. If it confirms that the first factory is the factory that produces the first product, it categorizes the production equipment required for producing the first product as essential production equipment and the IT equipment required for the operation of the first factory as essential IT equipment. Then, it matches the first floor plan with the required production equipment to generate a first matching result. This result is encoded to generate a first input signal, which is then input into a trained first AI model capable of selecting the optimal placement location of the production equipment on the floor plan. When the placement location of the essential production equipment is selected on the first floor plan using the first input signal, a first output signal indicating the placement location of the required production equipment is obtained from the first AI model. Based on this, the system then... The required production equipment is marked on the map to create a second floor plan. The second floor plan is then matched with the required IT equipment to generate a second matching result. This result is encoded to generate a second input signal, which is then input into a second artificial intelligence model that has been trained to select the optimal placement of IT equipment on the floor plan. When the location of the necessary IT equipment is selected on the second floor plan using the second input signal, a second output signal indicating the location of the necessary IT equipment is obtained from the second artificial intelligence model. Based on this, the required IT equipment is marked on the second floor plan to create a third floor plan. The third floor plan is then transmitted to the user terminal. This approach balances security and resource efficiency, providing automated solutions for the design and construction of IT equipment infrastructure for each industry.

Images