3D simulation system including real-time risk prediction and user history-based maintenance adjustment functions and 3D simulation method using the same

Abstract

A 3D simulation system that dynamically adjusts feedback and scenarios of maintenance simulation by reflecting the user's risk history and real-time response information, and a 3D simulation method using the same are provided. A 3D simulation system according to one aspect of the present invention includes a memory configured to store instructions and a processor configured to execute the instructions to: collect simulation data regarding a user's maintenance work input from a user terminal used by a user; analyze the user's real-time input, past work history, and error patterns from the simulation data to derive an analysis result reflecting the risk level during the maintenance work; generate 3D simulation content in which the scenario path and warning feedback level of the 3D simulation are adjusted based on the analysis result; and provide the generated 3D simulation content to the user terminal.

Description

Technology Field

[0001] The present invention relates to a 3D simulation system including real-time risk prediction and user history-based maintenance adjustment functions, and a 3D simulation method using the same. More specifically, it relates to a 3D simulation system that predicts risk factors based on a worker's past history and real-time response information, and automatically adjusts simulation scenarios and feedback intensity, and a 3D simulation method using the same. Background Technology

[0002] Devices composed of complex structures and diverse components, such as industrial facilities, machinery, and construction equipment, require a high level of proficiency and a precise understanding of procedures during maintenance processes. These devices are often operated in high-risk work environments, and failure to clearly understand their structure, operating principles, fault diagnosis, and replacement procedures can lead to serious accidents or equipment damage. Accordingly, maintenance workers must receive sufficient prior training on the structure and operation of the equipment and perform their duties based on an awareness of safety guidelines and regulations.

[0003] In response to this need, simulation-based virtual training systems have been introduced across various industries. In particular, with the advancement of 3D visualization technology and virtual reality (VR) and augmented reality (AR) technologies, users are now able to operate products or learn maintenance procedures in a virtual environment without directly handling actual equipment. These systems are receiving increasing attention for their ability to prevent risks that may occur in the field and to improve worker proficiency through repetitive learning.

[0004] Existing 3D simulation systems have focused on enabling users to intuitively understand the operating principles of a product by visually guiding them through the structure of the equipment, the location of parts, and operation methods. Additionally, some systems have provided functions that guide operation by following a sequence of tasks based on scenarios and provide feedback on operation results through animations in response to user input. However, most of these technologies are centered around fixed content, making it difficult to provide responsive content tailored to the user's proficiency or learning history, and resulting in reduced learning efficiency due to the repetitive experience of the same scenarios.

[0005] Furthermore, equipment and systems used in the field are subject to a wide variety of variables depending on the situation, and specific components or structures may entail exceptional risk factors. Accordingly, for more precise maintenance training, technology is required that goes beyond simply guiding users on equipment structure. It must be capable of adjusting simulation difficulty based on individual user understanding, past practice history, frequency of errors, and proficiency, while also providing advance prediction and real-time warnings for hazardous situations during operation.

[0006] Existing systems designed simulation content by considering various risk factors in advance, but they were limited to outputting warnings in specific situations or providing limited feedback. For example, while there is a feature that displays a red indicator or a warning sound when a specific part is mishandled, this is merely a reaction based on static judgment and falls short of providing customized warnings by learning a user's past behavioral history or repeated error patterns. As a result, users repeat the same mistakes without the opportunity to systematically correct them, and the educational effectiveness of the simulation remains limited.

[0007] Furthermore, since maintenance work is closely linked to relevant laws, safety regulations, and work standards, training systems require capabilities that go beyond mere structural understanding to determine compliance with these regulations. However, existing technologies fail to implement a system that automatically diagnoses whether a user's current operations are legally compliant by linking in real-time with external standard databases or public institution work guidelines, and provides guidance for correction in the event of a violation. Consequently, workers trained in the field may proceed with tasks in violation of regulations or unconsciously repeat dangerous procedures.

[0008] Furthermore, existing systems have a structural limitation in that the flow of simulation scenarios is linear and fixed, meaning the scenarios themselves cannot recognize or correct for situations where users repeatedly induce the same risks. This limitation leads to a failure to meet the practical objectives of correcting recurring risk tendencies and preventing accidents, particularly in maintenance training for high-risk equipment or complex multi-component devices.

[0009] As such, existing 3D simulation-based maintenance training systems have limitations such as static configuration of educational content, lack of personalized learning for users, absence of real-time risk assessment, insufficient regulatory integration capabilities, and unrealistic scenario correction, and these problems act as major obstacles to ensuring safety and work efficiency in actual industrial sites. The problem to be solved

[0010] The present invention was devised to solve the aforementioned problems, and aims to provide a 3D simulation system that dynamically adjusts the feedback and scenarios of a maintenance simulation by reflecting the user's risk history and real-time response information, and a 3D simulation method using the same.

[0011] However, the technical problems that the present invention aims to solve are not limited to those described above, and other unmentioned problems will be clearly understood by those skilled in the art from the description of the invention below. means of solving the problem

[0012] A 3D simulation system according to one aspect of the present invention includes a memory configured to store instructions and a processor configured to execute the instructions to: collect simulation data regarding a user's maintenance work input from a user terminal used by a user; analyze the user's real-time input, past work history, and error patterns from the simulation data to derive an analysis result reflecting the risk level during the maintenance work; generate 3D simulation content in which the scenario path and warning feedback level of the 3D simulation are adjusted based on the analysis result; and provide the generated 3D simulation content to the user terminal.

[0013] Preferably, the processor is configured to accumulate and store the user's repetitive error patterns appearing in the simulation data in chronological order, and to calculate the risk level based on a first factor regarding the operation step where the error occurred, a second factor regarding the type of part during the operation, a third factor regarding the user's reaction time, a fourth factor regarding whether the user accepted previous feedback, and a fifth factor regarding the frequency of repetition of the same error, whether a warning response occurred at the time of the error, and whether there was a subsequent change in behavior. For operation items with a high risk level, the processor may be configured to assign weights to each of the first to fifth factors according to the priority of providing warning feedback, the level of detailed explanation, the visual emphasis method, the warning intensity, and the number of repetitions when generating 3D simulation content.

[0014] Preferably, the processor selects a scenario path of the 3D simulation from a plurality of predefined scenario paths based on the user's risk-inducing history among the analysis results, corrects the selected scenario path of the 3D simulation by reflecting the user's error occurrence tendency and warning response history, and the risk-inducing history may include a high-risk operation history that occurred repeatedly in the same work step, frequent misoperations in a specific part group and the number of warnings caused thereby, the interval between errors relative to work time, the delay time in response to warnings, whether there were consecutive mistakes immediately after an error occurred, the correlation with a scenario section containing a violation of regulations, a response pattern that was not improved after a warning even though a high-risk warning was output during past 3D simulations, and the frequency of entering a dangerous operation even though attention-raising feedback was provided during the overall operation flow.

[0015] Preferably, the processor may be configured to be linked with a work standard database or a safety regulation database provided by a public institution, and to determine whether there is a violation of a standard or safety regulation by comparing the user's 3D simulation operation content shown in the simulation data with the standard provided from the work standard database or the safety regulation provided from the safety regulation database, and to output a warning message by assigning weights to each of the violation items for the risk level, number of violations, priority of application of the regulation, importance of the operation step, and user response, and to change the scenario of the 3D simulation.

[0016] Preferably, the processor may be configured to analyze the user's proficiency in 3D simulation operation by analyzing the user's operation accuracy, response time, acceptance of feedback, results of repeated execution of the same task, frequency of warning occurrence, and violation rate of regulations during the user's 3D simulation operation process, and to adjust the frequency, intensity, level of visual emphasis, and detail of explanation of providing feedback according to the proficiency analysis results.

[0017] Preferably, the processor may be configured to analyze the user's operation accuracy, response speed, type of error occurrence, tendency to accept feedback, improvement rate of iterative learning, fixation time, and recovery pattern after operation failure shown in the simulation data, and to adjust the difficulty of the feedback provided thereafter, the level of detail of the explanation, and the visual emphasis pattern for each user based on the analysis.

[0018] Preferably, the processor may be configured to generate 3D simulation content that allows a user to understand the operating principle of the device by identifying associated parts that are mechanically, electrically, or logically linked with a specific part provided in the device being manipulated during the 3D simulation, and by activating the state of the identified associated parts in real time to visually and simultaneously represent the structural connection or functional dependency relationship between the specific part and the associated parts.

[0019] A 3D simulation method according to one aspect of the present invention comprises: a step of collecting simulation data regarding a user's maintenance work input from a user terminal used by a user; a step of deriving an analysis result reflecting the risk level during maintenance work by analyzing the user's real-time input, past work history, and error patterns from the simulation data; a step of generating 3D simulation content in which the scenario path and warning feedback level of the 3D simulation are adjusted based on the analysis result; and a step of providing the generated 3D simulation content to the user terminal. Effects of the invention

[0020] According to an embodiment of the present invention, by providing a user-customized simulation flow and quantitatively analyzing the user's risk response capabilities to provide appropriate feedback, the ability to perform maintenance work and safety in actual industrial sites can be effectively improved.

[0021] In addition, various other additional effects may be achieved by various embodiments of the present invention. These various effects of the present invention are described in detail in each embodiment, or the description of effects that are easily understood by those skilled in the art is omitted. Brief explanation of the drawing

[0022] The following drawings attached to this specification illustrate preferred embodiments of the present invention and serve to further enhance understanding of the technical concept of the present invention together with the detailed description of the invention provided below; therefore, the present invention should not be interpreted as being limited only to the matters described in such drawings. FIG. 1 is a diagram illustrating components for performing a 3D simulation method according to an embodiment of the present invention. FIG. 2 is a drawing for explaining the elements constituting a system for performing a 3D simulation method according to an embodiment of the present invention. Figure 3 is a diagram showing an example of a 3D simulation method implemented by the system of Figure 2. FIG. 4 is a flowchart illustrating a 3D simulation method according to an embodiment of the present invention. Specific details for implementing the invention

[0023] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the attached drawings. Prior to this, terms and words used in this specification and claims should not be interpreted as being limited to their ordinary or dictionary meanings, and should be interpreted in a meaning and concept consistent with the technical spirit of the present invention, based on the principle that the inventor can appropriately define the concept of the terms to best describe his invention.

[0024] Therefore, the embodiments described in this specification and the configurations illustrated in the drawings are merely the most preferred embodiments of the present invention and do not represent all of the technical ideas of the present invention; thus, it should be understood that various equivalents and modifications that can replace them may exist at the time of filing this application.

[0025] FIG. 1 is a drawing for explaining components for performing a 3D simulation method according to an embodiment of the present invention, FIG. 2 is a drawing for explaining components constituting a system (100) for performing a 3D simulation method according to an embodiment of the present invention, and FIG. 3 is a drawing showing an example of a 3D simulation method implemented by the system (100) of FIG. 2.

[0026] Referring to FIGS. 1 to 3, each component for performing a 3D simulation method according to one embodiment of the present invention may include a system (100) for performing a 3D simulation method and a user terminal (200).

[0027] Specifically, the system (100) can exchange data with user terminals (200) via wired or wireless communication and provide facility simulation to the user terminals (200). In one embodiment, the user terminal (200) may be a mobile device or PC used by users. The system (100) may be a server device or a computer device for performing a 3D simulation method.

[0028] The 3D simulation method can be implemented in the form of a computer program or a mobile application. For example, processes for the operation of the 3D simulation method can be performed by the user terminal (200) running the computer program and the mobile app and the system (100) running the computer program.

[0029] Additionally, the server (100) includes an AI model and can provide a 3D simulation method to a customer terminal (200) through the learning of the AI ​​model.

[0030] The above AI model may be trained based on unsupervised learning methods, but is not limited thereto, and may also be trained based on supervised learning or semi-persistent learning methods.

[0031] In addition, an AI model may consist of one or more neural network layers, and each neural network may include one or more weights. An AI model can perform learning or inference by performing operations between one or more weights and input data.

[0032] Referring to FIG. 2, the system (100) may include memory (110) and a processor (120). However, it is not limited thereto, and other general-purpose components may be further included in the server (100).

[0033] Memory (110) may be configured to store instructions of a computer program or application that implements a 3D simulation method, and a processor (120) may execute the program or application by executing the instructions stored in memory (110). For example, memory (110) may be implemented as non-volatile memory such as ROM, PROM, EPROM, EEPROM, flash memory, PRAM, MRAM, RRAM, FRAM, etc., or volatile memory such as DRAM, SRAM, SDRAM, PRAM, RRAM, FeRAM, etc., and may be implemented in the form of HDD, SSD, SD, Micro-SD, etc., or a combination thereof. The processor (120) may be implemented as an array of logic gates, a microprocessor, a CPU, a GPU, an AP, or a combination thereof.

[0034] Specifically, the processor (120) can collect simulation data regarding the user's maintenance work input from the user terminal (200) used by the user. The simulation data collected at this time may include various forms of interaction data, such as the user's input method, operation flow, selected parts, viewpoint movement, touch or click response, and whether the operation is completed. This data is used as basic data to accurately identify the user's behavioral characteristics and work response. In particular, in the context of maintenance work, it is important to acquire detailed data in real time, including the order in which the user operated which devices, where errors were made, and how much time was spent on each task.

[0035] Additionally, the processor (120) can derive an analysis result that reflects the risk level during maintenance work by analyzing the user's real-time input, past work history, and error patterns from the simulation data. This analysis can be performed by comprehensively considering the user's learning history, the type and frequency of errors that occurred during previous simulations, warning response patterns, recovery behavior after operation errors, and whether the same error is repeated. For example, if repetitive errors occur during the operation of the same part, or if the user's reaction time is long at a specific stage, the corresponding factor is classified as high risk and can be reflected in the design of subsequent 3D simulation content. Furthermore, by patterning the types of errors and classifying them into lack of awareness, sequence errors, insufficient operation speed, etc., it is possible to make a detailed diagnosis of the user's weaknesses. These analysis results can be utilized as key insights that are directly reflected in the flow control of the simulation scenario, rather than as simple static statistical information.

[0036] In addition, the processor (120) can generate 3D simulation content in which the scenario path and the level of warning feedback of the 3D simulation are adjusted based on the analysis results. The 3D simulation content generated here refers to content that is not fixed to a single path, but is dynamically adjusted to suit the user. For example, for high-risk work stages, warning animations and voice guidance may be output simultaneously; for part operations where the user frequently made mistakes, step-by-step operation guidance may be reinforced; and for operations to which legal standards apply, warning messages linked to public standard information may be included. Furthermore, when an increase in the user's proficiency is recognized, unnecessary repetitive guidance may be omitted and the content may be structured around key elements, thereby increasing immersion and reducing learning fatigue, and the difficulty and level of explanation of the content may be adjusted.

[0037] Additionally, the processor (120) may be configured to provide the generated 3D simulation content to the user terminal (200). This 3D simulation content is transmitted from the system (100) to the user terminal (200) in real time, and the user terminal (200) outputs it through a display using a 3D graphic-based interface to induce user operation. The 3D simulation content provided to the user in this manner is composed of responsive content that changes dynamically according to user input, and visual feedback, warning messages, component activation effects, scenario branching, etc., can be immediately reflected according to the operation results. In particular, by designing this 3D simulation content to be executable via the web without separate installation on the user terminal (200), accessibility and applicability in various device environments (PC, tablet, kiosk, etc.) can be ensured.

[0038] According to this embodiment of the present invention, by providing a user-customized simulation flow and quantitatively analyzing the user's risk response capabilities to provide appropriate feedback, the ability to perform maintenance work and safety in actual industrial sites can be effectively improved.

[0039] In one embodiment, the processor (120) may be configured to accumulate and store the user's repetitive error patterns shown in the simulation data in chronological order, and to calculate the risk based on a first factor regarding the operation step where the error occurred, a second factor regarding the type of part during the operation, a third factor regarding the user's reaction time, a fourth factor regarding whether the user accepted previous feedback, and a fifth factor regarding the frequency of repetition of the same error, whether there was a warning response at the time the error occurred, and whether there was a change in behavior thereafter.

[0040] And, the processor (120) may be configured to assign weights to each of the first to fifth factors according to the ranking of warning feedback provision, level of detailed description, visual emphasis method, warning intensity, and number of repetitions when generating 3D simulation content for high-risk operation items.

[0041] Specifically, when determining the priority of providing warning feedback, the processor (120) may give greater weight to the fifth factor than to the first through fourth factors. This is because if the same error is repeated cumulatively but the user response has not improved, the item is considered a high-risk item requiring immediate guidance, and thus the warning priority must be raised. In addition, when determining the priority of providing warning feedback, the processor (120) may also give weight to the second factor (type of part). For example, if high-risk parts such as high-voltage parts, rotating parts, or hydraulic devices are associated with the error, the processor (120) may give greater weight to the second and fifth factors than to the first, third, and fourth factors.

[0042] Additionally, the processor (120) may give greater weight to the first and fourth factors than to the second, third, and fifth factors to include more detailed explanations, step-by-step guidance, diagrammatic representations, example-based explanations, etc., in relation to the level of detailed explanation, if the operation step corresponds to the functional core of the 3D simulation or if the user has a history of repeatedly ignoring or failing to understand the same feedback.

[0043] Additionally, the processor (120) may give greater weight to the third and fifth factors than to the first, second, and fourth factors so that visual stimuli such as highlight colors, flashing effects, enlarged displays, arrows, or path highlighting are reinforced on the screen, as it is considered that the user's reaction speed in a specific operation related to the visual highlighting method is slow, or the response after the warning is inappropriate, or the difficulty of perception is high or attention is scattered. Also, the processor (120) may give greater weight to the third and fifth factors than to the first, second, and fourth factors so that if the user has a history of repeating the same error multiple times related to the visual highlighting method, the visual warning display is applied first so that it is immediately output on the first screen of the feedback.

[0044] Additionally, the processor (120) can generate 3D simulation content based on a composite weight of a second factor (part type), a fifth factor (error repetition), and a fourth factor (whether previous feedback was ignored) in relation to the warning intensity condition. For example, if the same error is repeated in a high-risk part and the user feedback response is negative, the processor (120) can output a warning message with increased intensity, such as a simulation pause, a voice warning, or a red screen overlay, rather than a simple message. Conversely, if a one-time error occurs in a group of minor parts, the processor (120) can mitigate the intensity by replacing the warning with only a visual display.

[0045] Additionally, the processor (120) can generate 3D simulation content based on the composite weight of the fourth and fifth factors, which are closely related to whether user response is improved in relation to the number of repetitions. For example, the processor (120) can increase the number of repetition outputs for a user who has a low feedback acceptance rate while the same type of error occurs continuously, and can provide the same content repeatedly by transforming it into various ways (text → animation → voice). On the other hand, the processor (120) can set a feedback policy to maintain user concentration by omitting or minimizing the provision of repeated feedback when a history of success is confirmed more than a certain number of times.

[0046] According to this embodiment of the present invention, not only can the learning effect for each user be maximized, but the safety and educational effect of the entire system can also be enhanced by providing intensive and reinforced warnings regarding dangerous operations or operation items with a high probability of violating regulations.

[0047] In one embodiment, the processor (120) selects a scenario path for a 3D simulation from a plurality of predefined scenario paths based on the user's risk-inducing history among the analysis results, and The scenario path of the selected 3D simulation can be corrected by reflecting the user's error occurrence tendency and warning response history.

[0048] At this time, the risk-inducing history may include a history of high-risk operations that occurred repeatedly in the same work stage, frequent misoperations in specific parts groups and the number of warnings generated as a result, the interval between errors relative to work time, the delay time in responding to warnings, whether there were consecutive mistakes immediately after the error occurred, the association with scenario sections containing regulatory violations, response patterns that were not improved after warnings despite high-risk warnings being output during past 3D simulations, and the frequency of entering into hazardous operations despite providing alert feedback during the overall operation flow.

[0049] Each of these risk-inducing history indicates signs such as the user's failure to detect risks, lack of error recovery capabilities, insufficient awareness of regulations, and delays in task recognition, which serve as key criteria for determining the need for correction when designing scenario paths in 3D simulations.

[0050] Accordingly, the processor (120) may be configured to assign different weights to each of the above risk-inducing histories by considering conditions that influence the scenario correction judgment (e.g., threshold of the task, complexity of the operation, impact of system risk, etc.).

[0051] For example, a history of repeated high-risk operations in the same work step indicates that the user lacks structural understanding of a specific step. Since this item has high predictive power regarding the risk of repetition even within the overall scenario, the processor (120) can give a high weight to the item to prioritize presenting an automatic correction route in that work step or simplify the 3D simulation scenario to guide the user to enable repeated learning.

[0052] In addition, the magnitude of the risk may vary depending on the type of equipment or the characteristics of the parts, as there may be frequent errors in specific parts groups and the number of warnings generated as a result. In particular, if such history appears in high-risk equipment groups (e.g., hydraulic, electrical, rotating machinery, etc.), this item is considered a key risk trigger in the scenario correction criteria of the 3D simulation, so the processor (120) can automatically adjust the warning interval to be shorter and correct the 3D simulation path in a direction that increases the number of warning repetitions.

[0053] On the other hand, a short interval between errors relative to the work time may imply a decrease in user concentration or a failure to establish an operation plan. The processor (120) can analyze this interval information and, if errors are concentrated within a specific time, output a pre-warning notice before reaching that point, or configure the 3D simulation scenario to divide the concentration maintenance period and the repeated verification period. This item may be assigned a moderate weight as a time-based risk prediction signal.

[0054] The delay in response to a warning can also serve as a factor reflecting user perception and execution response. If a history of failing to respond immediately to the same warning accumulates, the warning type may not be suitable for the user's visual or auditory perception, so the processor (120) can correct the scenario flow by adding visual emphasis or changing the warning form according to this item. The average value and standard deviation of the response delay serve as indicators for determining situational awareness failure, and weights can be adjusted according to the impact.

[0055] A series of errors immediately following an error is a major indicator that can cause complex risks, which may include fatigue, confusion, and overconfidence in automation. When this item is detected, the processor (120) may insert a break period of a certain duration after performing a single scenario for the user, or automatically correct the configuration by simplifying the interaction between parts during that period. This item is an indicator that captures multiple errors within a short period of time and may have a high weight as an emergency correction route switching condition.

[0056] Since the association with a scenario section containing a violation of regulations directly affects the safety and compliance of the operation, that item can always be assigned the highest weight. In particular, if the current user's risk operation overlaps with a section where a violation of regulations occurred in the same scenario in the past, the processor (120) can automatically convert that section into a regulation-centered correction scenario that includes inserting a pre-guide, restricting operations, and branching scenarios.

[0057] Finally, if there is a high frequency of entering into a dangerous task despite warning feedback already being provided, the processor (120) determines that the user did not recognize or ignored the general feedback format and can correct the scenario for the item by resetting the feedback format (e.g., animation → voice warning), reinforcing the repeated output, or inserting a verification procedure after stopping the simulation. This is an item directly related to user compliance, and its weight can increase rapidly upon repeated accumulation.

[0058] In this way, the processor (120) can be configured to evaluate the relative importance and contribution of each of the multiple risk-inducing items based on predefined criteria and real-time analysis values, and thereby automatically adjust the correction direction and strength of the simulation path. Unlike existing systems that simply present a fixed scenario path, this method can provide a learning flow and risk response strategy optimized for each user, thereby contributing to securing both practicality and stability applicable to actual industrial environments.

[0059] In one embodiment, the processor (120) may be linked with a work standard database or safety regulation database provided by a public institution.

[0060] And, the processor (120) can determine whether there is a violation of the standard or safety regulation by comparing the user's 3D simulation operation content shown in the simulation data with the standard provided from the work standard database or the safety regulation provided from the safety regulation database.

[0061] Additionally, the processor (120) may be configured to output a warning message and change the scenario of the 3D simulation by assigning weights to each of the risk level, number of violations, priority of application of regulations, importance of operation steps, and user response for the violation item.

[0062] Specifically, the processor (120) outputs a warning message immediately if the user's operation violates a mandatory regulation, because the regulation is an item that must be complied with for legal or safety reasons, and even a single violation can lead to a physical accident or legal liability. For example, the processor (120) is configured to output a warning message immediately regardless of whether there is a violation of the regulation, such as checking insulation of high-voltage components or checking ventilation before opening a pressure valve. In this case, the highest weight can be given to the priority factor of the regulation application rather than the risk level, the number of violations, the importance of the operation step, or the user's reaction to the violation item.

[0063] On the other hand, the processor (120) outputs a step-by-step warning message in the case of arbitrary rule violations, taking into account the cumulative frequency and whether the previous feedback was accepted. In this case, since the user's learning tendency and level of familiarity with the rule must be reflected, a higher weight may be given to the factors of 'number of violations' and 'user response' than to other factors. For example, even if it is an arbitrary rule, if the user repeatedly violates the same item and does not respond to the warning, the processor (120) considers the importance of the rule to have substantially increased, so the next warning is visually and audibly reinforced, and in some cases, the workflow itself may be switched to a correction scenario.

[0064] Additionally, the processor (120) may raise the warning level regardless of user proficiency when the target of operation corresponds to a high-risk component under safety regulations (e.g., power system, rotating body, pressurized part, etc.) and may give the highest weight to the 'risk level' factor over other factors. This is because the risk of a specific component is directly linked to a fixed actual risk regardless of user proficiency or warning history. Therefore, a violation of regulations regarding the component can immediately lead to strong visual and auditory warnings and a restriction on re-entry into the scenario.

[0065] Additionally, the processor (120) may include a function to provide guidance on delaying operation or entering a correction scenario in addition to the warning message when the same user repeatedly violates the same regulation. This is a measure to help the user feel the importance of the regulation in an educational context, and in this case, the processor (120) may apply a composite weight to the 'number of violations' and 'user response status'. Furthermore, the processor (120) may change the scenario configuration so that the repeatedly violated item is displayed as a priority inspection item when entering a simulation later, or a pre-warning popup is displayed.

[0066] Finally, the processor (120) can adjust the priority of warning colors and notification messages according to the degree of exceeding the error range in the case of numerical-based rules such as work intervals, distances between devices, installation directions, and fastening strengths that are enforced by specific legal regulations. At this time, the processor (120) can apply quantitative weights in real time by adding a ‘numerical deviation of the degree of violation’ as an auxiliary factor. For example, if the error exceeds 10% of the standard value, a yellow warning may be issued, and if it exceeds 30%, a red warning may be issued along with a work stoppage measure.

[0067] As a result, the processor (120) can go beyond simply determining whether there is a violation of regulations and, through multidimensional analysis of each violation item, assign weights to five key judgment factors on a situational basis, and precisely control the warning message output method and scenario transition flow based on the weights, thereby guiding the user to achieve not only a customized learning effect but also the internalization of safety regulations and the improvement of worker risk sensitivity simultaneously.

[0068] In one embodiment, the processor (120) can analyze the user's proficiency in 3D simulation operation by analyzing the user's operation accuracy, response time, acceptance of feedback, results of repeated execution of the same task, frequency of warning occurrence, and violation rate of regulations during the user's 3D simulation operation process.

[0069] Additionally, the processor (120) may be configured to adjust the frequency, intensity, level of visual emphasis, and detail of the explanation of the feedback provided according to the results of the proficiency analysis.

[0070] Such proficiency judgment and feedback adjustment can be performed by assigning weights to each factor based on quantified evaluation items. The processor (120) can determine the weights according to the following conditions and dynamically change the feedback pattern accordingly.

[0071] First, operation accuracy is one of the most critical factors in determining proficiency, and can be evaluated by comparing the error rate or agreement rate of the actual operation performed with the standard operation set at each work step. Since users with high operation accuracy tend to perform specific actions quickly and accurately, the processor (120) gives a higher weight to that factor than other factors to evaluate the overall proficiency upward, thereby simplifying the details of the explanation and reducing the frequency of feedback.

[0072] Second, the response time can be used as an indicator of the user's situational awareness and ability to respond immediately. If a user's response delay to a specific warning or guidance message accumulates, it is determined that the user lacks real-time task response capability, and the processor (120) can increase the intensity of feedback (e.g., visual flashing, voice warning, etc.) by applying a time interval inverse weight to this factor. On the other hand, the processor (120) can adjust the feedback by reducing the frequency of visual emphasis for users whose response speed is faster than a certain level.

[0073] Third, the acceptance of feedback can be measured based on whether behavior changes in response to previously provided feedback. For example, if similar errors occur again even after repeated feedback on the same error, the feedback acceptance rate is judged to be low, and in this case, the user's proficiency may be considered low. The processor (120) gives a high weight to this factor, the detail of the explanation is increased, and the feedback format (e.g., animation, step-by-step diagram, etc.) can be automatically switched according to the user's response history.

[0074] Fourth, the results of repeated performance of the same task can be used as an item to evaluate the user's continuous learning effect and memory retention. If the error rate does not decrease or the performance time does not shorten despite performing the same task scenario multiple times, it is determined that proficiency has not increased, and in this case, the processor (120) can maintain or increase the frequency of feedback and the intensity of explanation. Conversely, if the success rate has significantly improved and the performance time has consistently decreased, the processor (120) can lower the intensity of feedback for the user and switch to a mode that provides summary guidance focusing on key warning items.

[0075] Fifth, the frequency of warning occurrences can be used as an indicator to judge the error density in the overall scenario flow rather than a single operation. Since users with narrow intervals and high frequency of warning occurrences during 3D simulation are likely to have a lack of overall risk awareness or procedural confusion, the processor (120) can give a higher weight to this factor than other factors to increase the visual emphasis level and warning intensity, and increase the number of repetition explanation cycles.

[0076] Sixth, the violation rate is a signal indicating the user's level of knowledge and compliance with regulations, and in particular, if the frequency of violating legal standards is high, it may operate to reinforce warning messages regardless of proficiency level. This serves as a judgment criterion for designing feedback by clearly distinguishing between operational errors caused by mistakes and non-compliance with regulations, and the processor (120) may add feedback components or increase the duration of the feedback by applying a separate legal violation correction factor to the relevant factor.

[0077] In this way, the processor (120) can calculate relative weights for each evaluation item based on the user's behavior data within the 3D simulation, and by adjusting the frequency, intensity, visual emphasis level, and explanation detail of the feedback according to the result, it can realize feedback adaptation optimization according to the user's proficiency.

[0078] Consequently, embodiments of the present invention move away from static manual-centered feedback and implement a learning-adaptive 3D simulation system that adjusts the feedback pattern in real time by reflecting user-specific behavioral patterns and skill levels, thereby contributing to simultaneously increasing the efficiency of repetitive learning and the acceptability of risk warnings.

[0079] In one embodiment, the processor (120) can analyze the user's operation accuracy, response speed, type of error occurrence, tendency to accept feedback, iterative learning improvement rate, fixation time, and recovery pattern after operation failure shown in the simulation data.

[0080] And, the processor (120) can be configured to adjust the difficulty of the feedback provided thereafter, the level of detail of the explanation, and the visual emphasis pattern for each user based on the analysis content.

[0081] More specifically, the processor (120) may assign weights to the relative importance of each item based on the following conditions and reflect the results in a feedback method.

[0082] First, the processor (120) evaluates the accuracy of the operation by analyzing the match rate between the input value for each operation step of the user and the predetermined target operation, and can adjust the difficulty of the feedback and the level of explanation to be simplified by giving a higher weight to the item as the accuracy is higher.

[0083] Additionally, the processor (120) can analyze the response speed by measuring the user's response time to a warning or guidance message in real time, and if the response time is long or delayed, it can increase the weight of the item to add visual emphasis to the feedback or extend the duration.

[0084] Additionally, the processor (120) classifies various types of errors that appear in the simulation and determines the need for educational intervention for each type, thereby selecting a feedback method based on the type of error that occurs, and can be configured to provide, for example, a step sequence emphasis animation in the case of frequent sequence errors, and a repetition reminder in the case of simple operation errors.

[0085] In addition, the processor (120) analyzes user response data to past feedback to determine the tendency to accept feedback, and if repeated non-acceptance responses to the same error are confirmed, it can control the feedback to provide multi-channel outputs such as warning sounds, vibrations, and voice guidance by increasing the weight of the item.

[0086] Additionally, the processor (120) calculates the improvement rate of repetitive learning by comprehensively evaluating whether the accuracy improves when the same scenario is repeated, the error reduction rate, and the reduction of work time, and if improvement is clearly evident, it adjusts the frequency of feedback and the level of explanation to decrease, and if improvement is insufficient, it maintains or increases the number of feedback repetitions.

[0087] In addition, the processor (120) measures the fixation time by analyzing how long the user's gaze was fixed on a specific UI element or warning message within the simulation screen, and if a pattern of short fixation time or distracted attention is detected, it can increase the visual emphasis intensity and duration by giving a high weight to the item.

[0088] Finally, the processor (120) analyzes the user's response pattern immediately after the error occurs to evaluate the recovery pattern after the operation failure, and if the error repetition or abnormal behavior is confirmed, it is configured to increase the weight of the item to strengthen detailed feedback or call an auxiliary simulation module, and conversely, if appropriate recovery measures are taken, it can be adjusted to reduce feedback for the same error.

[0089] In this way, the processor (120) can be configured to provide a feedback method (difficulty level, explanation level, visual emphasis pattern, etc.) optimized for the user's current proficiency and operation pattern by analyzing the characteristics of each of the multiple user behavior elements in real time and dynamically assigning weights that reflect the relative contribution of each item. This configuration can contribute to maximizing the educational effect in the pre-maintenance stage by simultaneously promoting repetitive learning and the improvement of risk avoidance ability.

[0090] In one embodiment, the processor (120) can identify associated parts that are mechanically, electrically, or logically linked with a specific part provided in the device being manipulated during the 3D simulation.

[0091] Additionally, the processor (120) can generate 3D simulation content to enable the user to understand the operating principles of the device by activating the state of the identified associated parts in real time and simultaneously visualizing the structural connection or functional dependency relationship between the specific parts and the associated parts.

[0092] At this time, the processor (120) may assign weights to each element based on multiple judgment conditions to differentiate the level, timing, and emphasis method of dynamic correlation visualization according to the simulation operation history and learning characteristics of each user.

[0093] Specifically, when a repetitive error is detected regarding a user's operation of a specific part, the processor (120) identifies the group of associated parts of the part and adjusts the weights in such a way that the visualization priority of the part with the most actual errors is increased, thereby clearly visualizing the functional linkage structure of the part.

[0094] Additionally, the processor (120) may provide an association structure visualization for a specific group of parts with emphasis effects such as flashing, color change, and animation magnification when the user's response speed is slow or the fixation time on a specific group of parts is excessively long. This is because it allows the user to clearly recognize the operating principle when they are not paying enough attention or lack information perception. Furthermore, if the same error is not repeated after the previously provided association structure visualization, the processor (120) may gradually lower the weight assigned to the structure visualization item and reduce the output frequency or emphasis level of the item in the future to accelerate the learning entry flow.

[0095] Additionally, the processor (120) can analyze which method of voice description, animation visualization, or text-based feedback the user showed a higher level of understanding in the user response record, and adjust the visualization format of the associated part to the user's preference based on the preferred feedback format.

[0096] In addition, if an error occurs continuously for a specific group of operations (e.g., rotating part operation, power device setting, etc.) within the same scenario, the processor (120) can break down the interlocking structure of the part containing the group of operations step by step and visualize it with a preview animation.

[0097] Additionally, the processor (120) can calculate the degree of operational confusion based on the time taken from the occurrence of a user error until recovery and the number of operations attempted immediately before recovery, and for parts with a high degree of confusion, it can enhance information recognition by adjusting the number of animation repetitions and the duration of emphasis of the associated structure visualization.

[0098] Finally, the processor (120) can adjust the entire 3D simulation scenario to be placed at an optimal difficulty level by maintaining the visualization intensity of the part group while configuring the output speed and the learning path entry flow quickly, in the case where the user showed various error patterns scattered at the beginning of the simulation but the learning improvement rate for a specific part group increased rapidly.

[0099] In this way, the processor (120) can provide an immersive learning environment for the structure and operating principles of the device by applying condition-based weights according to user understanding, responsiveness, and learning patterns when configuring 3D simulation content based on dynamic correlation between parts.

[0100] FIG. 4 is a flowchart illustrating a 3D simulation method according to an embodiment of the present invention.

[0101] Referring to FIG. 4, a 3D simulation method according to one embodiment of the present invention may include the following steps S1 to S4. However, it is not limited thereto, and other general steps may be further included in the 3D simulation method.

[0102] The 3D simulation method may consist of steps processed sequentially in the aforementioned system (100). Therefore, even if the content is omitted below, the description of the system (100) above may be equally applicable to the method of providing facility simulation.

[0103] In the above S1 step, the processor (120) collects simulation data regarding the user's maintenance work input from the user terminal (200).

[0104] In the above S2 step, the processor (120) analyzes the user's real-time input, past work history, and error patterns from the simulation data to derive an analysis result that reflects the risk level during maintenance work.

[0105] In the above S3 step, the processor (120) generates 3D simulation content in which the scenario path and warning feedback level of the 3D simulation are adjusted based on the analysis results.

[0106] In the above S4 step, the processor (120) provides the generated 3D simulation content to the user terminal (200).

[0107] Meanwhile, the 3D simulation method performed by the aforementioned system (100) may be implemented in the form of a computer program stored on a computer-readable storage medium. That is, the computer program may include instructions for implementing the 3D simulation method, and the instructions of the computer program may be stored on a computer-readable storage medium. The computer program may include a mobile application.

[0108] For example, computer-readable storage media may include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical media such as CD-ROMs and DVDs; magneto-optical media such as floptical disks; and hardware devices specifically configured to store and execute computer program instructions, such as ROM, RAM, and flash memory. Computer program instructions may include machine code generated by a compiler and high-level language code that can be executed by a computer using an interpreter, etc.

[0109] As described above, although the present invention has been explained by limited embodiments and drawings, the present invention is not limited thereto, and it is obvious that various modifications and variations are possible within the scope of the technical spirit of the present invention and the equivalent scope of the claims described below by those skilled in the art to which the present invention belongs.

[0110] Meanwhile, although terms indicating directions such as up, down, left, right, front, and back have been used in this invention, these terms are used merely for convenience of explanation, and it is obvious to those skilled in the art that they may vary depending on the position of the object or the position of the observer. Explanation of the symbols

[0111] 100 : System 110 : Memory 120 : Processor 200 : User terminal

Claims

Claim 1 Memory configured to store instructions; And by executing the above commands: a processor configured to collect simulation data regarding a user's maintenance work input from a user terminal used by the user, analyze the user's real-time input, past work history, and error patterns from the simulation data to derive an analysis result reflecting the risk level during the maintenance work, generate 3D simulation content in which the scenario path and level of warning feedback of the 3D simulation are adjusted based on the analysis result, and provide the generated 3D simulation content to the user terminal; the processor is configured to accumulate and store the user's repetitive error patterns appearing in the simulation data in chronological order, and to calculate the risk level based on a first factor regarding the operation step where the error occurred, a second factor regarding the type of part during the operation, a third factor regarding the user's reaction time, a fourth factor regarding whether the user accepted previous feedback, and a fifth factor regarding the frequency of repetition of the same error, whether a warning response was given at the time of the error, and whether there was a change in behavior thereafter; and for operation items with a high risk level, when generating the 3D simulation content, the first to fifth factors according to the priority of providing warning feedback, the level of detailed explanation, the visual emphasis method, the warning intensity, and the number of repetitions A 3D simulation system configured to assign weights to each. Claim 2 delete Claim 3 In claim 1, the processor selects a scenario path of the 3D simulation from a plurality of predefined scenario paths based on the user's risk-inducing history among the analysis results, and corrects the selected scenario path of the 3D simulation by reflecting the user's error occurrence tendency and warning response history, wherein the risk-inducing history includes a high-risk operation history that occurred repeatedly in the same work step, frequent misoperations in a specific part group and the number of warnings caused thereby, the interval between errors relative to work time, the response delay time to the warning, whether there were consecutive mistakes immediately after the error occurred, the correlation with a scenario section containing a violation of regulations, a response pattern that was not improved after the warning even though a high-risk warning was output during past 3D simulations, and the frequency of entering a dangerous operation even though attention-raising feedback was provided during the overall operation flow. Claim 4 A 3D simulation system according to claim 1, wherein the processor is linked to a work standard database or safety regulation database provided by a public institution, compares the user's 3D simulation operation content shown in the simulation data with the standard provided from the work standard database or the safety regulation provided from the safety regulation database to determine whether there is a violation of the standard or safety regulation, assigns weights to the risk level, number of violations, priority of application of the regulation, importance of the operation step, and user response for each violation item, outputs a warning message, and is configured to change the scenario of the 3D simulation. Claim 5 A 3D simulation method performed by a 3D simulation system comprises: a step of collecting simulation data regarding a user's maintenance work input from a user terminal used by a user; a step of deriving an analysis result reflecting the risk level during maintenance work by analyzing the user's real-time input, past work history, and error patterns from the simulation data; and a step of generating 3D simulation content in which the scenario path and the level of warning feedback of the 3D simulation are adjusted based on the analysis result. A 3D simulation method comprising the step of providing the generated 3D simulation content to the user terminal, wherein the 3D simulation system is configured to accumulate and store the user's repetitive error patterns appearing in the simulation data in chronological order, and to calculate the risk level based on a first factor regarding the operation step where the error occurred, a second factor regarding the type of part during the operation, a third factor regarding the user's reaction time, a fourth factor regarding whether the user accepted previous feedback, and a fifth factor regarding the frequency of repetition of the same error, whether a warning response occurred at the time of the error, and whether there was a subsequent change in behavior, and for the operation item with a high risk level, to assign weights to each of the first to fifth factors according to the priority of providing warning feedback, the level of detailed explanation, the visual emphasis method, the warning intensity, and the number of repetitions when generating the 3D simulation content.

Images