A system and method for early detection and warning of fire hazards

Abstract

The present invention relates to a system for detecting signs of fire in advance and alerting to danger. It comprises a Distributed Temperature Sensing (DTS)-based temperature sensing unit configured to collect location and temperature data, and an AI model unit configured to perform AI-based data analysis based on the collected data. The AI ​​model unit includes a data collection and preprocessing module configured to receive location and temperature data from the DTS-based temperature sensing unit, collect the location and temperature data, and perform data preprocessing for analysis preparation; an anomaly detection and pattern analysis module configured to receive data prepared for analysis from the data collection and preprocessing module, detect abnormal situations, and analyze temperature change patterns; a predictive modeling and risk assessment module configured to perform a warning function according to the risk level based on the analysis results of the anomaly detection and pattern analysis module; and an alarm and automated response module configured to report the risk assessment results to detect signs of fire early and send a pre-warning based on the risk assessment results by the predictive modeling and risk assessment module. Accordingly, through the convergence of DTS and AI, initial signs of fire occurrence can be detected in real time, and damage caused by fire can be minimized by providing a warning in advance.

Description

Technology Field

[0001] The present invention relates to a real-time fire pre-warning system combining a Distributed Temperature Sensing (DTS)-based temperature sensing system with Artificial Intelligence (AI). In particular, the present invention relates to a system that rapidly detects early signs of fire using fiber optic sensing technology and detects abnormal patterns in real time through AI-based data analysis to provide a pre-warning of risk factors before a fire occurs. Background Technology

[0002] Conventional fire detection systems are designed to react only after a fire occurs, primarily using smoke or high-temperature detectors, which has resulted in a problem of failing to effectively recognize early signs of fire. DTS is a technology capable of detecting temperature changes using optical fibers, allowing for real-time monitoring of temperature distribution over wide areas. However, DTS alone has limitations in accurately assessing fire risk under various temperature fluctuation conditions. Accordingly, there is a need for a system that utilizes AI technology to analyze temperature data collected by DTS, thereby predicting and alerting on the potential for fire in advance. The problem to be solved

[0003] This invention has been devised to solve the aforementioned problems. It proposes a method to minimize damage caused by fire by detecting early signs of fire occurrence in real time and providing advance warnings through the fusion of DTS and AI. Through this, the invention aims to enable a rapid response to and prevention of fires that may occur in large-scale facilities or hazardous areas.

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

[0005] To solve the above problem, a fire sign pre-detection and risk warning system according to one embodiment of the present invention comprises a DTS-based temperature detection unit configured to collect location and temperature data, and an AI model unit configured to perform AI-based data analysis based on the collected data. The AI ​​model unit comprises a data collection and pre-processing module configured to receive location and temperature data from the DTS-based temperature detection unit, collect the location and temperature data, and perform data pre-processing for analysis preparation; an anomaly detection and pattern analysis module configured to receive data prepared for analysis from the data collection and pre-processing module, detect anomalies, and analyze temperature change patterns; a predictive modeling and risk assessment module configured to perform a warning function according to the risk level based on the analysis results of the anomaly detection and pattern analysis module; and a warning and automated response module configured to report the risk assessment results to detect fire signs early and send a pre-warning based on the risk assessment results by the predictive modeling and risk assessment module.

[0006] In addition, a fire sign pre-detection and risk warning method, wherein each step is performed by a computer-implemented fire sign pre-detection and risk warning system according to one embodiment of the present invention, comprises: a DTS-based data integration detection step in which location and temperature data are collected and environmental factor data are integrated and utilized for risk analysis in a DTS-based temperature detection unit; a data integration and preprocessing step in which location, temperature, and environmental data are input from the DTS-based temperature detection unit and data preprocessing is performed for preparation for analysis in a data collection and preprocessing module of an AI model unit; an anomaly detection and multidimensional pattern analysis step in which abnormal patterns and predictability are derived and risk grades are assigned using the integrated data in an anomaly detection and pattern analysis module of an AI model unit; a prediction modeling and risk assessment step in which a warning function is performed according to the risk level based on the analysis results of the anomaly detection and multidimensional pattern analysis module in a prediction modeling and risk assessment module of an AI model unit; and an alarm and automated response step in which a risk assessment result is reported to detect fire signs early and send a pre-warning in a warning and automated response module of an AI model unit based on the risk assessment results by the prediction modeling and risk assessment module. Effects of the invention

[0007] According to the system and method for detecting signs of fire in advance and warning of danger according to the present invention, there is an advantage in that it can overcome the limitations of existing fire detection systems and establish a system that prevents the occurrence of fire by predicting signs of fire in advance.

[0008] In particular, by combining DTS technology capable of wide-ranging temperature detection with AI technology strong in pattern recognition, fire hazards can be detected early, which has the advantage of minimizing casualties and property loss.

[0009] More specifically, DTS technology utilizes long optical fibers to detect temperature changes over wide areas with high resolution, while AI technology analyzes this data in real time to detect abnormal patterns. The combination of these two technologies enables the prediction of fire risk with high accuracy and maximizes response time through early warning. Brief explanation of the drawing

[0010] FIG. 1 is a configuration diagram showing a fire sign pre-detection and danger warning system that fuses a DTS-based temperature sensor and AI according to an embodiment of the present invention. FIG. 2 is a flowchart illustrating a method for pre-detection of fire signs and danger warning by fusing a DTS-based temperature sensor and AI according to an embodiment of the present invention. FIG. 3 is a flowchart illustrating an AI fire sign detection and response case according to the first embodiment of the present invention. FIG. 4 is a flowchart illustrating an AI fire sign detection and response case according to a second embodiment of the present invention. FIG. 5 is a flowchart illustrating an AI fire sign detection and response case according to a third embodiment of the present invention. FIG. 6 is a flowchart illustrating an AI fire sign detection and response case according to the fourth embodiment of the present invention. FIG. 7 is a flowchart illustrating an AI fire sign detection and response case according to the fifth embodiment of the present invention. Specific details for implementing the invention

[0011] The fire sign pre-detection and danger warning system and method incorporating a DTS-based temperature sensor and AI according to the present invention will be described in detail below with reference to the attached drawings. The drawings presented below are provided as examples to ensure that the concept of the present invention is sufficiently conveyed to those skilled in the art. Accordingly, the present invention is not limited to the drawings presented below and may be embodied in other forms. Furthermore, throughout the specification, the same reference numerals indicate the same components.

[0012] Unless otherwise defined, technical and scientific terms used herein have the meaning commonly understood by those skilled in the art to which this invention pertains, and descriptions of known functions and configurations that could unnecessarily obscure the essence of the invention are omitted in the following description and accompanying drawings.

[0013] Furthermore, a system refers to a set of components, including devices, mechanisms, and means, that are organized and interact regularly to perform necessary functions.

[0014] A fire sign pre-detection and danger warning system and method according to one embodiment of the present invention relates to a technology for reducing property and human casualties by fusion of fiber optic temperature sensing and AI-based analysis to detect initial signs of fire in real time and provide early warnings.

[0015] FIG. 1 is a configuration diagram showing a fire sign pre-detection and danger warning system combining a DTS-based temperature sensor and an AI analysis module according to an embodiment of the present invention. With reference to FIG. 1, the fire sign pre-detection and danger warning system according to an embodiment of the present invention will be described in detail.

[0016] A fire sign pre-detection and danger warning system according to one embodiment of the present invention is preferably composed of a DTS-based temperature detection unit (101) and an AI model unit (102), as illustrated in FIG. 1. Additionally, the AI ​​model unit (102) is preferably composed of a data collection and preprocessing module (110), an anomaly detection and pattern analysis module (120), a prediction modeling and risk assessment module (130), a continuous learning and improvement module (140), a spatial information module (150), and an alarm and automated response module (160). Each of these components is preferably configured to perform operations by being individually or integrally included in at least one computational processing means including a computer.

[0017] The following describes each component in detail.

[0018] The data collection and preprocessing module (110) receives location and temperature data to be preprocessed from the DTS-based temperature sensing unit (101). The data collection and preprocessing module (110) is further composed of a real-time data collection module (111) that collects location and temperature data in real time to prepare for analysis, and a data normalization and filtering module (112) that removes unnecessary noise from the collected data and normalizes it to complete the preparation for analysis.

[0019] The above anomaly detection and pattern analysis module (120) receives data that has been prepared for analysis through the data normalization and filtering module (112), detects abnormal situations, and analyzes temperature change patterns. The above anomaly detection and pattern analysis module (120) is composed of a standard pattern learning module (121) that can set a standard for recognizing abnormal situations by learning normal temperature change patterns in more detail, and a real-time anomaly detection module (122) that monitors current data in real time and detects abnormal situations by comparing it with the standard pattern.

[0020] The above prediction modeling and risk assessment module (130) is a module for performing a warning function according to the level of risk, and more specifically, it is composed of a time series prediction modeling module (131) that predicts future risks based on current temperature change patterns and analyzes the possibility of potential fire occurrence, and a risk scoring and classification module (132) that assigns a score according to the level of risk and classifies according to the assigned score to determine the urgency of the warning.

[0021] The above continuous learning and improvement module (140) is composed of a real-time learning feedback module (141) for enabling the AI ​​model to continuously learn and improve performance through verified data, more specifically for the continuous learning and improvement of the AI ​​model, and a data update and retraining module (142) for improving the performance of the AI ​​model according to new data and increasing accuracy through real-time learning.

[0022] The spatial information module (150) performs spatial information mapping to visualize detected temperature and anomaly information, thereby enabling spatial identification and analysis of risk points. Through this spatial information mapping of the spatial information module (150), risk points can be identified more quickly and efficiently.

[0023] The above alarm and automation response module (160) performs the function of reporting risk assessment results so that fire signs can be detected early and a pre-alarm can be sent through the risk assessment results of the AI ​​model by the above prediction modeling and risk assessment module (130).

[0024] The fire sign pre-detection and risk warning system of the present invention according to the above-described configuration can be utilized, for example, in a logistics warehouse. In the logistics warehouse, DTS sensors installed in various zones continuously collect temperature data, and when an AI analysis module detects a rapid temperature rise in a specific zone, it sends a real-time fire sign pre-warning notification to the manager of that zone. This system according to the present invention helps to predict the possibility of fire signs occurring within the logistics warehouse at an early stage and enables the rapid implementation of response measures. The operation scenario of the system according to the present invention in a logistics warehouse will be described in more detail below.

[0025] FIG. 2 is a flowchart illustrating a method for pre-detection of fire signs and danger warning by fusing a DTS-based temperature sensor and AI according to an embodiment of the present invention. With reference to FIG. 2, a method for pre-detection of fire signs and danger warning according to an embodiment of the present invention will be described in detail.

[0026] A fire sign pre-detection and risk warning method according to one embodiment of the present invention is preferably configured to include, as illustrated in FIG. 2, a DTS-based data integration detection step (S210), a data integration and preprocessing step (S220), an anomaly detection and multidimensional pattern analysis step (S230), a predictive modeling and risk assessment step (S240), a continuous learning and improvement step (S250), a spatial information mapping step (S260), and a warning and automated response step (S270). Additionally, each step is performed by a computer-implemented specific image search-based de-identification processing system.

[0027] Each step is described in detail below.

[0028] In the DTS-based data integration detection step (S210), location and temperature data are collected from the DTS-based temperature detection unit (101), environmental factor data (e.g., humidity, etc.) is integrated and utilized for risk analysis, and the integrated data is transmitted to the data collection and preprocessing module (110).

[0029] In the above data integration and preprocessing step (S220), the integrated data to be preprocessed is input through the real-time data collection module (111) of the data collection and preprocessing module (110) to prepare for analysis, and unnecessary noise in the integrated data is removed and the data is normalized by the data normalization and filtering module (112), and preparation for multidimensional analysis is performed. That is, the data collected through the real-time data collection module (111) is refined by undergoing data normalization, noise removal, and filtering processes through the data normalization and filtering module (112).

[0030] In the above anomaly detection and pattern analysis step (S230), data prepared for analysis through the data normalization and filtering module (112) is input, abnormal patterns and predictability are derived, and a risk grade is assigned. At this time, the standard pattern learning module (121) learns the normal temperature change pattern to set a standard for recognizing abnormal situations, and the real-time anomaly detection module (122) monitors the current data in real time and detects abnormal situations by comparing it with the standard pattern. That is, the data refined through the data normalization and filtering module (112) is compared with a normal pattern through the standard pattern learning module (121), and abnormal signs are detected by detecting abnormal patterns through the real-time anomaly detection module (122).

[0031] In the above prediction modeling and risk assessment step (S240), a warning function is performed according to the risk level. At this time, the time series prediction modeling module (131) predicts future risks based on current temperature change patterns to analyze the potential possibility of fire occurrence, and the risk scoring and classification module (132) assigns a score according to the risk level and classifies it according to the assigned score to determine the urgency of the warning. That is, when an abnormal sign is detected through the real-time anomaly detection module (122), the risk is evaluated through the time series prediction modeling module (131) and the risk scoring and classification module (132).

[0032] The above continuous learning and improvement step (S250) is a step for the continuous learning and improvement of the AI ​​model, enabling the AI ​​model to continuously learn and improve performance through data verified by the real-time learning feedback module (141), and to improve the performance of the AI ​​model according to new data and increase accuracy through real-time learning by the data update and retraining module (142). In other words, the AI ​​model can be continuously improved through the real-time learning feedback module (141) and the data update and retraining module (142).

[0033] In the spatial information mapping step (S260) above, spatial information mapping is performed to visualize detected temperature and anomaly information so that risk points can be spatially identified and analyzed. Through this spatial mapping, risk points can be identified more quickly and efficiently.

[0034] In the above alarm and automated response step (S270), the risk assessment results are reported through the risk assessment results of the AI ​​model by the prediction modeling and risk assessment module (130) so that signs of fire can be detected early and a pre-alarm can be sent. At this time, if the risk assessment result indicates a high risk level, a notification is sent through the alarm system of the alarm and automated response module (160), and if necessary, the automatic action system of the alarm and automated response module (160) is activated to respond to the emergency situation.

[0035] Through this, the fire sign pre-detection and risk warning system and method according to one embodiment of the present invention has the advantage of overcoming the limitations of existing fire detection systems and establishing a system that prevents fire occurrence by predicting fire signs in advance. In particular, by combining DTS technology capable of wide-ranging temperature detection with AI technology strong in pattern recognition, fire risk factors can be detected early, thereby minimizing casualties and property loss.

[0036] In the following, with reference to FIGS. 3 to 7, an example of how an AI model assigns a risk grade score to each case and issues an alarm to respond to a fire is explained according to the analysis case of temperature data received from the DTS-based temperature detection unit (101). In light of the step of receiving and analyzing temperature data from the DTS-based temperature detection unit (101), five cases of temperature data analysis may occur as described below.

[0037] FIG. 3 is a flowchart illustrating an AI fire sign detection and response case according to the first embodiment of the present invention. Referring to FIG. 3 as an example of an abnormal temperature rise pattern detection according to the first embodiment of the present invention, an abnormal temperature rise pattern occurs (310), and a rapid temperature rise compared to normal occurs at a specific location, and it is detected that the temperature has risen by more than 3°C in one minute (320). When detected in this way, the risk scoring and classification module (132) scores the risk level as high risk (80 / 100) (330). In this case, since there is a high probability that there is a fire sign at the location, a warning notification transmission function is performed to immediately conduct an on-site inspection (340).

[0038] FIG. 4 is a flowchart illustrating an AI fire sign detection and response case according to a second embodiment of the present invention. Referring to FIG. 4 as an example of an abnormal temperature rise pattern detection according to the second embodiment of the present invention, a localized temperature rise and movement pattern occurs (410), and it is detected that a high-temperature point starting in Zone A spreads to adjacent Zones B and C within 5 minutes (420). When detected in this way, the risk scoring and classification module (132) scores the risk level as very dangerous (90 / 100) (430). In this case, the automatic firefighting system is switched to a ready state, and a function to transmit an emergency situation notification to the field inspection team is performed (440).

[0039] FIG. 5 is a flowchart illustrating an AI fire sign detection and response case according to a third embodiment of the present invention. Referring to FIG. 5 as an example of an abnormal temperature rise pattern detection according to the third embodiment of the present invention, multiple abnormal patterns occur simultaneously (510), and a high temperature rise is detected in Zone A and a low concentration of smoke is detected in Zone D (520). When detected in this way, the risk scoring and classification module (132) scores the risk level as a strong fire sign (85 / 100) (530). In this case, the automatic sprinkler is switched to a ready state, and a warning notification transmission function is performed to instruct the manager to conduct an immediate on-site inspection (540).

[0040] FIG. 6 is a flowchart illustrating an AI fire sign detection and response case according to the fourth embodiment of the present invention. Referring to FIG. 6 as an example of an abnormal temperature rise pattern detection according to the fourth embodiment of the present invention, a pattern of rapid temperature rise followed by rapid temperature drop occurs (610), and it is detected that the temperature rises by 10°C within 10 seconds and then immediately returns to the original temperature (620). When detected in this way, the risk scoring and classification module (132) scores the risk score as medium risk (50 / 100) (630). In this case, since the risk is not high, an on-site inspection is recommended, but emergency measures are unnecessary. Therefore, the function of periodically monitoring and observing the progress is performed (640).

[0041] FIG. 7 is a flowchart illustrating an AI fire sign detection and response case according to the fifth embodiment of the present invention. Referring to FIG. 7 as an example of an abnormal temperature rise pattern detection according to the fifth embodiment of the present invention, a continuous temperature rise pattern occurs (710), and the temperature is gradually rising for 30 minutes, and the current temperature is detected to be 5°C higher than the reference value (720). When detected in this way, the risk scoring and classification module (132) scores the risk score as "caution needed" (70 / 100) (730). In this case, the rate of temperature rise is confirmed through additional monitoring, and a warning notification transmission function is performed if the situation deteriorates (740).

[0042] The fire sign pre-detection and risk warning system integrating a DTS-based temperature sensor and AI according to one embodiment of the present invention as described above can be installed in an industrial logistics warehouse to detect temperature change patterns in real time and automatically transmit a warning to an administrator upon detection of abnormal signs. Since logistics warehouses are environments where damage is significant in the event of a fire due to their large space, high volume of stored goods, and high-density storage conditions, the present invention can be effectively utilized.

[0043] More specifically, a fire sign pre-detection and risk warning system fused with a DTS-based temperature sensor and AI according to one embodiment of the present invention can be utilized to detect fire risks that may occur in large-scale logistics warehouses in advance, and the operation scenario is as follows.

[0044] - DTS sensors are installed in key temperature-changing zones within the logistics warehouse (e.g., electrical equipment rooms, machinery areas, stored chemical areas, etc.) to detect temperature changes in real time.

[0045] - When the temperature of the electric panel in a specific area rises above the reference temperature (e.g., 60℃) due to an electrical overload, the DTS sensor transmits the corresponding data.

[0046] The AI-based analysis system monitors DTS sensor data in real time to detect abnormal temperature patterns. For example, if a rapid temperature rise is detected in a specific area, the AI ​​system assigns a risk score of 80 and identifies it as an early sign of fire.

[0047] - In cases of high risk (e.g., when the rate of temperature rise and changes in the surrounding environment exceed a threshold), the alarm system activates immediately and sends a notification to the logistics warehouse manager and safety officer.

[0048] According to the aforementioned operating scenario, upon receiving an alarm notification, measures can be taken to automatically shut off power to the hazardous area and simultaneously activate cooling systems, such as sprinklers within the warehouse, to prevent initial spread. Additionally, the logistics warehouse manager can monitor the situation in real time via a mobile dashboard and connect with the fire department to direct a rapid fire suppression response.

[0049] Through the measures described above, potential fires in large-scale logistics warehouses can be detected in advance, thereby minimizing property and human casualties. In particular, measures based on temperature rise patterns and risk scores enable efficient fire prevention and rapid response. Furthermore, there is the advantage of being able to provide customized fire detection and response solutions tailored to the contents and environment within the warehouse.

[0050] In addition, the fire sign pre-detection and danger warning system fused with a DTS-based temperature sensor and AI according to one embodiment of the present invention as described above can be installed in an electric vehicle charging area within an underground parking lot and utilized to detect a temperature rise caused by electrical abnormalities at an early stage and to warn of the possibility of a fire in advance.

[0051] The method described above may be provided as a computer program stored on a computer-readable recording medium for execution on a computer. The medium may continuously store a computer-executable program, or temporarily store it for execution or download. Additionally, the medium may be various recording or storage means in the form of a single or multiple hardware components, and may not be limited to a medium directly connected to a computer system but may exist distributed over a network. Examples of media may include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical recording media such as CD-ROMs and DVDs; magneto-optical media such as floptical disks; and media configured to store program instructions, including ROM, RAM, and flash memory. Furthermore, other examples of media may include recording or storage media managed by app stores that distribute applications or sites and servers that supply or distribute various other software.

[0052] The methods, operations, or techniques of the present disclosure may be implemented by various means. For example, these techniques may be implemented in hardware, firmware, software, or a combination thereof. Those skilled in the art will understand that the various exemplary algorithmic steps described in connection with the disclosure herein may be implemented in electronic hardware, computer software, or a combination of both. To clearly explain such interchangeability between hardware and software, various exemplary steps have been generally described above in terms of their functional aspects. Whether such functions are implemented in hardware or in software depends on the design requirements imposed on the specific application and the overall system. Those skilled in the art may implement the described functions in various ways for each specific application, but such implementations should not be construed as departing from the scope of the present disclosure.

[0053] In a hardware implementation, the processing unit used to perform the techniques may be implemented in one or more ASICs, DSPs, digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, electronic devices, other electronic units designed to perform the functions described in this disclosure, computers, or combinations thereof.

[0054] Accordingly, the various exemplary steps described in connection with the present disclosure may be implemented or performed by any combination of general-purpose processors, DSPs, ASICs, FPGAs or other programmable logic devices, discrete gate or transistor logic, discrete hardware components, or those designed to perform the functions described herein. A general-purpose processor may be a microprocessor, but alternatively, the processor may be any conventional processor, controller, microcontroller, or state machine. The processor may also be implemented as a combination of computing devices, for example, a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors coupled with a DSP core, or any other combination of configurations.

[0055] In firmware and / or software implementations, techniques may be implemented as instructions stored on a computer-readable medium such as random access memory (RAM), read-only memory (ROM), non-volatile random access memory (NVRAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable PROM (EEPROM), flash memory, compact disc (CD), magnetic or optical data storage devices, etc. The instructions may be executable by one or more processors, and may cause the processor(s) to perform specific aspects of the functions described in this disclosure.

[0056] Where implemented in software, the steps may be stored on a computer-readable medium as one or more instructions or code, or transmitted through a computer-readable medium. Computer-readable media include both computer storage media and communication media, including any medium that facilitates the transmission of a computer program from one place to another. Storage media may be any available media accessible by a computer. As a non-limiting example, such computer-readable media may include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium accessible by a computer that can be used to transfer or store desired program code in the form of instructions or data structures. Additionally, any connection is appropriately referred to as a computer-readable medium.

[0057] For example, if software is transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair cable, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, coaxial cable, fiber optic cable, twisted pair cable, digital subscriber line, or wireless technologies such as infrared, radio, and microwave are included within the definition of a medium. As used herein, disk and disc include CD, laser disc, optical disc, DVD (digital versatile disc), floppy disk, and Blu-ray disc, wherein disks usually play data magnetically, whereas discs play data optically using a laser. The above combinations should also be included within the scope of computer-readable media.

[0058] The software module may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, a hard disk, a removable disk, a CD-ROM, or any other known form of storage medium. An exemplary storage medium may be connected to a processor so that the processor can read information from the storage medium or write information to the storage medium. Alternatively, the storage medium may be integrated into the processor. The processor and the storage medium may exist within an ASIC. The ASIC may exist within a user terminal. Alternatively, the processor and the storage medium may exist as separate components within the user terminal.

[0059] Although the embodiments described above have been described as utilizing aspects of the subject matter disclosed herein in one or more standalone computer systems, the present disclosure is not limited thereto and may be implemented in conjunction with any computing environment, such as a network or a distributed computing environment. Furthermore, aspects of the subject matter in the present disclosure may be implemented in a plurality of processing chips or devices, and storage may be similarly affected across a plurality of devices. Such devices may include PCs, network servers, and portable devices.

[0060] The embodiments according to the present invention are not limited to those described above, and may be implemented with various alternatives, modifications, and changes within the scope obvious to those skilled in the art in relation to the present invention. Explanation of the symbols

[0061] 101: DTS-based temperature sensor 102: AI Model Department 110: Data Collection and Preprocessing Module 111: Real-time data collection module 112: Data normalization and filtering module (112) 120: Anomaly Detection and Pattern Analysis Module 121: Standard Pattern Learning Module 122: Real-time Anomaly Detection Module 130: Predictive Modeling and Risk Assessment Module 131: Time Series Forecasting Modeling Module 132: Risk Scoring and Classification Module 140: Continuous Learning and Improvement Module 141: Real-time Learning Feedback Module 142: Data Update and Retraining Module 150: Spatial Information Module 160: Alarm and Automation Response Module

Claims

Claim 1 A system for detecting signs of fire in advance and warning of danger comprises: a Distributed Temperature Sensing (DTS) based temperature sensing unit (101) configured to collect location and temperature data; and an AI model unit (102) configured to perform AI-based data analysis based on the collected data, wherein the AI ​​model unit (102) A data collection and preprocessing module (110) configured to receive location and temperature data from the DTS-based temperature sensing unit (101), collect the location and temperature data, and perform data preprocessing for preparation for analysis; An anomaly detection and pattern analysis module (120) configured to receive data ready for analysis from the above data collection and preprocessing module (110), detect anomalies, and analyze temperature change patterns; A predictive modeling and risk assessment module (130) configured to perform a warning function according to the risk level based on the analysis results of the above-mentioned anomaly detection and pattern analysis module (120); and A fire sign pre-detection and risk warning system comprising an alarm and automated response module (160) configured to report risk assessment results to detect fire signs early and send a pre-warning based on the risk assessment results by the above-mentioned predictive modeling and risk assessment module (130). Claim 2 A fire sign pre-detection and danger warning system according to claim 1, further comprising a continuous learning and improvement module (140) configured to perform learning in real time through verified data and new data for the continuous learning and improvement of the AI ​​model unit (102). Claim 3 A fire sign pre-detection and danger alarm system according to claim 1, further comprising a spatial information module (150) configured to perform spatial information mapping to enable spatial identification and analysis of danger points by visualizing detected temperature and abnormal information. Claim 4 A fire sign pre-detection and danger warning system according to claim 1, wherein the data collection and preprocessing module (110) comprises: a real-time data collection module (111) that collects location and temperature data in real time and prepares for analysis; and a data normalization and filtering module (112) that removes unnecessary noise from the collected data and normalizes it to complete the preparation for analysis. Claim 5 A fire sign pre-detection and danger warning system according to claim 1, wherein the anomaly detection and pattern analysis module (120) comprises: a reference pattern learning module (121) capable of learning normal temperature change patterns to set criteria for recognizing an abnormal situation; and a real-time anomaly detection module (122) that monitors current data in real time and detects an abnormal situation by comparing it with a reference pattern. Claim 6 A fire sign pre-detection and risk warning system according to claim 1, wherein the prediction modeling and risk assessment module (130) comprises: a time series prediction modeling module (131) that predicts future risks based on current temperature change patterns and analyzes the possibility of potential fire occurrence; and a risk scoring and classification module (132) that assigns scores according to risk levels and classifies according to assigned scores to determine the urgency of the warning. Claim 7 In claim 2, the continuous learning and improvement module (140) is characterized by being composed of: a real-time learning feedback module (141) for enabling the AI ​​model to continuously learn and improve performance through verified data; and a data update and retraining module (142) for improving the performance of the AI ​​model according to new data and increasing accuracy through real-time learning, in a fire sign pre-detection and danger warning system. Claim 8 A fire sign pre-detection and risk alarm system according to claim 1, wherein the alarm and automated response module (160) is also configured to perform a case-by-case response function that sets an alarm for each risk level based on fire sign detection and response cases and performs an appropriate response for the corresponding risk level. Claim 9 A method for fire sign pre-detection and risk warning, wherein each step is performed by a computer-implemented fire sign pre-detection and risk warning system, comprising: a DTS-based data integration detection step (S201) in which a DTS-based temperature detection unit collects location and temperature data and integrates environmental factor data to utilize for risk analysis; a data integration and preprocessing step (S202) in which a data collection and preprocessing module of an AI model unit receives location, temperature, and environmental factor data from the DTS-based temperature detection unit and performs data preprocessing for preparation for analysis; an anomaly detection and multidimensional pattern analysis step (S203) in which an anomaly detection and pattern analysis module of the AI ​​model unit derives abnormal patterns and predictability and assigns a risk grade using the integrated data; and a predictive modeling and risk assessment step (S204) in which a predictive modeling and risk assessment module of the AI ​​model unit performs a warning function according to the risk level based on the analysis results of the anomaly detection and multidimensional pattern analysis module. A method for early detection of fire signs and risk warning, comprising an alarm and automation response step (S207) in the alarm and automation response module of the AI ​​model unit, which reports a risk assessment result to detect fire signs early and send a pre-warning based on the risk assessment result by the prediction modeling and risk assessment module. Claim 10 A method for pre-detection of fire signs and danger warning, wherein, in the continuous learning and improvement module of the AI ​​model unit, the continuous learning and improvement step (S205) for performing learning in real time through verified data and new data for the continuous learning and improvement of the AI ​​model unit. Claim 11 A method for pre-detection of fire signs and danger warning according to claim 9, further comprising a spatial information mapping step (S206) in which spatial information mapping is performed in the spatial information module of the AI ​​model part to visualize detected temperature and abnormal information so as to spatially identify and analyze danger points. Claim 12 A method for pre-detection of fire signs and danger warning according to claim 9, wherein the data integration and preprocessing step (S202) comprises: a step of collecting location and temperature data in real time and preparing for analysis in a real-time data collection module of the data collection and preprocessing module; and a step of removing unnecessary noise from the collected data, normalizing the data, and preparing for multidimensional analysis in a data normalization and filtering module of the data collection and preprocessing module. Claim 13 In claim 9, the anomaly detection and multidimensional pattern analysis step (S203) is characterized by comprising: a step of setting a standard for recognizing an abnormal situation by learning a normal temperature change pattern in a standard pattern learning module of the anomaly detection and pattern analysis module; and a step of detecting an abnormal situation by monitoring current data in real time and comparing it with a standard pattern in a real-time anomaly detection module of the anomaly detection and pattern analysis module. Claim 14 In claim 9, the prediction modeling and risk assessment step (S204) is characterized by comprising: a step of analyzing the possibility of a potential fire occurrence by predicting future risks based on current temperature change patterns in the time series prediction modeling module of the prediction modeling and risk assessment module; and a step of determining the urgency of a warning by assigning a score according to the risk level and classifying according to the assigned score in the risk scoring and classification module of the prediction modeling and risk assessment module. Claim 15 A method for pre-detection of fire signs and danger warning, wherein, in claim 10, the continuous learning and improvement step (S205) comprises: a step of enabling an AI model to continuously learn and improve performance through verified data in a real-time learning feedback module of the continuous learning and improvement module; and a step of improving the performance of the AI ​​model according to new data and increasing accuracy through real-time learning in a data update and retraining module of the continuous learning and improvement module. Claim 16 In claim 9, the alarm and automated response step (S207) is characterized by performing a case-specific response function that sets an alarm suitable for each risk level based on fire sign detection and response cases, and performs an appropriate response suitable for the corresponding risk level. Claim 17 A computer-readable recording medium storing a computer program for executing a method according to any one of claims 9 through 16 on a computer system.

Images