Search

KR-102964432-B1 - SYSTEM FOR CALCULATING WORKER CUSTOMERIZED RISK INDEX BASED ON STANDARD RISK ASSESSMENT FOR ELECTRIC POWER INDUSTRY AUTONOMOUS SAFETY, AND METHOD OF THE SAME

KR102964432B1KR 102964432 B1KR102964432 B1KR 102964432B1KR-102964432-B1

Abstract

A system for calculating a worker-customized risk index based on a standard risk assessment for autonomous safety in the power industry, which can be efficiently applied to different workers, is disclosed. The system is characterized by including a biometric detection device that detects the biometric information of a worker working at a specific site and generates biometric information, a camera device that monitors the worker and generates image information, and a computer that calculates a worker-customized risk index (RI) for the specific site based on a standard risk assessment using the biometric information and the image information.

Inventors

  • 이동엽

Assignees

  • 한국전력공사

Dates

Publication Date
20260513
Application Date
20221028

Claims (18)

  1. A bio-sensing device (210a to 210n) that detects the biometric information of a worker working at a specific site and generates biometric information; Camera devices (220a to 220n) that monitor the above-mentioned worker and generate video information; and A computer (230) that calculates a worker-customized risk index (RI) for the specific site based on a standard risk assessment using the above biometric information and the above image information; is included. The above worker-customized risk index is a mathematical formula A worker-customized risk index calculation system for autonomous safety in the power industry, characterized by being defined as follows: (wherein RI represents a risk index, RI i represents a risk level assigned through a standard risk assessment (a value normalized between 0 and 1), Sparm j represents the j-th safety parameter capable of reducing the risk level, and Sratio j represents the weight for the j-th safety parameter).
  2. In Article 1, The above computer (230) is, A worker-customized risk index calculation system for autonomous safety in the power industry, characterized by including an artificial intelligence (AI) engine (360) implemented as a complex deep learning ensemble model that calculates a worker-customized risk index (RI) using the above biometric information and the above image information as inputs.
  3. In Article 2, The above artificial intelligence engine (360) is, Detection modules (410a to 410n) for detecting the above biological information and the above image information; and A worker-customized risk index calculation system for autonomous safety in the power industry, characterized by including: an ensemble module (420a to 420n) that analyzes the above biometric information and the above image information to identify the condition of the worker and calculates the worker's risk index from the risk level assigned through the above standard risk assessment.
  4. In Paragraph 3, The above artificial intelligence engine (360) is, A worker-customized risk index calculation system for autonomous safety in the power industry, characterized by including a notification module (430) that provides notifications related to the worker's condition based on the above risk index.
  5. In Paragraph 3, The above ensemble modules (420a to 420n) are, A plurality of AI units (510a to 510n) that individually analyze the above biometric information and generate analysis results; A judgment unit (520) that determines the corresponding step currently in progress corresponding to a standard work procedure that is pre-set based on the above analysis results, and confirms the risk level assigned to the corresponding step through a standard risk assessment; and It includes a calculation unit (530) that calculates a worker-specific risk index from the risk level using safety parameters selected from a preset worker profile, and A worker-customized risk index calculation system for autonomous safety in the power industry, characterized in that a plurality of the above-mentioned AI units (510a to 510n) utilize a voting method or a bagging method.
  6. In Article 5, A worker-customized risk index calculation system for autonomous safety in the power industry, characterized in that when the above ensemble modules (420a to 420n) are bio-ensemble modules and implemented by the voting method, a plurality of the above AI units (510a to 510n) are each designed with different types of TAD (time series anomaly detection)-GAN (generative adversarial network) and each trained with the same dataset, or when the above ensemble modules (420a to 420n) are bio-ensemble modules and implemented by the bagging method, they are each designed with the same type of TAD-GAN and each trained with different datasets.
  7. In Article 5, A worker-customized risk index calculation system for autonomous safety in the power industry, characterized in that when the above ensemble modules (420a to 420n) are bio-ensemble modules and implemented by the voting method, a plurality of the above AI units (510a to 510n) are each designed with full-subnets of different types and each trained with the same data set, or when the above ensemble modules (420a to 420n) are bio-ensemble modules and implemented by the bagging method, they are designed with full-subnets of the same type and each trained with different data sets.
  8. In Article 5, A worker-customized risk index calculation system for autonomous safety in the power industry, characterized in that when the above ensemble modules (420a to 420n) are image ensemble modules and implemented using the voting method, a plurality of the above AI units (510a to 510n) are each designed with different types of YOLO (You Only Look Once) and each trained with the same dataset, or when the above ensemble modules (420a to 420n) are image ensemble modules and implemented using the bagging method, each is designed with different types of full-subnets and each trained with the same dataset.
  9. delete
  10. (a) A process in which a bio-sensing device (210a to 210n) detects the bio of a worker working at a specific site and generates bio-information; (b) a process in which camera devices (220a to 220n) monitor the worker and generate image information; and (c) a process in which a computer (230) calculates a worker-customized risk index (RI) for a specific site based on a standard risk assessment using the biometric information and the image information; and The above worker-customized risk index is a mathematical formula A method for calculating a worker-customized risk index for autonomous safety in the power industry, characterized by being defined as follows: (wherein RI represents a risk index, RI i represents a risk level assigned through a standard risk assessment (a value normalized between 0 and 1), Sparm j represents the j-th safety parameter capable of reducing the risk level, and Sratio j represents the weight for the j-th safety parameter).
  11. In Article 10, The above (c) process is, (c-1) A process in which an artificial intelligence (AI) engine (360) is implemented as a complex deep learning ensemble model that calculates a worker-customized risk index (RI) using the above biometric information and the above image information as inputs; a method for calculating a worker-customized risk index for autonomous safety in the power industry.
  12. In Article 11, The above (c) process is, (c-2) A process in which detection modules (410a to 410n) detect the bio-information and the image information; and (c-3) A process in which an ensemble module (420a to 420n) analyzes the biometric information and the image information to determine the condition of the worker, and calculates the worker's risk index from the risk level assigned through the standard risk assessment; characterized by including a method for calculating a worker-customized risk index for autonomous safety in the power industry.
  13. In Article 12, The above (c) process is, (c-4) A method for calculating a worker-customized risk index for autonomous safety in the power industry, characterized by including a process in which a notification module (430) provides a notification related to the worker's condition based on the risk index.
  14. In Article 12, The above (c-3) process is, A process in which a plurality of AI units (510a to 510n) individually analyze the bio-information to generate analysis results; A process in which a judgment unit (520) determines the corresponding step currently in operation corresponding to a standard work procedure that is pre-set based on the analysis results, and confirms the risk level assigned to the corresponding step through a standard risk assessment; and The process includes a calculation unit (530) calculating a worker-specific risk index from the risk level using safety parameters selected from a preset worker profile; A method for calculating a worker-customized risk index for autonomous safety in the power industry, characterized in that a plurality of the above-mentioned AI units (510a to 510n) use a voting method or a bagging method.
  15. In Article 14, A method for calculating a worker-customized risk index for autonomous safety in the power industry, characterized in that when the above ensemble modules (420a to 420n) are bio-ensemble modules and implemented by the voting method, a plurality of the above AI units (510a to 510n) are each designed with different types of TAD (time series anomaly detection)-GAN (generative adversarial network) and each are trained with the same dataset, or when the above ensemble modules (420a to 420n) are bio-ensemble modules and implemented by the bagging method, they are each designed with the same type of TAD-GAN and each are trained with different datasets.
  16. In Article 14, A method for calculating a worker-customized risk index for autonomous safety in the power industry, characterized in that when the above ensemble modules (420a to 420n) are bio-ensemble modules and implemented by the voting method, a plurality of the above AI units (510a to 510n) are each designed with full-subnets of different types and each trained with the same data set, or when the above ensemble modules (420a to 420n) are bio-ensemble modules and implemented by the bagging method, they are designed with full-subnets of the same type and each trained with different data sets.
  17. In Article 14, A method for calculating a worker-customized risk index for autonomous safety in the power industry, characterized in that when the above ensemble modules (420a to 420n) are image ensemble modules and implemented by the voting method, a plurality of the above AI units (510a to 510n) are each designed with different types of YOLO (You Only Look Once) and each trained with the same dataset, or when the above ensemble modules (420a to 420n) are image ensemble modules and implemented by the bagging method, each is designed with different types of full-subnets and each trained with the same dataset.
  18. delete

Description

System for Calculating Worker Customized Risk Index Based on Standard Risk Assessment for Electric Power Industry Autonomous Safety and Method thereof The present invention relates to a risk assessment technology for autonomous safety in the power industry, and more specifically, to a system and method for calculating a worker-customized risk index based on a standard (e.g., ISO 45001) risk assessment for autonomous safety in the power industry. Generally, work related to the power industry is carried out at specific sites, and such work may involve risks. Consequently, a standard risk assessment may be conducted for such work. Examples of standard risk assessments include ISO 45001 risk assessment, which represents an international standard occupational safety and health management system. Specifically, safety and health experts can assign a degree of risk to each step by individually evaluating the steps of the standard operating procedure (SOP) related to the work. However, since the workers actually carrying out the work differ in terms of skill level and physical condition, there is a problem in that applying the same level of risk to different workers is not efficient. FIG. 1 is a drawing for exemplarily illustrating a standard risk assessment according to an embodiment of the present invention. FIG. 2 is a schematic block diagram of a system according to the disclosure of an embodiment of the present invention. Figure 3 is a detailed block diagram of the computer shown in Figure 2. Figure 4 is a detailed block diagram of the artificial intelligence (AI) engine illustrated in Figure 3. Figure 5 is a detailed configuration block diagram of the ensemble module shown in Figure 4. FIGS. 6 and 7 are drawings illustrating an exemplary concept of a bio-ensemble implemented through voting or bagging according to an embodiment of the present invention. FIGS. 8 and 9 are drawings illustrating an exemplary bio-ensemble concept implemented through voting or bagging according to an embodiment of the present invention. FIG. 10 is a diagram exemplarily illustrating an image ensemble concept implemented through voting according to an embodiment of the present invention. FIG. 11 is a drawing that exemplarily illustrates an image ensemble concept implemented through bagging according to an embodiment of the present invention. FIG. 12 is a flowchart showing the process of calculating a risk index for a worker and generating a notification using a risk level according to an embodiment of the present invention. The present invention is capable of various modifications and may have various embodiments, and specific embodiments are illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the invention to specific embodiments, and it should be understood that the invention includes all modifications, equivalents, and substitutions that fall within the spirit and scope of the invention. When describing each drawing, similar reference numerals are used for similar components. Terms such as first, second, etc., may be used to describe various components, but said components should not be limited by said terms. Such terms are used solely for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be named the second component, and similarly, the second component may be named the first component. The term "and/or" includes a combination of a plurality of related described items or any of a plurality of related described items. Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as generally understood by those skilled in the art to which this invention pertains. Terms such as those defined in commonly used dictionaries should be interpreted as having meanings consistent with their meanings in the context of the relevant technology, and should not be interpreted in an ideal or overly formal sense unless explicitly defined in this application. A worker-customized risk index calculation system and method based on standard risk assessment for autonomous safety in the power industry according to an embodiment of the present invention will be described in detail below with reference to the attached drawings. FIG. 1 is a drawing illustrating an exemplary standard risk assessment according to an embodiment of the present invention. Referring to FIG. 1, generally, work related to the power industry is carried out at a specific site, and there may be risks involved in such work. As a result, a standard risk assessment for the relevant work can be conducted. Standard risk assessments include, for example, ISO 45001 risk assessment, which can represent an international standard occupational safety and health management system. Specifically, safety and health experts can assign a risk level to each step by individually evalua