CN-121987169-A - Human body sign evaluation method and device under high-temperature high-altitude operation environment, terminal equipment and storage medium

CN121987169ACN 121987169 ACN121987169 ACN 121987169ACN-121987169-A

Abstract

The invention discloses a human body sign evaluation method and device under a high-temperature high-altitude operation environment, terminal equipment and a storage medium, and belongs to the technical field of risk identification. The method comprises the steps of extracting time domain features, frequency domain features and time-space features according to motion sensing data of a person to be monitored, inputting the time domain features, the frequency domain features and the time-space features into a preset residual error network, identifying operation action types to obtain action classification results, calculating to obtain physiological load indexes according to maximum heart rate, current body temperature data and current heart rate data in acquired physical examination data, carrying out Gaussian process regression calculation according to the motion sensing data, the acquired environment temperature data and altitude data to obtain heat stress risk scores, and carrying out weighted evaluation according to the action classification results, arrhythmia probabilities obtained according to analysis of the physical examination data, the physiological load indexes, the heat stress risk scores, the motion sensing data, the environment temperature data and the altitude data to obtain health risk grades.

Inventors

LU SHASHA
You Linnan
HE PENGJUN
LI HUALIANG
SHEN YALI
YANG WENHAN
WANG QIRU
LI LINYONG
WANG YIFAN
DENG WEN
Wang yalong

Assignees

广东电网有限责任公司电力科学研究院
广东电网有限责任公司

Dates

Publication Date: 20260508
Application Date: 20260320

Claims (10)

1. A method for human body sign assessment in a high temperature high altitude operation environment, comprising: acquiring motion sensing data, current body temperature data, current heart rate data and physical examination data of a person to be monitored, and environment temperature data and altitude data of the current environment; Performing feature extraction according to the motion sensing data to obtain a time domain feature for representing the sudden change of the instantaneous amplitude of the motion, a frequency domain feature for representing the difference of motion mode frequencies, and a space-time feature for representing the continuity of the motion, and inputting the time domain feature, the frequency domain feature and the space-time feature into a preset residual error network to perform operation motion type recognition to obtain a motion classification result; Performing time sequence feature analysis on the physical examination data based on a preset time sequence convolution network to obtain arrhythmia probability, and calculating to obtain a physiological load index according to the maximum heart rate, the current body temperature data and the current heart rate data in the physical examination data; Carrying out Gaussian process regression calculation according to the motion sensing data, the environmental temperature data and the altitude data to obtain a heat stress risk score; And carrying out comprehensive weighted evaluation according to the action classification result, the arrhythmia probability, the physiological load index, the heat stress risk score, the motion sensing data, the environmental temperature data and the elevation data to obtain the health risk level of the personnel to be monitored.
2. The method for evaluating the physical sign of a human body in a high-temperature high-altitude operation environment according to claim 1, wherein the motion sensing data comprises triaxial acceleration data and angular velocity data; Extracting features according to the motion sensing data to obtain a time domain feature for representing the sudden change of the instantaneous amplitude of the motion, a frequency domain feature for representing the difference of the frequency components of the motion mode, and a space-time feature for representing the continuity of the motion, wherein the feature extraction comprises the following steps: Extracting the mean value, variance and peak value of the triaxial acceleration data and the angular velocity data according to the motion sensing data, and taking the mean value, variance and peak value as time domain features for representing the instantaneous amplitude mutation of the motion; extracting a main frequency component of the motion sensing data based on fast Fourier transform, and taking the main frequency component as a frequency domain characteristic for representing a difference of motion mode frequency composition; Modeling the continuous acquisition sequence of the motion sensing data based on a preset long-short-term memory network to obtain space-time characteristics for representing motion continuity.
3. The method for evaluating human body signs in a high-temperature high-altitude operation environment according to claim 2, wherein the action classification result comprises climbing, walking, falling and standing; Inputting the time domain features, the frequency domain features and the space-time features into a preset residual error network for operation action type identification to obtain action classification results, wherein the operation classification results comprise: Inputting the time domain features, the frequency domain features and the space-time features into a preset residual error network for dimensional alignment and feature fusion processing to obtain a fusion feature sequence; The method comprises the steps of carrying out feature extraction according to a fusion feature sequence to obtain an action deep layer distinguishing feature, wherein the action deep layer distinguishing feature comprises a vertical motion dimension feature, a time domain dimension distinguishing feature, a frequency domain dimension distinguishing feature and a time-space sequence dimension association feature, the vertical motion dimension feature comprises an acceleration fluctuation interval, an amplitude change rate, a duration and a gravity component duty ratio in the vertical direction, the time domain dimension distinguishing feature comprises a mutation gradient of an action amplitude, a peak duration, a variance change trend and a mean value offset, the frequency domain dimension distinguishing feature comprises a main frequency distribution interval, a frequency spectrum energy concentration degree, a harmonic duty ratio and a frequency band energy change rate of an action signal, and the time-space sequence dimension association feature comprises a time sequence dependency relationship, an action state conversion feature and a continuous action duration of a continuous action sequence; And identifying the action type according to the action depth discriminating characteristics to obtain an action classification result.
4. The method for evaluating the physical sign of a human body in a high-temperature high-altitude working environment according to claim 1, wherein the calculating the physiological load index according to the maximum heart rate in the physical examination data, the current body temperature data and the current heart rate data comprises: carrying out normalization processing according to the maximum heart rate, the current body temperature data and the current heart rate data in the physical examination data, taking the ratio of the current heart rate data after normalization processing to the maximum heart rate as a heart rate relative load rate, and taking the ratio of the current body temperature data after normalization processing to a preset normal body temperature as a body temperature relative load rate; And carrying out weighted summation calculation according to the heart rate relative load rate, the body temperature relative load rate and preset weights to obtain a physiological load index.
5. The method for evaluating human body signs in a high-temperature high-altitude operation environment according to claim 4, wherein performing gaussian process regression calculation according to the motion sensing data, the ambient temperature data and the altitude data to obtain a heat stress risk score comprises: Performing thermal drift compensation according to the triaxial acceleration data in the motion sensing data and the ambient temperature data to obtain compensated triaxial acceleration data; and carrying out nonlinear fitting and regression calculation according to the environmental temperature data, the elevation data and the compensated triaxial acceleration data, and obtaining a heat stress risk score for representing the personnel to be monitored in the current working environment and the motion state by fitting the mapping relation of the environmental temperature data, the elevation data and the motion state with the human heat stress level.
6. The method for evaluating human body signs in a high-temperature high-altitude operation environment according to claim 5, wherein the comprehensively weighted evaluation is performed according to the action classification result, the arrhythmia probability, the physiological load index, the heat stress risk score, the environmental temperature data and the altitude data to obtain a health risk level of a person to be monitored, comprising: determining corresponding allocation weights according to the action classification result, the environmental temperature data and the elevation data; performing falling risk node assessment according to the action classification result and the weightlessness characteristic of the triaxial acceleration data of the motion sensing data to obtain a falling risk index; performing heat stress risk node assessment according to the heat stress risk score and the physiological load index to obtain a heat stress risk index; Carrying out cardiovascular risk node assessment according to the arrhythmia probability and the heart rate relative load rate in the physiological load index to obtain a cardiovascular risk index; Performing musculoskeletal injury risk assessment according to the climbing action duration and the acceleration variance of the motion sensing data in the action classification result to obtain musculoskeletal risk indexes; evaluating according to the falling risk index, the heat stress risk index, the cardiovascular risk index, the musculoskeletal risk index and the assigned weight to obtain a comprehensive evaluation index; Judging according to the comprehensive evaluation index and the operation tolerance threshold of the personnel to be monitored, and determining the health risk level; And under the condition that the motion classification result is falling and the vertical motion distance in the motion sensing data exceeds a preset vertical threshold, directly judging that the health risk level is the highest level.
7. The method for evaluating human body signs in a high-temperature high-altitude operation environment according to claim 1, further comprising triggering safety alarm information of a corresponding level according to the health risk level, and transmitting the safety alarm information to a monitoring center in real time.
8. A human body sign evaluation device in a high-temperature high-altitude operation environment, comprising: The to-be-detected data acquisition module is used for acquiring motion sensing data, current body temperature data, current heart rate data and physical examination data of a to-be-monitored person, and environment temperature data and altitude data of the current environment; The feature extraction module is used for carrying out feature extraction according to the motion sensing data to obtain a time domain feature used for representing the sudden change of the instantaneous amplitude of the motion, a frequency domain feature used for representing the difference of motion mode frequencies and a space-time feature used for representing the continuity of the motion, and inputting the time domain feature, the frequency domain feature and the space-time feature into a preset residual error network to carry out operation motion type recognition to obtain a motion classification result; The physiological index calculation module is used for carrying out time sequence feature analysis on the physical examination data based on a preset time sequence convolution network to obtain arrhythmia probability, and calculating to obtain a physiological load index according to the maximum heart rate, the current body temperature data and the current heart rate data in the physical examination data; The heat stress risk assessment module is used for carrying out Gaussian process regression calculation according to the motion sensing data, the environmental temperature data and the elevation data to obtain a heat stress risk score; And the health risk level evaluation module is used for comprehensively weighting and evaluating according to the action classification result, the arrhythmia probability, the physiological load index, the heat stress risk score, the motion sensing data, the environment temperature data and the elevation data to obtain the health risk level of the personnel to be monitored.
9. A terminal device comprising a processor, a memory and a computer program stored in the memory and configured to be executed by the processor, the processor implementing a human body physical sign assessment method under a high-temperature high-altitude operation environment according to any one of claims 1 to 7 when executing the computer program.
10. A computer readable storage medium, characterized in that the computer readable storage medium comprises a stored computer program, wherein the computer program, when run, controls a device in which the computer readable storage medium is located to perform a human body physical sign assessment method in a high-temperature high-altitude operation environment according to any one of claims 1 to 7.

Description

Human body sign evaluation method and device under high-temperature high-altitude operation environment, terminal equipment and storage medium Technical Field The present invention relates to the field of risk identification technologies, and in particular, to a method and apparatus for evaluating a human body sign in a high-temperature high-altitude operation environment, a terminal device, and a storage medium. Background The operation and maintenance operation of the power grid is a core link for guaranteeing the stable operation of the power system, the outdoor operation scene of power grid operators is often accompanied with complex and severe operation conditions such as high temperature, high altitude and the like, the physiological state of the operators can be obviously influenced under the environment, health problems are easy to cause, and even safety accidents such as high altitude falling and operation errors are possibly caused by abnormal signs, so that the real-time and accurate assessment and monitoring of the power grid operators in the high-temperature and high-altitude operation environment are key requirements for the safety production of the power industry, and the safety of workers is very important. At present, human body characteristics in an indoor calm state are mainly collected for health sign monitoring and evaluation of operators, or the normal outdoor operation scene of a low-altitude area is only adapted, no specific design is made for the high-temperature high-frightening environment of power grid operation, the collected sign data are prone to error, the evaluation result output by a model has larger deviation from the actual physiological state of the operators, and effective sign evaluation of the operators in the high-temperature high-altitude environment cannot be realized. Disclosure of Invention The embodiment of the invention provides a human body sign evaluation method, device, terminal equipment and storage medium in a high-temperature high-altitude operation environment, which can effectively solve the problem that the prior art mainly collects human body features in an indoor calm state or is only suitable for a common outdoor operation scene in a low-altitude area, and cannot realize effective sign evaluation of operators in the high-temperature high-altitude environment. An embodiment of the present invention provides a human body sign evaluation method in a high-temperature high-altitude operation environment, including: acquiring motion sensing data, current body temperature data, current heart rate data and physical examination data of a person to be monitored, and environment temperature data and altitude data of the current environment; Performing feature extraction according to the motion sensing data to obtain a time domain feature for representing the sudden change of the instantaneous amplitude of the motion, a frequency domain feature for representing the difference of motion mode frequencies, and a space-time feature for representing the continuity of the motion, and inputting the time domain feature, the frequency domain feature and the space-time feature into a preset residual error network to perform operation motion type recognition to obtain a motion classification result; Performing time sequence feature analysis on the physical examination data based on a preset time sequence convolution network to obtain arrhythmia probability, and calculating to obtain a physiological load index according to the maximum heart rate, the current body temperature data and the current heart rate data in the physical examination data; Carrying out Gaussian process regression calculation according to the motion sensing data, the environmental temperature data and the altitude data to obtain a heat stress risk score; And carrying out comprehensive weighted evaluation according to the action classification result, the arrhythmia probability, the physiological load index, the heat stress risk score, the motion sensing data, the environmental temperature data and the elevation data to obtain the health risk level of the personnel to be monitored. Further, the motion sensing data includes triaxial acceleration data and angular velocity data; Extracting features according to the motion sensing data to obtain a time domain feature for representing the sudden change of the instantaneous amplitude of the motion, a frequency domain feature for representing the difference of the frequency components of the motion mode, and a space-time feature for representing the continuity of the motion, wherein the feature extraction comprises the following steps: Extracting the mean value, variance and peak value of the triaxial acceleration data and the angular velocity data according to the motion sensing data, and taking the mean value, variance and peak value as time domain features for representing the instantaneous amplitude mutation of the motion; extracting a main frequency compo