CN-121971048-A - Exoskeleton health monitoring and fatigue early warning method and system for high-altitude operation

CN121971048ACN 121971048 ACN121971048 ACN 121971048ACN-121971048-A

Abstract

The invention discloses an exoskeleton health monitoring and fatigue early warning method and system for aloft work, which relate to the technical field of exoskeleton control and comprise the steps of collecting parameters of operators and body surface temperature data in real time through an exoskeleton integrated sensor; the method comprises the steps of carrying out noise reduction and feature extraction pretreatment on acquired data, constructing a model based on multidimensional features, evaluating muscle fatigue, joint load and cardiovascular state, generating health state scores and fatigue accumulation values, dynamically adjusting exoskeleton assistance modes and output forces according to evaluation results, setting a safety threshold and a grading early warning mechanism, sending out prompt or limiting high-strength actions when triggering early warning, and recording related data. According to the invention, through dynamic linkage of the health state and the power-assisted control, the wearing comfort and the operation persistence are improved, the high-altitude falling risk is reduced by the grading early warning mechanism, and double health and safety guarantees are provided for high-altitude operation such as ultra-high voltage power overhaul.

Inventors

WANG SHENLI
DONG QIONG
MAO ZHU
WANG MAN
HAN HAO
DU YONG
WU JUN
SHI YI
HU LONGJIANG
ZHANG SHANHE
JIA XIANG
LI BAISONG

Assignees

国网湖北省电力有限公司超高压公司
湖北省超能电力有限责任公司

Dates

Publication Date: 20260505
Application Date: 20251223

Claims (10)

1. An exoskeleton health monitoring and fatigue early warning method for aloft work is characterized by comprising the following steps: the method comprises the steps that through an exoskeleton integrated biological sensor and a motion sensor, heart rate, electromyographic signals, joint moment, gait parameters and body surface temperature data of an operator are collected in real time, and all the data are synchronously stored in an exoskeleton local cache; Noise reduction processing is carried out on the collected original data, high-frequency interference is removed from heart rate data by adopting moving average filtering, time domain root mean square and frequency domain center frequency characteristics are extracted from electromyographic signals through wavelet transformation, joint moment data are subjected to Kalman filtering smoothing processing, cycle characteristics and variation coefficients are extracted from gait parameters through time sequence analysis, and trend extraction is carried out on body surface temperature data so as to eliminate environmental temperature interference; constructing a health state evaluation model based on the preprocessed characteristic data, judging cardiovascular load through heart rate variability, evaluating muscle fatigue by utilizing the characteristic attenuation degree of the electromyographic signal frequency domain, analyzing joint load by combining joint moment peak value and duration, and judging exercise coordination by fusing gait parameter variation coefficients to generate health state score and fatigue accumulation value; according to the health status score and the fatigue accumulation value, dynamically adjusting a power-assisted output strategy of the exoskeleton, increasing the power-assisted proportion of corresponding muscle groups when the muscle fatigue is high, reducing the moment output of the joint when the joint load exceeds a threshold value, and switching to a low-strength power-assisted mode when the cardiovascular load is abnormal; setting a health state safety threshold and a fatigue early warning threshold, when the health state score is lower than the safety threshold or the fatigue accumulation value is higher than the early warning threshold, sending out visual early warning through the exoskeleton wearable display screen, sending out voice prompt through the bone conduction earphone, or triggering the exoskeleton semi-locking state to limit high-strength action, and simultaneously recording all data at the early warning moment for subsequent analysis.
2. The exoskeleton health monitoring and fatigue early warning method for high-voltage operation according to claim 1, further comprising a signal anti-interference processing step in a high-voltage environment, wherein for an electromagnetic interference environment of ultra-high voltage power operation, adaptive notch filter processing is adopted for electromyographic signals and joint moment signals, interference frequencies are automatically identified through analysis of signal frequency spectrum characteristics, interference components are specifically removed through adjustment of notch filter parameters, a differential amplification circuit is adopted to enhance the signal-to-noise ratio of the electromyographic signals, and the joint moment signals are transmitted through a shielding cable and grounded.
3. The exoskeleton health monitoring and fatigue early warning method for high-altitude operation according to claim 1, further comprising a health index baseline calibration step, wherein heart rate, myoelectric signal baseline value and joint moment zero drift value of an operator in a resting state are collected before operation, a personalized baseline database is established according to age, weight and past operation history of the operator, all health indexes are compared and analyzed with corresponding baseline values in the operation process, evaluation deviation caused by individual difference is eliminated, the baseline values are collected and updated once every 3 months, and manual triggering calibration flow is supported when the physical state of the operator changes.
4. The method for monitoring the health of an exoskeleton and early warning fatigue for aloft work according to claim 1, wherein a fatigue accumulation assessment model is introduced in the fatigue state assessment step, and the fatigue accumulation value is calculated by weighting a plurality of dimension indexes, wherein the calculation formula is as follows Wherein In order to obtain the value of the fatigue accumulation, For the heart rate load weight, As the weight of myoelectric fatigue, For the weight of the joint load, For the operation duration weight, the sum of the four is 1, For a real-time heart rate, As a resting heart rate baseline value, As an upper limit for the safe heart rate, For the real-time myoelectric signal frequency domain center frequency, For the initial myoelectric signal frequency domain center frequency, For the average moment of the joint, For the joint safety moment threshold value, For the duration of the continuous operation, Is a safe continuous operation time.
5. The exoskeleton health monitoring and fatigue early warning method for high-altitude operation according to claim 1, wherein an abnormal gait recognition step is added in gait parameter analysis, abnormal gait such as lameness and unbalance is recognized by calculating the gait cycle variation coefficient, step symmetry and deviation of the support phase time ratio, when the gait cycle variation coefficient exceeds 15%, the step symmetry deviation exceeds 20% or the support phase time ratio deviates from the normal range by more than 10%, abnormal gait is determined, and the system automatically reduces the power output of the corresponding lower limb joint and enhances balance assistance to give out gait adjustment prompt.
6. The exoskeleton health monitoring and fatigue early warning method for high-altitude operation according to claim 1, wherein a multi-feature fusion analysis step is adopted in myoelectric signal processing, five feature parameters including time domain root mean square, peak factor, waveform factor, frequency domain average power frequency and center frequency of the myoelectric signal are extracted, the myoelectric fatigue feature value is obtained through weighted summation, weight distribution is set according to the functional importance of a muscle group, the weights of biceps brachii and triceps brachii in an upper limb muscle group are higher than those of other muscles, the weights of quadriceps femoris and popliteal cord muscle in a lower limb muscle group are higher than those of other muscles, and multi-feature fusion reduces single feature misjudgment.
7. The method for exoskeleton health monitoring and fatigue early warning for aloft work according to claim 1, wherein a dynamic assistance output formula is introduced in the adaptive assistance adjustment step, and an assistance coefficient is dynamically calculated according to a health status score and a joint load, wherein the calculation formula is Wherein For the dynamic power-assisting coefficient, Is used as a basic power-assisted coefficient, The value range is 0-1 for the standardized grading of the health status, As the health status influencing factor, For the joint load compensation factor, And (3) with The sum is 0.8-1.0, For the moment of the joint in real time, And the power output is dynamically matched with the health state and the joint load for the joint safety moment threshold.
8. The exoskeleton health monitoring and fatigue early warning method for aloft work according to claim 1, further comprising a multi-mode data fusion step, wherein evaluation results of heart rate, myoelectricity, joint moment and gait parameters are fused by adopting a D-S evidence theory, a health evaluation result of heart rate data is used as a first evidence source, a fatigue evaluation result of myoelectricity is used as a second evidence source, a load evaluation result of joint moment is used as a third evidence source, a coordination evaluation result of gait parameters is used as a fourth evidence source, reliability factors of all evidence sources are set, and a final evaluation result is obtained through evidence combination rules.
9. The exoskeleton health monitoring and fatigue early warning method for the high-altitude operation according to claim 1 is characterized in that a hierarchical early warning mechanism is adopted in the early warning prompting step, three-level early warning is divided according to fatigue accumulation values and health state scores, the first-level early warning corresponds to the fatigue accumulation values of 0.6-0.8 or the health state scores of 0.7-0.8, only voice prompts are sent out to suggest proper rest, the second-level early warning corresponds to the fatigue accumulation values of 0.8-0.9 or the health state scores of 0.6-0.7, voice and vision dual early warning is sent out, meanwhile exoskeleton boosting strength is reduced by 10% -20%, the third-level early warning corresponds to the fatigue accumulation values exceeding 0.9 or the health state scores being lower than 0.6, strong acousto-optic early warning is sent out, the exoskeleton limits high-strength actions and prompts to stop operation immediately, and thresholds of different early warning grades can be dynamically adjusted according to operation difficulty.
10. An overhead operation-oriented exoskeleton health monitoring and fatigue warning system for implementing the overhead operation-oriented exoskeleton health monitoring and fatigue warning method as set forth in any one of claims 1 to 9, comprising: the multi-dimensional data acquisition module is integrated with a biological sensor and a motion sensor, wherein the biological sensor comprises a heart rate sensor, a myoelectric sensor and a body surface temperature sensor, and is respectively used for acquiring heart rate, key muscle group myoelectric signals and body surface temperature data of an operator in real time; The system comprises a signal preprocessing and feature extraction module, a time domain and frequency domain analysis module, a time sequence analysis module and a time sequence analysis module, wherein the signal preprocessing and feature extraction module is used for performing noise reduction and feature extraction operation on collected original data, heart rate data adopts moving average filtering, electromyographic signals extract time domain and frequency domain features through wavelet transformation, joint moment data are smoothed through Kalman filtering, gait parameters extract periodic features and variation coefficients through time sequence analysis, and body surface temperature data are subjected to trend extraction to eliminate environmental interference; The health and fatigue evaluation module is internally provided with a health state evaluation model and a fatigue accumulation evaluation model, analyzes cardiovascular load through heart rate variability, evaluates muscle fatigue based on electromyographic signal frequency domain characteristic attenuation, analyzes joint load in combination with joint moment peak value and duration, and fuses gait parameter variation coefficients to judge motion coordination, so as to generate health state scores and fatigue accumulation values, and supports multi-mode data fusion and abnormal gait recognition; The self-adaptive power-assisted control module is communicated with the health and fatigue evaluation module, dynamically adjusts an exoskeleton power-assisted output strategy according to the health status score and the fatigue accumulation value, comprises targeted adjustment of power-assisted proportion, output moment and response speed, and is internally provided with a dynamic power-assisted coefficient calculation unit to realize real-time matching of power-assisted output with the health status of a human body and joint load; The early warning and intervention module is used for setting a health state safety threshold and a fatigue early warning threshold, configuring a wearable display screen, a bone conduction earphone and a joint half-locking unit, triggering visual early warning, voice prompt or high-strength action limitation according to early warning grades, and synchronously recording full-quantity data at early warning time; The auxiliary functional module comprises a high-voltage anti-interference unit, a health index baseline calibration unit and a post-operation health compound disc unit, wherein the high-voltage anti-interference unit guarantees data acquisition accuracy in a high-voltage environment through self-adaptive notch filtering and shielding transmission, the baseline calibration unit establishes a personalized baseline database and supports periodic updating, and the health compound disc unit automatically generates a health state and fatigue change report and transmits the health state and fatigue change report to the management platform.

Description

Exoskeleton health monitoring and fatigue early warning method and system for high-altitude operation Technical Field The invention relates to the technical field of exoskeleton robot control, in particular to an exoskeleton health monitoring and fatigue early warning method and system for high-altitude operation. Background With the rapid development of the power industry, the maintenance and overhaul work of the ultra-high voltage transmission line is more frequent, and the high-altitude operation, in particular to the climbing of an iron tower, the replacement of an insulator, the high-altitude operation and other power overhaul scenes, has extremely high requirements on the physical ability and safety of operators. The existing exoskeleton technology focuses on the sport assistance function, achieves the labor-saving effect by optimizing joint driving, track planning and the like, but generally ignores the health state monitoring and fatigue early warning requirements of operators in overhead operation, and fails to fuse health protection with operation safety. The exoskeleton often adjusts the power-assisted mode only according to the motion data, lacks real-time perception of the physiological state of a human body, and is difficult to predict misoperation or high-altitude falling risk caused by muscle fatigue and cardiovascular overload. The electric power overhaul type high-altitude operation has remarkable specificity that operators need to carry overhaul tools to climb and stay on complex structures such as iron towers for a long time, muscles are continuously in a high-load state, joints repeatedly bear impact loads, and meanwhile physical energy consumption is further accelerated due to high concentration of spirit in a high-pressure environment. The data show that more than 60% of safety accidents of high-altitude operation are related to action instability caused by physical overdrawing and muscle fatigue of operators. However, most of the existing health monitoring devices are independently worn on a bracelet, a chest strap and the like, only basic heart rate data can be provided, the basic heart rate data cannot be linked with an exoskeleton system, the data dimension is single and pertinence is lacking, dynamic changes of muscle fatigue and joint load in the operation process cannot be accurately reflected, and an emergency adjustment or early warning mechanism of the exoskeleton cannot be timely triggered when risks occur. Meanwhile, the prior exoskeleton technology combining the health monitoring has obvious limitations that on one hand, the monitoring data are limited to single physiological indexes, motion data such as joint moment and gait characteristics are not fused, so that fatigue evaluation and load judgment are not accurate enough, on the other hand, the early warning mechanism is mainly a passive prompt, dynamic linkage with a power assisting mode is lacking, output force cannot be optimized in real time according to the health state, and even human body load is possibly aggravated due to improper power assisting. In addition, the problem of signal interference in the high-voltage electromagnetic environment is not effectively solved, so that the accuracy of health data acquisition is insufficient, and the monitoring and early warning effects are further affected. These defects make the prior art difficult to meet the double requirements of health protection and operation safety in high-altitude operation scenes such as electric power overhaul and the like, and cannot fundamentally reduce the occupational injury risk caused by fatigue. Disclosure of Invention The exoskeleton health monitoring and fatigue early warning method for the overhead operation provided by the invention aims to solve the problems in the prior art. In order to achieve the purpose, the invention adopts the following technical scheme that the exoskeleton health monitoring and fatigue early warning method for high-altitude operation comprises the following steps: A multidimensional physiological and motion data acquisition step, namely acquiring heart rate, electromyographic signals, joint moment, gait parameters and body surface temperature data of an operator in real time through an exoskeleton integrated biological sensor and a motion sensor, wherein the heart rate sampling frequency is not lower than 50Hz, the electromyographic signals are acquired to cover key muscle groups such as upper limb biceps, triceps brachii and quadriceps of lower limb, popliteal and the like, the joint moment is acquired for four joints of shoulder, hip, knee and ankle, the gait parameters comprise step frequency, step length and support phase ratio, and all the data are synchronously stored in an exoskeleton local cache; A signal preprocessing and feature extraction step, namely carrying out noise reduction processing on the collected original data, removing high-frequency interference on heart rate data by adopting moving av