Fatigue index screening system and integration method thereof in LoRa monitoring terminal

Abstract

The invention discloses a fatigue index screening system and an integration method thereof in a LoRa monitoring terminal. According to the method, comprehensive and accurate evaluation of the fatigue state of the operator is realized by constructing a multi-dimensional index system and a scientific screening method. The system integrates multi-dimensional indexes such as electrocardio, heart rate, respiratory rate, skin temperature, three-axis posture, exercise intensity score and physical state six-minute walking test, can be used for real-time monitoring, high-altitude noninvasive, individual difference correction, synchronism verification, scene adaptive adjustment and Spearman correlation analysis by combining a five-step deletion screening method, eliminates indexes with strong interference or lag, and can be used for real-time monitoring, high-altitude noninvasive, individual difference correction, synchronism verification, scene adaptive adjustment and Spearman correlation analysis. Core indexes highly related to the fatigue development trend are reserved, it is ensured that all-dimensional characteristics covering physiological reaction, psychological perception and behavioral performance of operators are evaluated, the accuracy of fatigue state judgment is remarkably improved, stable operation can still be achieved in a strong electromagnetic field environment based on the LoRa technology and anti-interference design, and the reliability of data transmission is effectively guaranteed.

Description

Technical Field

[0001] This invention belongs to the field of power grid transmission operation safety technology, specifically a fatigue index screening system and its integration method in a LoRa monitoring terminal. Background Technology

[0002] In high-altitude power transmission operations, the assessment and management of worker fatigue is a crucial aspect of ensuring operational safety. With the increasing workload of power grid construction and maintenance, the fatigue risks faced by workers are becoming increasingly prominent. Currently, research in the field of human factors safety management in power systems is still in its early stages, with only a small amount of intelligent personal protective equipment technology development. The environment for high-altitude power transmission operations is complex, with problems such as strong electromagnetic interference and poor signal coverage in remote areas. Furthermore, the work is demanding and prolonged, meaning that worker fatigue directly impacts operational safety and efficiency. Therefore, there is an urgent need for a system and method capable of accurately screening fatigue indicators and achieving real-time monitoring to improve operational safety management.

[0003] However, existing fatigue assessment technologies have the following main problems: Limited indicator selection: Many models only use one or a few indicators such as heart rate and skin conductance, which cannot comprehensively reflect the fatigue status of workers. For example, the existing general-purpose model only includes two indicators, heart rate and skin conductance, which is difficult to cope with complex working conditions and individual differences.

[0004] Insufficient anti-interference capability: In strong electromagnetic field environments, data transmission is easily interfered with, leading to inaccurate monitoring results.

[0005] Limited transmission distance: Traditional monitoring equipment has a short data transmission distance, which cannot meet the needs of long-distance monitoring in high-altitude power transmission operations.

[0006] Low integration: The fatigue index screening system and the monitoring terminal are not well integrated, making it difficult to achieve real-time and accurate fatigue monitoring and early warning. Summary of the Invention

[0007] The purpose of this invention is to provide a fatigue index screening system and its integration method in a LoRa monitoring terminal in order to solve the problems mentioned above.

[0008] The technical solution adopted in this invention is as follows: a fatigue index screening system, comprising: a multi-dimensional data acquisition module, a dynamic adaptive index screening module, an intelligent weight calculation and optimization module, a fatigue assessment model construction module, a LoRa communication and anti-interference module, and a real-time early warning and feedback module; The modules in the fatigue index screening system are closely connected through data flow: The output of the multi-dimensional data acquisition module is directly connected to the input of the dynamic adaptive index filtering module; The output of the dynamic adaptive index screening module is connected to the input of the intelligent weight calculation and optimization module; The output of the intelligent weight calculation and optimization module is connected to the input of the fatigue assessment model construction module; The output of the fatigue assessment model construction module is connected to the input of the LoRa communication and anti-interference module; The output of the LoRa communication and anti-interference module is connected to the input of the real-time early warning and feedback module; The output of the real-time early warning and feedback module can be connected in reverse to the input of the LoRa communication and anti-interference module to transmit the adjustment instructions from the ground control terminal back to the front-end device, forming a closed-loop connection for data interaction.

[0009] In a preferred embodiment, the multi-dimensional data acquisition module covers the physiological, psychological, behavioral, and environmental data of personnel working at heights in power transmission lines, providing multi-source data support for subsequent analysis; The multi-dimensional data acquisition module includes: Physiological indicator acquisition module: integrates PPG sensor, GSR sensor, surface electromyography sensor, accelerometer, etc., to realize non-invasive, real-time physiological signal acquisition; Biochemical indicator acquisition module: Through a wearable microfluidic chip, it detects biochemical indicators such as electrocardiogram, heart rate, respiratory rate, skin temperature, three-axis posture, and exercise intensity, solving the limitation of not being able to draw blood during high-altitude operations; Subjective psychological data collection module: Integrates a lightweight mobile questionnaire system, supports quick completion during work breaks, and combines voice interaction function to reduce the burden of filling out questionnaires; Environmental and operational data acquisition module: Records operational scene characteristics through the terminal's built-in electromagnetic field sensor, temperature and humidity sensor, and operational type tags.

[0010] In a preferred embodiment, the dynamic adaptive index screening module comprises a real-time monitoring screening unit, a high-altitude non-invasive screening unit, an individual difference correction unit, a synchronization verification unit, and a scene adaptive adjustment unit. The real-time monitoring screening unit, based on the equipment's acquisition capabilities, excludes indicators requiring invasive detection or high-frequency sampling costs, retaining only physiological indicators collected in real time by wearable devices. The high-altitude non-invasive screening unit, targeting high-altitude work scenarios, eliminates indicators requiring contact sampling or interfering with the work, retaining only indicators that can be collected by surface sensors or microfluidic chips. The individual difference correction unit uses historical data to establish an individual baseline model, filtering indicators with large group fluctuations and retaining indicators with strong consistency. The synchronization verification unit uses timestamp alignment technology to screen indicators that change synchronously with the fatigue development trend. The scene adaptive adjustment unit combines environmental sensors and work type labels to identify the current scene and dynamically adjusts the screening thresholds of the first four steps. For example, in strong electromagnetic field scenarios, indicators with strong anti-interference are preferentially retained, and in high temperature and high humidity scenarios, the weight of skin temperature indicators is increased.

[0011] In a preferred embodiment, the dynamic adaptive index screening module is based on a dual screening logic of "five-step funnel method + scenario adaptive mechanism" to solve the problem that traditional screening cannot adapt to complex working conditions, specifically including the following steps: Step 1: Real-time monitoring and screening: Exclude indicators that require invasive testing or have high costs for high-frequency sampling, and retain only indicators that can be collected in real time by wearable devices; Step 2: High-altitude non-invasive screening: Eliminate indicators that require contact sampling or interfere with the operation, and retain only indicators that can be collected by surface sensors or microfluidic chips; Step 3: Individual difference correction: Establish an "individual baseline model" through historical data, automatically filter out indicators with large individual fluctuations, and retain only indicators with strong group consistency; Step 4: Synchronization verification: Based on timestamp alignment technology, screen indicators that change synchronously with the fatigue development trend and exclude lagging or leading indicators. Step 5: Adaptive adjustment of the scene: Introduce a scene perception module to dynamically adjust the screening thresholds of the first four steps; for example, in the "strong electromagnetic field scene", prioritize the retention of the skin conductance index with strong anti-interference ability; in the "high temperature and high humidity scene", increase the screening weight of the skin temperature index.

[0012] In a preferred embodiment, the intelligent weight calculation and optimization module consists of an initial weight calculation submodule and a dynamic optimization submodule. The initial weight calculation submodule combines domain expert experience with Spearman correlation analysis to generate an initial weight matrix using a weighted scoring method. Expert experience is used to set the sensitivity benchmark for the indicators, and Spearman analysis calculates the monotonic correlation between the indicators and the scores of the subjective fatigue scale. The dynamic optimization submodule is based on machine learning algorithms such as random forests to train on historical sample data and automatically correct the weight values. If the correlation between a certain indicator and fatigue changes significantly in a specific scenario, its weight is adjusted accordingly to ensure that the weight allocation matches the actual working conditions.

[0013] In a preferred embodiment, the fatigue assessment model construction module includes a basic assessment layer, a scenario correction layer, and a grading standard setting layer. The basic assessment layer integrates physiological, biochemical, psychological, and physical fitness test indicators, and calculates the initial fatigue value through a linear weighted formula, in which the indicator values ​​and weights correspond one-to-one. The scenario correction layer introduces an environmental risk coefficient and a task difficulty coefficient. The environmental risk coefficient is determined based on environmental data such as electromagnetic fields, temperature, and humidity, while the task difficulty coefficient is set according to the task type. Both adjust the initial fatigue value together. The grading standard setting layer divides the threshold into three levels: safety level, warning level, and danger level, based on the sample data of 36 portrait sets from the simulated climbing test, each corresponding to different fatigue state judgment standards.

[0014] In a preferred embodiment, the LoRa communication and anti-interference module consists of a hardware communication submodule and a software anti-interference submodule. The hardware communication submodule uses a LoRa wireless communication chip to support Mesh networking, integrates adaptive spread spectrum technology to adjust the transmission rate according to the interference intensity, and optimizes the spiral anti-interference antenna design to improve the stability of long-distance transmission. The software anti-interference submodule reduces the data loss rate through forward error correction coding, uses frequency hopping technology to switch the operating frequency band every 5 seconds to avoid strong electromagnetic field interference, and combines CRC cyclic redundancy check to ensure the integrity of transmitted data, jointly ensuring reliable data transmission in complex environments.

[0015] In a preferred embodiment, the real-time early warning and feedback module consists of a terminal early warning component and a ground feedback component. The terminal early warning component has a built-in vibration motor, LED light, and buzzer. It triggers different early warning modes according to the fatigue level. The dangerous level triggers high-frequency vibration, red light, and buzzer, while the early warning level triggers low-frequency vibration and yellow light. The ground feedback component pushes fatigue value, early warning level, and key indicators to the ground control terminal in real time via a 4G / 5G network. This allows management personnel to view the fatigue heat map of the workers and remotely send rest instructions or adjust work tasks, realizing two-way interaction between the front end and the management end.

[0016] In a preferred embodiment, a method for integrating a fatigue index screening system into a LoRa monitoring terminal includes the following steps: S1: Hardware integration; integrates physiological information acquisition devices such as PPG, GSR, and accelerometer; adopts LoRa wireless communication chip to support long-distance transmission and Mesh networking; optimizes power management system to improve battery life; and designs a high-protection-level structure to adapt to complex outdoor environments. S2: Software algorithm development and system integration; develop fatigue calculation algorithms to process multi-dimensional index data in real time, realize software integration between the fatigue index screening system and LoRa monitoring terminal, and support real-time reception and updating of index weights, thresholds and other parameters; S3: Anti-interference mechanism design; integrates anti-electromagnetic interference technology into the LoRa communication module, including forward error correction coding, frequency hopping technology and CRC cyclic redundancy check, to ensure the stability of data transmission in strong electromagnetic field environments; S4: Real-time data transmission and processing; Real-time transmission of data in sets per second is achieved through the LoRa network, transmitting the collected physiological index data to the edge computing terminal or ground control terminal, and simultaneously completing data preprocessing and fatigue value calculation. S5: Early warning function integration; The monitoring terminal has a built-in sound, light and vibration module. When the fatigue value exceeds the preset threshold, a graded early warning signal is triggered, and the early warning information is synchronized to the ground control terminal.

[0017] In a preferred embodiment, in step S2, the fatigue assessment model integrates multi-dimensional indicators and weights to achieve quantitative assessment and classification of fatigue state. Basic assessment layer: Based on physiological indicators: heart rate variability, electromyography, biochemical indicators: cortisol, psychological indicators: occupational burnout score and physical fitness test results, the initial fatigue value (F0) is calculated using a linear weighted formula: F=∑(indicator value i×weight i); Scene correction layer: Introduces environmental risk coefficients (such as strong electromagnetic field coefficient 1.2, high temperature coefficient 1.1) and operation difficulty coefficients (such as "equipotential live operation" coefficient 1.5, "tower erection" coefficient 1.3), and dynamically adjusts the thresholds; the final fatigue value (F) is: F = F × environmental risk coefficient × operation difficulty coefficient; Grading criteria: Based on simulated climbing test data (36 portrait samples), three threshold levels are set: Safety level (F < 60): No obvious fatigue, can work normally; Warning level (60≤F<80): Work rhythm needs to be adjusted or energy replenished; Hazard level (F≥80): Work must be stopped and rest required.

[0018] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: 1. This invention achieves a comprehensive and accurate assessment of worker fatigue by constructing a multi-dimensional indicator system and employing scientific screening methods. The system integrates multi-dimensional indicators such as electrocardiogram, heart rate, respiratory rate, skin temperature, triaxial posture, exercise intensity score, and a 6-minute walking test of physical fitness. Combined with a five-step elimination screening method, it enables real-time monitoring, non-invasive high-altitude monitoring, individual difference correction, synchronization verification, scene adaptive adjustment, and Spearman correlation analysis. This process eliminates highly interfering or lagging indicators, retaining core indicators highly correlated with fatigue development trends. This ensures the assessment covers all dimensions of workers' physiological responses, psychological perceptions, and behavioral performance, significantly improving the accuracy of fatigue state assessment.

[0019] 2. In this invention, relying on LoRa technology and anti-interference design, it can still operate stably in strong electromagnetic field environments, effectively ensuring the reliability of data transmission. The LoRa communication module adopts adaptive spread spectrum technology, frequency hopping mechanism to switch the operating frequency band every 5 seconds, and forward error correction coding. Combined with the spiral anti-interference antenna design, it can resist strong electromagnetic interference such as power frequency electric field and magnetic induction intensity fluctuations in scenarios such as 500kV live lines. At the same time, CRC cyclic redundancy check technology ensures data integrity and avoids transmission errors caused by signal interference, enabling the system to continuously and stably transmit monitoring data in complex electromagnetic environments.

[0020] 3. In this invention, the long-distance transmission capability of LoRa technology meets the special monitoring needs of high-altitude power transmission operations, solving the problem of limited transmission distance of traditional equipment. The system uses a LoRa wireless communication chip to support Mesh networking, with a theoretical transmission distance of ≥500 meters. In actual operation of the 500kV Yanping No. 2 line grounding wire, data can still be received normally at a maximum distance of 210 meters. The reliability of its long-distance transmission has been verified in various high-altitude operation scenarios such as tower erection and power outage troubleshooting, ensuring that the ground control terminal can obtain fatigue data of high-altitude workers in real time, breaking through the monitoring blind zone limitation caused by the short transmission distance of traditional monitoring equipment.

[0021] 4. In this invention, by calculating fatigue levels in real time and triggering graded early warnings, the fatigue state of workers can be detected in a timely manner, effectively reducing the risk of safety accidents. The monitoring terminal collects and transmits a set of physiological data every second. After being processed in real time by the edge computing terminal or the ground control terminal, the fatigue level is quickly calculated by integrating basic indicators and scenario correction coefficients through a fatigue assessment model. When the value exceeds the preset thresholds for safety level, early warning level, and danger level, the terminal immediately issues graded early warning signals through a vibration motor, LED lights, and a buzzer. For example, the danger level triggers high-frequency vibration, red lights, and a buzzer. The ground control terminal simultaneously receives the early warning information and intervenes, realizing an instant response of monitoring-assessment-early warning, providing key protection for the safety of workers.

[0022] 5. In this invention, the high integration of the fatigue index screening system and the LoRa monitoring terminal realizes an integrated process from data acquisition, transmission, processing to early warning, significantly improving the system's practicality and reliability. At the hardware level, the terminal integrates sensors such as PPG and GSR, a LoRa communication module, a new power architecture design, and an IP68 protection rating structure, forming an integrated hardware system of acquisition, transmission, power supply, and protection. At the software level, the fatigue calculation algorithm is deeply integrated with the terminal, supporting real-time reception of index weight and threshold updates. Combined with data interaction between 4G / 5G networks, ground control terminals, and headquarters servers, it achieves fully automated operation. This highly integrated design reduces the complexity of multi-device collaboration, minimizes human error, and makes the system more suitable for the actual application needs of high-altitude power transmission operations. Attached Figure Description

[0023] Figure 1 This is an overall system block diagram of the present invention; Figure 2 This is a diagram of the fatigue quantification assessment model for the work process in this invention; Figure 3 This is a schematic diagram of the indicator set in this invention; Figure 4 This is a flowchart of the specific indicator screening method in this invention; Figure 5 This is a flowchart of the multi-dimensional representation mapping method in this invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0025] Example: Reference Figure 1-5 , A fatigue index screening system includes: a multi-dimensional data acquisition module, a dynamic adaptive index screening module, an intelligent weight calculation and optimization module, a fatigue assessment model construction module, a LoRa communication and anti-interference module, and a real-time early warning and feedback module. The modules in the fatigue index screening system are closely connected through data flow: The output of the multi-dimensional data acquisition module is directly connected to the input of the dynamic adaptive index filtering module; The output of the dynamic adaptive index screening module is connected to the input of the intelligent weight calculation and optimization module; The output of the intelligent weight calculation and optimization module is connected to the input of the fatigue assessment model construction module; The output of the fatigue assessment model building module is connected to the input of the LoRa communication and anti-interference module; The output of the LoRa communication and anti-interference module is connected to the input of the real-time early warning and feedback module; The output of the real-time early warning and feedback module can be connected in reverse to the input of the LoRa communication and anti-interference module to transmit the adjustment commands from the ground control terminal back to the front-end device, forming a closed-loop connection for data interaction.

[0026] The multi-dimensional data acquisition module covers physiological, psychological, behavioral, and environmental data of personnel working at heights in power transmission lines, providing multi-source data support for subsequent analysis; The multi-dimensional data acquisition module includes: Physiological indicator acquisition module: integrates PPG (photoplethysmography) sensor (for heart rate and blood oxygen), GSR (gamma-ray skin response) sensor (for emotional stress), surface electromyography sensor (for muscle fatigue), accelerometer (for movement trajectory), etc., to achieve non-invasive, real-time physiological signal acquisition.

[0027] Biochemical index collection module: The wearable microfluidic chip saliva collection module detects cortisol biochemical indexes, overcoming the limitation of not being able to draw blood during high-altitude operations.

[0028] Subjective psychological data collection module: Integrates a lightweight mobile questionnaire system (based on modified SF-36, SCL-90, and Maslach scales), supports quick completion during work breaks, and combines voice interaction functions (such as "Please describe your current fatigue feelings") to reduce the burden of filling out questionnaires.

[0029] Environmental and operational data acquisition module: The terminal uses built-in electromagnetic field sensors (monitoring power frequency electric field and magnetic induction intensity), temperature and humidity sensors (environmental comfort), and operational type tags (such as "pole erection" and "conductor and ground wire deployment") to record the characteristics of the operational scene.

[0030] The dynamic adaptive indicator screening module consists of a real-time monitoring screening unit, a high-altitude non-invasive screening unit, an individual difference correction unit, a synchronization verification unit, and a scenario adaptive adjustment unit. The real-time monitoring screening unit, based on the equipment's data acquisition capabilities, excludes indicators requiring invasive testing or costly high-frequency sampling, retaining only physiological indicators collected in real-time by wearable devices. The high-altitude non-invasive screening unit, targeting high-altitude work scenarios, eliminates indicators requiring contact sampling or interfering with the work, retaining only indicators that can be collected by surface sensors or microfluidic chips. The individual difference correction unit uses historical data to establish an individual baseline model, filtering indicators with large group fluctuations and retaining indicators with strong consistency. The synchronization verification unit uses timestamp alignment technology to screen indicators that change synchronously with fatigue development trends. The scenario adaptive adjustment unit combines environmental sensors and work type labels to identify the current scenario and dynamically adjusts the screening thresholds of the first four steps; for example, in strong electromagnetic field scenarios, indicators with strong anti-interference capabilities are prioritized, and in high-temperature and high-humidity scenarios, the weight of skin temperature indicators is increased.

[0031] The dynamic adaptive index screening module is based on a dual screening logic of "five-step funnel method + scenario adaptive mechanism" to solve the problem that traditional screening cannot adapt to complex working conditions. Specifically, it includes the following steps: Step 1: Real-time monitoring and screening: Exclude indicators that require invasive testing (such as blood glucose) or have high costs for high-frequency sampling (such as electroencephalography), and retain only indicators that can be collected in real time through wearable devices (such as heart rate variability and skin conductance).

[0032] Step 2: High-altitude non-invasive screening: Eliminate indicators that require contact sampling (such as venous blood) or interfere with the operation (such as eye tracking), and retain only indicators that can be collected by surface sensors or microfluidic chips (such as surface electromyography and salivary cortisol).

[0033] Step 3: Individual difference correction: Establish an "individual baseline model" using historical data (such as the basic indicator range of people with different BMIs and ages), automatically filter out indicators with large individual fluctuations (such as blood pressure), and retain only indicators with strong group consistency (such as heart rate variability).

[0034] Step 4: Synchronization verification: Based on timestamp alignment technology, screen indicators that change synchronously with the fatigue development trend (such as electromyographic signals increasing synchronously with muscle fatigue), and exclude lagging or leading indicators (such as some biochemical indicators).

[0035] Step 5: Adaptive Scene Adjustment: Introduce a work scene perception module (identifying the current scene through environmental sensors and work type labels, such as "500kV live-line work" or "220kV tower erection") to dynamically adjust the screening thresholds from the first four steps. For example, in a "strong electromagnetic field scene," skin conductance indicators with strong anti-interference capabilities are prioritized; in a "high temperature and high humidity scene," the screening weight of skin temperature indicators is increased.

[0036] The intelligent weight calculation and optimization module consists of an initial weight calculation submodule and a dynamic optimization submodule. The initial weight calculation submodule combines domain expert experience with Spearman correlation analysis to generate an initial weight matrix using a weighted scoring method. Expert experience is used to set the sensitivity benchmark for the indicators, and Spearman analysis calculates the monotonic correlation between the indicators and the scores of the subjective fatigue scale. The dynamic optimization submodule, based on machine learning algorithms such as random forests, trains on historical sample data and automatically adjusts the weight values. If the correlation between a certain indicator and fatigue level changes significantly under a specific scenario, its weight is adjusted accordingly to ensure that the weight allocation matches the actual working conditions.

[0037] The fatigue assessment model construction module comprises a basic assessment layer, a scenario correction layer, and a grading standard setting layer. The basic assessment layer integrates physiological, biochemical, psychological, and physical fitness test indicators, calculating the initial fatigue value using a linear weighted formula, where each indicator value corresponds to a weight. The scenario correction layer introduces an environmental risk coefficient and a work difficulty coefficient. The environmental risk coefficient is determined based on environmental data such as electromagnetic fields, temperature, and humidity, while the work difficulty coefficient is set according to the work type (e.g., equipotential live-line work, pole erection). Both factors work together to adjust the initial fatigue value. The grading standard setting layer, based on 36 portrait sample data from simulated climbing tests, divides the data into three threshold levels: safety, warning, and danger, each corresponding to different fatigue state judgment standards.

[0038] in: Physiological indicators include heart rate variability and blood oxygen saturation.

[0039] Currently, there are nearly 700 physiological indicators reflecting fatigue that are available in the literature. This application employs a five-stage funnel screening process. The first step involves moving from a basic pool of indicators to those that can be monitored in real time. For example, blood glucose fluctuations are related to fatigue, but blood glucose is considered an indicator that cannot be monitored in real time because it cannot be measured in real time. The second step involves screening for indicators that can be monitored non-invasively at heights. Simply put, it's impossible to draw blood while people are working at heights, so non-invasive high-altitude monitoring indicators must be selected. The third step eliminates individual variability, as some indicators exhibit significant individual variations, such as blood pressure, which varies considerably between different individuals. In practice, these cannot be fused. The final part involves filtering out indicators that exhibit synchronous changes after these four steps. That is, because the timestamps are always consistent during the detection process, these indicators must exhibit synchronous changes with consistent timestamps to be screened. Asynchronous changes would indicate that the indicator reacts slower than other indicators, rendering it meaningless. This is called synchronous screening, followed by indicator fusion. Therefore, after these screenings were completed, this application conducted a correlation analysis of the subjective questionnaire, thus completing the entire indicator screening process of this application.

[0040] Psychological indicators include: modified SF-36 (quality of life), SCL-90 (mental health), and Maslach (burnout) scale scores.

[0041] Behavioral and performance parameters.

[0042] Workload index.

[0043] Background information: Unit type, age, BMI, etc. (used to define this portrait set).

[0044] A total of more than 25,000 basic data sets of individuals are distributed across these 36 portrait sets.

[0045] Key features and objectives: Structured and refined: The sample database is not a chaotic collection of data, but rather the result of predefined grouping dimensions (units, age, BMI) based on research needs (analyzing differences in fatigue among different populations) and refined segmentation accordingly. The 36 portrait sets represent 36 pre-defined research scenarios or population subgroups.

[0046] Ensuring representativeness and universality: By covering three types of units, different ages, and different BMIs, 36 combinations were constructed to ensure that the research sample represents the diversity of actual power transmission workers at heights as broadly as possible. This makes the final research conclusions (fatigue assessment model, influencing factors, and management recommendations) more universally applicable and able to provide more targeted recommendations for different groups of people.

[0047] Support for in-depth comparative analysis: One of the core purposes of establishing a portrait collection sample library is to facilitate comparative studies between groups. For example: Compare the fatigue responses of employees in their main job, outsourced employees, and employees in industrial units at the same age and BMI.

[0048] Compare the fatigue differences between younger and older employees under the same workplace type and BMI.

[0049] Compare the fatigue status of individuals with different BMI levels under the same workplace and age.

[0050] This study analyzes the impact of factors such as age and BMI on fatigue in a specific unit type.

[0051] Multidimensional data integration: Each portrait set integrates fatigue-related data from all aspects, including physiological, psychological, behavioral, and subjective feelings, providing a rich and comprehensive analytical foundation for the research.

[0052] Research repository: This sample repository is the most important core asset of the project. Based on it, data mining, correlation analysis, and model construction (such as fatigue prediction models) are carried out to draw "conclusions with practical guiding significance".

[0053] In the project of fatigue assessment for high-altitude power transmission operations, the "portrait set sample repository" refers to a research database containing 36 independent data subsets (portrait sets). Each "portrait set" corresponds to a specific group of workers defined by the nature of the unit, age group, and BMI level, and contains all-round information such as the physiological indicators (heart rate variability, blood oxygen, etc.), psychological scale scores (SF-36, SCL-90, Maslach), behavioral data, and workload of individuals within the group. The entire sample repository (the sum of 36 portrait sets) systematically and structurally covers the key feature combinations of the research target population. Its core purpose is to ensure the representativeness and universality of the research, and support the refined comparative analysis and modeling of the fatigue status of different characteristic populations, providing a solid data foundation for the final formation of an effective fatigue assessment and control system. It is the core organizational form for data collection, collation, and analysis in this project. BMI refers to Body Mass Index, which is usually translated into Chinese as "body mass index" or simply "physical fitness index".

[0054] Core definition and calculation of BMI: Quantitative indicator: BMI is a simple and standardized numerical indicator used to measure the relationship between adult body weight and height.

[0055] Calculation formula: BMI = weight (kg) / [height (m)] 2 For example: A person with a height of 1.75 meters and a weight of 70 kilograms: Square of height = 1.75 * 1.75 = 3.0625 square meters BMI = 70 / 3.0625 ≈ 22.86 Main uses and significance of BMI: Preliminary judgment of weight status: BMI is a tool widely adopted by the World Health Organization (WHO) and national health agencies for quickly screening whether the body weight of adults is within the healthy range (underweight, normal, overweight or obese).

[0056] Health risk assessment indicator: Too low BMI: May indicate malnutrition, certain wasting diseases or other health problems.

[0057] Too high BMI: Is an important risk factor for many chronic diseases, especially when it reaches the level of obesity. Risks associated with too high BMI include: Cardiovascular diseases (such as hypertension, coronary heart disease, stroke), type 2 diabetes, certain cancers, osteoarthritis (which increases the burden on joints), sleep apnea, metabolic syndrome... as well as issues of concern to the project such as fatigue, decreased energy levels, and impaired work performance.

[0058] Specific applications and importance in fatigue assessment projects for high-altitude operations in power transmission: Key grouping variables: As mentioned, the project included BMI as one of the core variables (alongside employer type and age) to construct 36 “portrait sets.” This indicates that the project highly values ​​the impact of weight status on the fatigue experience of workers at heights.

[0059] Factors affecting fatigue: Energy consumption: Being overweight or obese increases the burden on the body, especially in high-altitude work that requires physical activities such as climbing, balancing, and carrying heavy loads. This leads to faster energy consumption, earlier physical exhaustion, and slower recovery, thus exacerbating fatigue.

[0060] Cardiopulmonary burden: Excessive weight increases the burden on the cardiovascular and respiratory systems, reduces overall physical fitness and endurance, and makes it easier to feel fatigued in environments that place high demands on cardiopulmonary function, such as working at heights.

[0061] Heat stress and comfort: High body fat percentage may affect thermoregulation, making individuals more prone to discomfort and fatigue when working in hot or cold environments. Heavy protective gear may also place an additional burden on larger individuals.

[0062] Musculoskeletal stress: Extra weight increases stress on joints (such as the knees and back), and prolonged work can lead to musculoskeletal pain and strain, which is also a major source of fatigue.

[0063] Associated with potential health problems: Abnormal BMI (too high or too low) is often accompanied by other health problems (such as sleep problems, endocrine problems, and decreased cardiovascular function), which themselves can significantly affect energy and fatigue levels.

[0064] Refined fatigue assessment: By using BMI as a stratification factor, the project is able to: Comparative analysis: This study investigates whether there are significant differences in physiological indicators (heart rate variability, blood oxygenation), psychological feelings (scale scores), behavioral performance, and subjective fatigue among workers of different BMI levels (e.g., normal, overweight, obese) under similar unit conditions and age.

[0065] Identify high-risk groups: Determine which BMI groups are more prone to fatigue problems in high-altitude work environments to provide a basis for developing more targeted fatigue intervention and health management measures (e.g., physical training, nutritional guidance, or work arrangement adjustments for overweight / obese employees).

[0066] Enhancing model accuracy: BMI is an important input variable when building fatigue prediction models, which helps improve the accuracy of the model in predicting fatigue risk for people of different body types.

[0067] Important things to note about BMI: Screening tool, not diagnostic tool: BMI is a preliminary screening tool that primarily reflects the overall relationship between weight and height, and cannot distinguish the proportion of muscle and fat in weight. Therefore: For individuals with very developed muscles (such as bodybuilders or certain manual laborers), their BMI may be high, classifying them as "overweight" or even "obese," but this does not necessarily mean that their body fat percentage is high or that they have health risks.

[0068] For older adults with insufficient muscle mass but relatively normal body fat, their BMI may be within the "normal" range, but this may conceal risks such as sarcopenia.

[0069] Other assessments are needed: To more accurately assess an individual's health status or degree of obesity, it is usually necessary to combine other indicators, such as: Waist circumference (measures abdominal fat and is more directly related to health risks) Body fat percentage (direct measurement of fat content) Laboratory indicators such as blood pressure, blood sugar, and blood lipids.

[0070] Overall health assessment and medical history.

[0071] In the fatigue assessment project for high-altitude operations in power transmission, BMI (Body Mass Index) is a core categorical variable and analytical dimension in the project design. By grouping workers according to different BMI levels, the project aims to delve into the specific impact of weight status, a crucial physiological factor, on the physiological, psychological, and behavioral manifestations of fatigue in high-altitude work environments. This is critical for identifying high-risk fatigue groups, understanding the mechanisms of fatigue (especially in terms of physical exertion and health risks), and developing precise and effective fatigue management strategies. The project team clearly recognizes the particular importance of BMI as a key indicator of weight-related health risks and physical exertion in high-intensity, high-risk high-altitude work environments. • Construction of a multi-dimensional indicator system: Physiological indicators include heart rate variability (HRV), skin conductance, respiratory rate, skin temperature, and surface electromyography.

[0072] Subjective psychological indicators: perceived fatigue, anxiety, depression, workload, etc.

[0073] Physical fitness indicators include the results of the 6MWT test (6-minute maximum walking distance test), the blindfolded single-leg standing test, the squat support test, and the simulated climbing test.

[0074] • Indicator selection method: A weighted scoring method was adopted, assigning different weights to each indicator based on its effectiveness and sensitivity in fatigue assessment. The weight calculation formula is: Overall Score = Σ (Indicator Weight × Indicator Score).

[0075] By leveraging big data analytics and machine learning algorithms, multi-dimensional indicators are screened and optimized to identify core monitoring metrics. For example, the Spearman correlation coefficient is used to calculate the correlation between each indicator and fatigue levels, allowing for the selection of indicators with strong correlations.

[0076] (Spearman's core idea: Spearman correlation coefficient (usually denoted as ρ or r)) s The correlation coefficient (CCC) measures the strength and direction of the monotonic relationship between two variables. It focuses on whether the rank of the two variables is consistent, rather than on the linear relationship of the original values, as the Pearson correlation coefficient does.

[0077] The LoRa communication and anti-interference module consists of a hardware communication submodule and a software anti-interference submodule. The hardware communication submodule uses a LoRa wireless communication chip to support Mesh networking, integrates adaptive spread spectrum technology to adjust the transmission rate according to the interference intensity, and optimizes the spiral anti-interference antenna design to improve the stability of long-distance transmission. The software anti-interference submodule reduces data loss rate through forward error correction coding, uses frequency hopping technology to switch the operating frequency band every 5 seconds to avoid strong electromagnetic field interference, and combines CRC cyclic redundancy check to ensure the integrity of transmitted data, jointly ensuring reliable data transmission in complex environments.

[0078] The real-time early warning and feedback module consists of a terminal early warning component and a ground feedback component. The terminal early warning component has a built-in vibration motor, LED light, and buzzer. It triggers different early warning modes according to the fatigue level classification. The dangerous level triggers high-frequency vibration, red light, and buzzer, while the early warning level triggers low-frequency vibration and yellow light. The ground feedback component pushes fatigue value, early warning level, and key indicators to the ground control terminal in real time via 4G / 5G network. This allows management personnel to view the fatigue heat map of the workers and remotely send rest instructions or adjust work tasks, realizing two-way interaction between the front end and the management end.

[0079] A fatigue index screening system and its integration method in a LoRa monitoring terminal include the following steps: S1: Hardware Integration. Integrates physiological information acquisition devices such as PPG, GSR, and accelerometer sensors. Employs LoRa wireless communication chips to support long-distance (≥500 meters) transmission and Mesh networking. Optimizes the power management system (such as a new power architecture) to improve battery life. Designed with a high protection rating (IP68) structure to adapt to complex outdoor environments.

[0080] S2: Software Algorithm Development and System Integration. Develop a fatigue calculation algorithm to process multi-dimensional indicator data in real time, and achieve software integration between the fatigue indicator screening system and the LoRa monitoring terminal, supporting real-time reception and updating of indicator weights, thresholds, and other parameters.

[0081] S3: Anti-interference mechanism design. Anti-electromagnetic interference technology is integrated into the LoRa communication module, including forward error correction coding, frequency hopping technology (switching the operating frequency band every 5 seconds), and CRC cyclic redundancy check, to ensure the stability of data transmission in strong electromagnetic field environments (such as 500kV live lines).

[0082] S4: Real-time data transmission and processing. Data is transmitted in real-time at a rate of one packet per second via the LoRa network, transmitting collected physiological indicator data to edge computing terminals or ground control terminals, simultaneously completing data preprocessing and fatigue value calculation.

[0083] S5: Integrated early warning function. The monitoring terminal has a built-in sound, light and vibration module (such as a vibration motor, LED light and buzzer). When the fatigue value exceeds the preset threshold, it triggers a graded early warning signal (such as dangerous high-frequency vibration + red light + buzzer) and synchronizes the early warning information to the ground control terminal.

[0084] In step S2, the fatigue assessment model integrates multi-dimensional indicators and weights to achieve quantitative assessment and classification of fatigue status.

[0085] Basic assessment layer: Based on physiological indicators (heart rate variability, electromyography), biochemical indicators (cortisol), psychological indicators (occupational burnout score), and physical fitness test results (6-minute walking distance), the initial fatigue value (F0) is calculated using a linear weighted formula: F = ∑(indicator value i × weight i) Scenario Correction Layer: This layer introduces environmental risk coefficients (e.g., strong electromagnetic field coefficient 1.2, high temperature coefficient 1.1) and operational difficulty coefficients (e.g., "equipotential live-line work" coefficient 1.5, "tower erection" coefficient 1.3) to dynamically adjust thresholds. The final fatigue value (F) is: F = F × Environmental Risk Coefficient × Operational Difficulty Coefficient Grading criteria: Based on simulated climbing test data (36 portrait samples), three threshold levels are set: Safety level (F < 60): No obvious fatigue, can work normally; Warning level (60≤F<80): Work rhythm needs to be adjusted or energy replenished; Hazard level (F≥80): Work must be stopped and rest required.

[0086] This invention achieves a comprehensive and accurate assessment of worker fatigue by constructing a multi-dimensional indicator system and employing scientific screening methods. The system integrates multi-dimensional indicators such as electrocardiogram, heart rate, respiratory rate, skin temperature, triaxial posture, exercise intensity score, and a 6-minute walking test of physical fitness. Combined with a five-step elimination screening method, it enables real-time monitoring, non-invasive high-altitude monitoring, individual difference correction, synchronization verification, scene adaptive adjustment, and Spearman correlation analysis. This process eliminates highly interfering or lagging indicators, retaining core indicators highly correlated with fatigue development trends. This ensures the assessment covers all dimensions of workers' physiological responses, psychological perceptions, and behavioral performance, significantly improving the accuracy of fatigue state assessment.

[0087] This invention, relying on LoRa technology and anti-interference design, can still operate stably in strong electromagnetic field environments, effectively ensuring the reliability of data transmission. The LoRa communication module adopts adaptive spread spectrum technology, frequency hopping mechanism to switch the operating frequency band every 5 seconds, and forward error correction coding. Combined with a spiral anti-interference antenna design, it can resist strong electromagnetic interference such as power frequency electric field and magnetic induction intensity fluctuations in scenarios such as 500kV live lines. At the same time, CRC cyclic redundancy check technology ensures data integrity and avoids transmission errors caused by signal interference, enabling the system to continuously and stably transmit monitoring data in complex electromagnetic environments.

[0088] In this invention, the long-distance transmission capability of LoRa technology meets the special monitoring needs of high-altitude power transmission operations, solving the problem of limited transmission distance of traditional equipment. The system uses a LoRa wireless communication chip to support Mesh networking, with a theoretical transmission distance of ≥500 meters. In actual operation of the 500kV Yanping No. 2 line grounding wire, data can still be received normally at a maximum distance of 210 meters. The reliability of its long-distance transmission has been verified in various high-altitude operation scenarios such as tower erection and power outage troubleshooting, ensuring that the ground control terminal can obtain fatigue data of high-altitude workers in real time, breaking through the monitoring blind zone limitation caused by the short transmission distance of traditional monitoring equipment.

[0089] In this invention, by calculating fatigue levels in real time and triggering tiered early warnings, the fatigue status of workers can be detected promptly, effectively reducing the risk of safety accidents. The monitoring terminal collects and transmits a set of physiological data every second. After real-time processing by an edge computing terminal or ground control terminal, the fatigue level is quickly calculated by integrating basic indicators and scenario correction coefficients through a fatigue assessment model. When the value exceeds preset thresholds for safety, warning, and danger levels, the terminal immediately issues tiered early warning signals via a vibration motor, LED lights, and a buzzer. For example, a danger level triggers high-frequency vibration, red lights, and a buzzer. The ground control terminal simultaneously receives the early warning information and intervenes, achieving an instant response of monitoring, assessment, and early warning, providing crucial protection for worker safety.

[0090] In this invention, the high integration of the fatigue index screening system and the LoRa monitoring terminal realizes an integrated process from data acquisition, transmission, processing to early warning, significantly improving the system's practicality and reliability. At the hardware level, the terminal integrates sensors such as PPG and GSR, a LoRa communication module, a new power architecture design, and an IP68 protection rating structure, forming an integrated hardware system of acquisition, transmission, power supply, and protection. At the software level, the fatigue calculation algorithm is deeply integrated with the terminal, supporting real-time reception of index weight and threshold updates. Combined with data interaction between 4G / 5G networks, ground control terminals, and headquarters servers, it achieves fully automated operation. This highly integrated design reduces the complexity of multi-device collaboration, minimizes human error, and makes the system more suitable for the actual application needs of high-altitude power transmission operations.

[0091] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the term "comprising" or any other variations thereof is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0092] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A fatigue index screening system, characterized in that: include: Multi-dimensional data acquisition module, dynamic adaptive index screening module, intelligent weight calculation and optimization module, fatigue assessment model construction module, LoRa communication and anti-interference module, and real-time early warning and feedback module; The modules in the fatigue index screening system are closely connected through data flow: The output of the multi-dimensional data acquisition module is directly connected to the input of the dynamic adaptive index filtering module; The output of the dynamic adaptive index screening module is connected to the input of the intelligent weight calculation and optimization module; The output of the intelligent weight calculation and optimization module is connected to the input of the fatigue assessment model construction module; The output of the fatigue assessment model construction module is connected to the input of the LoRa communication and anti-interference module; The output of the LoRa communication and anti-interference module is connected to the input of the real-time early warning and feedback module; The output of the real-time early warning and feedback module can be connected in reverse to the input of the LoRa communication and anti-interference module to transmit the adjustment instructions from the ground control terminal back to the front-end device, forming a closed-loop connection for data interaction.

2. The fatigue index screening system as described in claim 1, characterized in that: The multi-dimensional data acquisition module covers physiological, psychological, behavioral, and environmental data of personnel working at heights in power transmission lines, providing multi-source data support for subsequent analysis; The multi-dimensional data acquisition module includes: Physiological indicator acquisition module: integrates PPG sensor, GSR sensor, surface electromyography sensor, accelerometer, etc., to realize non-invasive, real-time physiological signal acquisition; Biochemical indicator acquisition module: Through a wearable microfluidic chip, it detects biochemical indicators such as electrocardiogram, heart rate, respiratory rate, skin temperature, three-axis posture, and exercise intensity, solving the problem of not being able to draw blood during high-altitude operations; Subjective psychological data collection module: Integrates a lightweight mobile questionnaire system, supports quick completion during work breaks, and combines voice interaction function to reduce the burden of filling out questionnaires; Environmental and operational data acquisition module: Records operational scene characteristics through the terminal's built-in electromagnetic field sensor, temperature and humidity sensor, and operational type tags.

3. The fatigue index screening system as described in claim 1, characterized in that: The dynamic adaptive index screening module consists of a real-time monitoring screening unit, a high-altitude non-invasive screening unit, an individual difference correction unit, a synchronization verification unit, and a scenario adaptive adjustment unit. The real-time monitoring screening unit excludes indicators that require invasive detection or have high cost for high-frequency sampling based on the equipment's acquisition capabilities, retaining only physiological indicators that can be collected in real time by wearable devices. The high-altitude non-invasive screening unit, targeting high-altitude work scenarios, eliminates indicators that require contact sampling or interfere with the work, retaining only indicators that can be collected by surface sensors or microfluidic chips. The individual difference correction unit uses historical data to establish an individual baseline model, filters out indicators with large group fluctuations, and retains indicators with strong consistency; the synchronization verification unit uses timestamp alignment technology to screen indicators that change synchronously with the fatigue development trend; the scene adaptive adjustment unit combines environmental sensors and work type labels to identify the current scene and dynamically adjusts the screening thresholds of the first four steps. For example, in strong electromagnetic field scenes, indicators with strong anti-interference are retained first, and in high temperature and high humidity scenes, the weight of skin temperature indicators is increased.

4. The fatigue index screening system as described in claim 1, characterized in that: The dynamic adaptive index screening module is based on a dual screening logic of "five-step funnel method + scenario adaptive mechanism" to solve the problem that traditional screening cannot adapt to complex working conditions. Specifically, it includes the following steps: Step 1: Real-time monitoring and screening: Exclude indicators that require invasive testing or have high costs for high-frequency sampling, and retain only indicators that can be collected in real time by wearable devices; Step 2: High-altitude non-invasive screening: Eliminate indicators that require contact sampling or interfere with the operation, and retain only indicators that can be collected by surface sensors or microfluidic chips; Step 3: Individual difference correction: Establish an "individual baseline model" through historical data, automatically filter out indicators with large individual fluctuations, and retain only indicators with strong group consistency; Step 4: Synchronization verification: Based on timestamp alignment technology, screen indicators that change synchronously with the fatigue development trend and exclude lagging or leading indicators. Step 5: Adaptive adjustment of the scene: Introduce a scene perception module to dynamically adjust the screening thresholds of the first four steps; for example, in the "strong electromagnetic field scene", skin conductance index with strong anti-interference is retained first; in the "high temperature and high humidity scene", the screening weight of skin temperature index is increased.

5. The fatigue index screening system as described in claim 1, characterized in that: The intelligent weight calculation and optimization module consists of an initial weight calculation submodule and a dynamic optimization submodule. The initial weight calculation submodule combines domain expert experience with Spearman correlation analysis to generate an initial weight matrix using a weighted scoring method. Expert experience is used to set the sensitivity benchmark for the indicators, and Spearman analysis calculates the monotonic correlation between the indicators and the scores of the subjective fatigue scale. The dynamic optimization submodule is based on machine learning algorithms such as random forests, trains on historical sample data, and automatically corrects the weight values. If the correlation between a certain indicator and fatigue changes significantly in a specific scenario, its weight is adjusted accordingly to ensure that the weight allocation matches the actual working conditions.

6. The fatigue index screening system as described in claim 1, characterized in that: The fatigue assessment model construction module includes a basic assessment layer, a scenario correction layer, and a grading standard setting layer. The basic assessment layer integrates physiological, biochemical, psychological, and physical fitness test indicators and calculates the initial fatigue value through a linear weighted formula, in which the indicator value and weight correspond one-to-one. The scenario correction layer introduces an environmental risk coefficient and a task difficulty coefficient. The environmental risk coefficient is determined based on environmental data such as electromagnetic fields, temperature, and humidity, while the task difficulty coefficient is set according to the task type. Both adjust the initial fatigue value together. The grading standard is set based on 36 portrait samples from simulated climbing tests, dividing the threshold into three levels: safety, warning, and danger, each corresponding to different fatigue state judgment criteria.

7. The fatigue index screening system as described in claim 1, characterized in that: The LoRa communication and anti-interference module consists of a hardware communication sub-module and a software anti-interference sub-module. The hardware communication sub-module uses a LoRa wireless communication chip to support Mesh networking, integrates adaptive spread spectrum technology to adjust the transmission rate according to the interference intensity, and optimizes the spiral anti-interference antenna design to improve the stability of long-distance transmission. The software anti-interference submodule reduces data loss rate through forward error correction coding, uses frequency hopping technology to switch the operating frequency band every 5 seconds to avoid strong electromagnetic field interference, and combines CRC cyclic redundancy check to ensure the integrity of transmitted data, together ensuring reliable data transmission in complex environments.

8. The fatigue index screening system as described in claim 1, characterized in that: The real-time early warning and feedback module consists of a terminal early warning component and a ground feedback component. The terminal early warning component has a built-in vibration motor, LED light and buzzer. It triggers different early warning modes according to the fatigue level. The dangerous level triggers high-frequency vibration, red light and buzzer, and the early warning level triggers low-frequency vibration and yellow light. The ground feedback component pushes fatigue values, early warning levels, and key indicators to the ground control terminal in real time via 4G / 5G network, allowing management personnel to view the fatigue heat map of workers and remotely send rest instructions or adjust work tasks, thus realizing two-way interaction between the front end and the management end.

9. The integration method of the fatigue index screening system in a LoRa monitoring terminal as described in claim 1, characterized in that: The method includes the following steps: S1: Hardware integration; integrates physiological information acquisition devices such as PPG, GSR, and accelerometer; adopts LoRa wireless communication chip to support long-distance transmission and Mesh networking; optimizes power management system to improve battery life; and designs a high-protection-level structure to adapt to complex outdoor environments. S2: Software algorithm development and system integration; develop fatigue calculation algorithms to process multi-dimensional index data in real time, realize software integration between the fatigue index screening system and LoRa monitoring terminal, and support real-time reception and updating of index weights, thresholds and other parameters; S3: Anti-interference mechanism design; integrates anti-electromagnetic interference technology into the LoRa communication module, including forward error correction coding, frequency hopping technology and CRC cyclic redundancy check, to ensure the stability of data transmission in strong electromagnetic field environments; S4: Real-time data transmission and processing; Real-time transmission of data in sets per second is achieved through the LoRa network, transmitting the collected physiological index data to the edge computing terminal or ground control terminal, and simultaneously completing data preprocessing and fatigue value calculation. S5: Early warning function integration; The monitoring terminal has a built-in sound, light and vibration module. When the fatigue value exceeds the preset threshold, a graded early warning signal is triggered, and the early warning information is synchronized to the ground control terminal.

10. The fatigue index screening system as described in claim 1 and its integration method in a LoRa monitoring terminal, characterized in that: In step S2, the fatigue assessment model integrates multi-dimensional indicators and weights to achieve quantitative assessment and classification of fatigue state. Basic assessment layer: Based on physiological indicators: heart rate variability, electromyography, biochemical indicators: cortisol, psychological indicators: occupational burnout score and physical fitness test results, the initial fatigue value (F0) is calculated using a linear weighted formula: F=∑(indicator value i×weight i); Scene correction layer: Introduces environmental risk coefficients (such as strong electromagnetic field coefficient 1.2, high temperature coefficient 1.1) and operation difficulty coefficients (such as "equipotential live operation" coefficient 1.5, "tower erection" coefficient 1.3), and dynamically adjusts the thresholds; the final fatigue value (F) is: F = F × environmental risk coefficient × operation difficulty coefficient; Grading criteria: Based on simulated climbing test data (36 portrait samples), three threshold levels are set: Safety level (F < 60): No obvious fatigue, can work normally; Warning level (60≤F<80): Work rhythm needs to be adjusted or energy replenished; Hazard level (F≥80): Work must be stopped and rest required.

Images