Personnel monitoring method and system for limited space and electronic equipment

Abstract

The invention provides a personnel monitoring method and system for a limited space and electronic equipment, and the method comprises the steps: building a low-cost positioning mechanism which does not need extra hardware in a well through employing the regular attenuation characteristic of an inherent structure of the limited space for a wireless signal, and enabling the positioning precision to be improved. The defects of high cost and incapability of remote positioning in the traditional scheme are effectively overcome; by performing fusion analysis on the rough distance data acquired in real time and various sensing data, the crossing from static threshold alarm to dynamic trend early warning is realized, the predictability of potential risks and the timeliness of early warning are remarkably improved, and a more intelligent and reliable safety guarantee is provided for operating personnel in a limited space.

Description

Technical Field

[0001] This invention relates to the technical field of communication well inspection, and more specifically to a method, system, and electronic device for personnel monitoring in confined spaces. Background Technology

[0002] A confined space safety monitoring system is a monitoring system used to monitor the internal conditions of closed or semi-closed hazardous environments (pipelines, storage tanks, underground wells). It relies on the fixed installation of various sensors (e.g., gas detectors, temperature and humidity sensors) inside the space and the transmission of monitoring data (gas concentration, temperature) to an external monitoring center via wired means to achieve centralized monitoring of environmental parameters and alarms for exceeding limits.

[0003] In related technologies, security monitoring systems for confined spaces suffer from technical defects such as a single monitoring dimension and a lack of location context association, which leads to the inability to achieve dynamic security risk assessment based on changes in personnel location. Summary of the Invention

[0004] The purpose of this invention is to overcome the above-mentioned technical deficiencies and provide a method, system and electronic device for personnel monitoring in confined spaces, so as to solve the technical defects of related technologies, such as the single monitoring dimension and lack of location context association in security monitoring systems for confined spaces, which makes it impossible to realize dynamic security risk assessment based on personnel location changes.

[0005] To achieve the above-mentioned technical objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a method for personnel monitoring in a confined space, the method being executed by a monitoring terminal located outside the confined space, the method comprising: Receives wireless signals transmitted by a portable terminal carried by a person via a wireless communication module; wherein the wireless signals carry sensor data packets; Measure the signal strength of the received wireless signal; Based on a preset signal strength-distance mapping relationship, the distance data of the portable terminal entering the confined space corresponding to the signal strength is determined; wherein, the signal strength-distance mapping relationship characterizes the attenuation law of the wireless signal by the confined space structure; Based on the distance data and the sensor data included in the sensor data packet, a security monitoring operation is performed.

[0006] Furthermore, the signal strength-distance mapping relationship is defined by a machine learning model, which is pre-trained from at least training data pairs of signal strength-distance data. The step of determining the distance data of the portable terminal entering the confined space corresponding to the signal strength based on a preset signal strength-distance mapping relationship includes: Based on the signal strength and the machine learning model, the machine learning model outputs a predicted value of the current depth distance to obtain the distance data.

[0007] Furthermore, during pre-training, the machine learning model's input training data also includes at least one of communication well type identifier and / or well metal density identifier.

[0008] Further, the step of obtaining the distance data by outputting a predicted value of the current depth distance from the machine learning model based on the signal strength and the machine learning model includes: Obtain the environmental feature identifier of the current communication well; wherein, the environmental feature identifier includes a communication well type identifier and / or a metal density identifier within the well; The signal strength and the environmental feature identifier are input into the machine learning model, and the machine learning model outputs the distance data.

[0009] Furthermore, the sensing data package includes gas concentration data collected by a gas sensor and / or vital sign data collected by a vital sign sensor; The step of performing security monitoring operations based on the distance data and the sensor data included in the sensor data packet includes: A fusion analysis is performed based on the changing trends of the distance data, the gas concentration data, and / or the vital signs data. When the analysis results indicate a security risk, an alert command is generated and sent.

[0010] Furthermore, the step of fusing and analyzing the trend of change in the distance data with the trend of change in the gas concentration data and / or the trend of change in the vital signs data includes: Determine whether, within a preset time window, the distance data shows a first trend of continuous increase, and whether the gas concentration data shows a second trend of continuous increase; If both the first trend and the second trend are true, then the analysis results indicate a security risk.

[0011] Further, the step of determining whether the distance data shows a first trend of continuous increase and whether the gas concentration data shows a second trend of continuous increase within a preset time window includes: Within the preset time window, the distance data is sampled to generate a first sequence, and the gas concentration data is sampled to generate a second sequence; Calculate the difference between adjacent sample values ​​in the first sequence to obtain the first difference sequence, and calculate the difference between adjacent sample values ​​in the second sequence to obtain the second difference sequence; The number of consecutive positive values ​​in the first difference sequence is counted to obtain the first consecutive positive difference count, and the number of consecutive positive values ​​in the second difference sequence is counted to obtain the second consecutive positive difference count. If the first consecutive positive difference count is greater than or equal to the first threshold, and the second consecutive positive difference count is greater than or equal to the second threshold, then it is determined that the first trend and the second trend are both valid.

[0012] Furthermore, after determining that the first trend and the second trend are simultaneously established, the method further includes: Continue to count the number of consecutive positive differences and the number of consecutive positive differences for the sampling data of subsequent preset rounds; If, within the subsequent preset rounds, the number of consecutive positive errors obtained is consistently not lower than their respective thresholds, a risk confirmation is triggered, and an upgraded warning operation is executed. The upgraded warning operation includes at least one of the following: increasing the level of the warning instruction, shortening the sending interval of the warning instruction, or activating mandatory reporting to the remote monitoring center.

[0013] Secondly, the present invention provides a personnel monitoring system for confined spaces, the personnel monitoring system comprising: A portable terminal is configured to be carried into the confined space by a person, collect sensor data within the confined space, and then transmit the sensor data wirelessly. A monitoring terminal is configured to receive wireless signals transmitted by a portable terminal carried by a person via a wireless communication module; wherein the wireless signals carry sensor data packets; measure the signal strength of the received wireless signals; determine the distance data of the portable terminal penetrating the confined space corresponding to the signal strength based on a preset signal strength-distance mapping relationship; wherein the signal strength-distance mapping relationship characterizes the attenuation law of the wireless signals by the confined space structure; and perform security monitoring operations based on the distance data and the sensor data included in the sensor data packets.

[0014] Thirdly, the present invention provides an electronic device, comprising: a memory, and one or more processors communicatively connected to the memory; the memory stores instructions executable by the one or more processors, the instructions being executed by the one or more processors to cause the one or more processors to implement the method described above.

[0015] Beneficial effects: The personnel monitoring method provided by this invention utilizes the regular attenuation characteristics of wireless signals caused by the inherent structure of confined spaces to establish a low-cost positioning mechanism that requires no additional hardware inside the well. This effectively overcomes the shortcomings of traditional solutions, such as high cost and inability to provide remote positioning. By fusing and analyzing real-time coarse distance data with data from multiple sensors, it achieves a leap from static threshold alarms to dynamic trend warnings, significantly improving the predictability of potential risks and the timeliness of warnings, thus providing more intelligent and reliable safety protection for personnel working in confined spaces. Attached Figure Description

[0016] The accompanying drawings, which form part of this specification, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings: Figure 1 One of the flowcharts for a personnel monitoring method in a confined space provided in this embodiment; Figure 2 This is a second flowchart of a personnel monitoring method for confined spaces provided in this embodiment; Figure 3 A block diagram of a personnel monitoring system for confined spaces provided in this embodiment; Figure 4 This is a block diagram of the electronic device provided in this embodiment. Detailed Implementation

[0017] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0018] The following detailed description is exemplary and intended to provide further detailed explanation of the invention. Unless otherwise specified, all technical terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in this invention is for describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention.

[0019] In related technologies, confined spaces such as communication wells and underground utility tunnels are important components of urban infrastructure. Their internal structures are mostly linear and deep, and contain a large number of metal cables, supports, and equipment. Workers need to enter these spaces periodically for inspection and maintenance.

[0020] The environment inside wells is complex and enclosed, and there is a risk of the accumulation of harmful gases such as methane and hydrogen sulfide or oxygen deficiency, posing a potential threat to the lives of workers. Currently, the typical system architecture for safety monitoring in such environments mainly relies on two types of equipment: a fixed environmental sensor network deployed inside the well and portable gas detectors carried by workers.

[0021] In related technologies, both of the above-mentioned mainstream solutions have significant technical limitations: For fixed sensor networks, the challenge lies in the fact that to fully cover communication wells tens or even hundreds of meters deep, a large number of sensor nodes need to be deployed at intervals within the wells. This results in extremely high hardware costs, wiring costs, and long-term maintenance costs. At the same time, due to the low frequency of maintenance operations, these fixed devices are idle for most of the time, resulting in a significant problem of low utilization and making them difficult to promote economically.

[0022] The main drawback of portable gas detectors is their limited functionality. Existing devices typically only have local audible and visual alarms, which can alert the user but cannot remotely transmit alarm information (especially the real-time location of the person) to monitoring personnel outside the well. This means that if a worker becomes incapacitated deep inside the well due to environmental hazards, those outside cannot determine their exact location, significantly delaying rescue efforts.

[0023] A deeper technical deficiency lies in the fact that, regardless of whether it's a fixed system or a simple portable system, its monitoring dimensions are singular and isolated. It may be able to detect and report gas concentrations, but it completely lacks awareness and understanding of the crucial context of the personnel's "location." The system cannot know "where in the well a certain concentration of harmful gas was detected," let alone dynamically analyze "how environmental risks change as personnel move."

[0024] The root cause of the aforementioned technical deficiencies lies in the fact that the relevant technical solutions have failed to effectively address the core challenge of obtaining real-time personnel location data in the well without significantly increasing costs and system complexity. Traditional precise positioning technologies (such as UWB and Bluetooth beacons) require pre-installed equipment inside the well, which is costly and unsuitable for temporary operations. Therefore, the safety data (mainly gas concentration) collected by existing systems is isolated and lacks spatial correlation, ultimately leading to the technical deficiency of a single monitoring dimension and a lack of location context correlation, making it impossible to achieve dynamic safety risk assessment based on changes in personnel location.

[0025] The present invention aims to solve the above-mentioned technical problems. Its core inventive concept is to establish a mapping relationship between signal strength and personnel depth by utilizing the predictable attenuation law of specific wireless signals (e.g., LoRa) caused by the inherent structure of confined space. This allows for the low-cost acquisition of personnel's approximate location without relying on additional positioning facilities inside the well. This location information is then fused with sensor data to ultimately achieve a dynamic and intelligent safety monitoring and risk assessment operation.

[0026] like Figure 1 As shown, this embodiment provides a method for personnel monitoring in a confined space. The method is executed by a monitoring terminal located outside the confined space, and the method includes: Step S12: Receive a wireless signal sent by a portable terminal carried by a person through a wireless communication module; wherein the wireless signal carries a sensor data packet.

[0027] In this embodiment, the monitoring terminal refers to a dedicated electronic device located outside the confined space for receiving and processing data from a portable terminal.

[0028] In one possible and specific implementation, the monitoring terminal may specifically include: The wireless communication module serves as an interface for establishing a data link between the monitoring terminal and the portable terminal. It includes at least one LoRa receiver for receiving LoRa signals from the portable terminal at a preset frequency and spreading factor, and is capable of measuring Received Signal Strength Indication (RSSI). Furthermore, the wireless communication module may also include a 4G / 5G mobile network module or a Wi-Fi module for uploading data to a remote cloud platform or sending alarm information when an alert is triggered.

[0029] The processing module, which serves as the computing core of the monitoring terminal, can internally store program instructions for executing the methods and signal strength-distance mapping data (this mapping relationship can be represented as a data table or machine learning model parameters). The processor is configured to perform the following operations: parse received sensor data packets to acquire sensor data such as gas concentration and vital signs; read real-time signal strength values ​​from the wireless communication module; convert RSSI values ​​into approximate distance data for the portable terminal based on the pre-stored mapping relationship; run a risk analysis algorithm to perform a time-series fusion analysis of the distance data and sensor data to determine if there is a risk trend of increasing distance and synchronously increasing concentration; and generate corresponding control commands (e.g., triggering an alarm) based on the analysis results.

[0030] The human-machine interface module may further include a display unit, an alarm unit, and an input unit. Specifically, the display unit may be an LCD screen used to display real-time information such as the portable terminal's distance, gas concentration curve, signal strength, battery level, and system status, providing external monitoring personnel with intuitive situational awareness. The alarm unit may include a high-sound-pressure buzzer and a high-brightness LED warning light, used to issue a strong audible and visual alarm when the processor triggers a warning command, attracting the attention of monitoring personnel. The input unit may include physical buttons or a touchscreen for interactive operations such as system power on / off, parameter settings (such as alarm thresholds), and well type selection.

[0031] The power module can power the entire monitoring terminal. It can use a high-capacity lithium battery and is equipped with a power management integrated circuit to meet the needs of long-term outdoor operation.

[0032] In this embodiment, the portable terminal can be referred to as an integrated monitoring device carried by personnel working in a confined space. It is used to collect environmental and personnel data in real time and transmit the data to an external monitoring terminal via wireless communication.

[0033] In one possible and specific implementation, the portable terminal may include: The sensing module may further include a gas sensor and a vital signs sensor. Specifically, the gas sensor can be used to detect and collect ambient gas concentration data in a confined space in real time, such as the concentrations of methane, hydrogen sulfide, carbon monoxide, and oxygen, and can employ a sensor module based on electrochemical or infrared principles.

[0034] Vital signs sensors can be integrated with sensors such as heart rate sensors and blood oxygen sensors to collect some physiological parameter data of workers, providing additional information for safety monitoring.

[0035] The wireless communication module establishes a one-way or two-way wireless data link with an external monitoring terminal. Specifically, it can be a LoRa transmitter or LoRa transceiver, operating in a matched frequency band and configured with a preset spreading factor and coding rate. This wireless communication module can be configured to periodically or event-triggeredly transmit sensor data in a low-power, long-range manner.

[0036] The processing module, which is the control core of the portable terminal, is responsible for scheduling the work of each module. Specifically, it is programmed to periodically wake up the sensing modules to collect data; package the collected sensing data into a sensing data packet containing information such as sensor type, concentration value, timestamp, and device ID; and control the wireless communication module to send the data packet out.

[0037] The power module provides power to the portable terminal.

[0038] The local alarm module may include a vibration motor and a high-brightness LED light. It can be triggered by receiving a warning command from the monitoring terminal or by the processing module (the processing module of the portable terminal) based on the locally collected ultra-high concentration value, and can issue a warning to the wearer in the form of vibration and light.

[0039] In one specific implementation, the portable terminal may be designed in physical form as a handheld device or a wearable device (e.g., fixed to a helmet or shoulder).

[0040] In this embodiment, the confined space can be defined as a closed or semi-closed facility with a complex internal structure, restricted access, poor natural ventilation, not designed for long-term human residence, and potentially hazardous environment.

[0041] Specifically, the confined space can be an underground communication pipeline, such as a communication well, telecommunications pipeline, or fiber optic network maintenance hole. Its interior is often densely packed with metal cables, optical cables, junction boxes, and metal supports, forming an electromagnetic environment that has a regular attenuation characteristic for wireless signals.

[0042] The confined space can be a power cable channel, such as a power cable well, cable tunnel, or substation cable layer. Such spaces also have abundant metallic structures and may pose a risk of accumulating harmful gases due to the aging and decomposition of insulation materials.

[0043] It should be noted that the core characteristic of the confined space lies in the predictability and repeatability of the attenuation pattern of the wireless signal due to its physical structure. Specifically, due to the presence of a large number of regularly arranged metal conductors (cables, supports) inside, electromagnetic waves will produce definite scattering, reflection, and absorption effects when propagating within it. This results in a modelable, regular attenuation of signal strength as the propagation distance increases, rather than a random, chaotic attenuation. This characteristic is the core premise for this invention to pre-establish a "signal strength-distance" mapping relationship and achieve rough positioning based on it. Therefore, the application scenario of this invention is not all enclosed spaces, but rather confined spaces with significant and stable internal structural characteristics (especially metal structures) that enable the wireless signal to exhibit a deterministic attenuation pattern.

[0044] In this embodiment, the wireless communication module of the portable terminal can be a low-power, long-range wireless communication chip and its peripheral circuitry. It can modulate and transmit sensor data packets in the electromagnetic environment of the confined space, ensuring that the signal strength attenuation pattern with distance is predictable and stable, thus meeting the requirement for distance mapping based on signal strength.

[0045] In this embodiment, the wireless communication module of the portable terminal can be a LoRa (Long Range) module.

[0046] In this embodiment, the wireless communication module of the portable terminal can also be a low-power wide-area network (LPWAN) module. Specifically, it can be a Sigfox module, an NB-IoT module, or a LoRaWAN module, etc.

[0047] In this embodiment, the wireless communication module of the portable terminal may also be a proprietary protocol module based on FSK modulation, or a wireless module based on CSS technology, etc.

[0048] Step S14: Measure the signal strength of the received wireless signal.

[0049] In this embodiment, the wireless communication module of the monitoring terminal (e.g., a LoRa receiver), while demodulating the wireless signal and extracting the sensor data packets, can obtain a physical quantity representing the power of the received radio waves, namely the Received Signal Strength Indicator (RSSI), from its internal chip (e.g., the RSSI pin of a LoRa modem chip) or through software instructions. This RSSI value can be a negative value in dBm (e.g., -110 dBm); the larger the absolute value, the weaker the signal strength. This measurement process is a low-level hardware function in the wireless communication system, automatically completed by the communication chip. To ensure the stability and accuracy of the measurement values ​​and avoid errors caused by instantaneous signal fluctuations, the processing module of the monitoring terminal can be configured to perform the following optimization operation: perform multiple RSSI samplings within a short period (e.g., within the reception window of a single data packet or when receiving multiple data packets consecutively). The acquired RSSI sample values ​​are then digitally filtered, for example, by calculating their arithmetic mean or median, to obtain a more stable and reliable final RSSI measurement value that better represents the actual signal strength.

[0050] Step S16: Based on a preset signal strength-distance mapping relationship, determine the distance data of the portable terminal entering the confined space corresponding to the signal strength; wherein, the signal strength-distance mapping relationship characterizes the attenuation law of the wireless signal by the confined space structure.

[0051] In this embodiment, the preset signal strength-distance mapping relationship can be a preset data mapping table. Specifically, the data mapping table can adopt a two-dimensional relational structure, where the first column stores the quantized signal strength value (unit: dBm), and the second column stores the corresponding estimated distance value (unit: meters), representing a full range of coverage from zero distance from the wellhead to the maximum depth of the confined space. The data mapping table can be constructed in any of the following ways: within the target confined space, by actually measuring the signal strength values ​​at different known distance points, original data pairs are formed, and the final mapping table is generated after smoothing; based on the geometric parameters and electromagnetic propagation model of the confined space, the theoretical signal strength at each distance point is calculated through ray tracing simulation to generate the corresponding mapping relationship.

[0052] In this embodiment, the preset signal strength-distance mapping relationship can also be a predictive model built based on machine learning algorithms. This predictive model can automatically learn the complex nonlinear relationship between signal attenuation and propagation distance in a confined space by analyzing communication data collected in historical environments.

[0053] In this embodiment, the preset signal strength-distance mapping relationship can also be a physical attenuation model based on wireless channel propagation theory. This physical attenuation model can directly describe the physical relationship between signal strength and propagation distance using mathematical formulas. Specifically, this physical attenuation model can be a logarithmic distance path loss model, a dual-slope model, or an improved model that considers multipath fading, etc.

[0054] It is understood that the signal strength-distance mapping relationship can be a pre-established data model or data table stored in the monitoring terminal. This model describes the quantitative relationship between the propagation path loss and transmission distance of a wireless signal from the portable terminal's transmitting point to the monitoring terminal's receiving point within a confined space. This mapping relationship characterizes the attenuation law of wireless signals propagating within a confined space. Because such spaces (communication wells) contain a large number of regular metal cables, supports, and equipment, electromagnetic waves undergo reflection, scattering, diffraction, and absorption during propagation, resulting in a non-free-space, but systematic, attenuation of power with increasing distance. This attenuation mode is predictable, repeatable, and modelable due to the relatively fixed layout of the metal structure.

[0055] Step S18: Based on the distance data and the sensor data included in the sensor data packet, perform a security monitoring operation.

[0056] In this embodiment, the monitoring terminal can time-align the received distance data with sensor data (e.g., gas concentration, vital signs) to ensure that the analysis is based on data from the same timestamp or within the same time window. The data is then filtered and smoothed to eliminate transient fluctuations and measurement noise, providing a high-quality data sequence for trend analysis.

[0057] In this embodiment, the monitoring terminal does not judge whether the instantaneous values ​​of distance or concentration exceed the standard in isolation, but analyzes the changing trend of both within a preset time window. That is, it judges whether the trends of "whether personnel are going deeper into the well" (whether the distance data shows a continuous increasing trend) and "whether the environmental risk at their location is increasing synchronously" (for example, whether the concentration of a certain harmful gas shows a continuous increasing trend) occur simultaneously. Once they occur simultaneously, a high-risk mode is determined to be established, and the monitoring terminal immediately generates an early warning command. This early warning command can be sent to a portable terminal carried by personnel to trigger the portable terminal's audible and visual alarm, or it can trigger the local (monitoring terminal) audible and visual alarm, or it can automatically send an alarm message to a remote monitoring center or the mobile terminal of the relevant safety officer through the monitoring terminal's uplink communication interface (4G / 5G module). The message content can include key information such as personnel ID, real-time distance, type of hazardous gas, concentration value, and time of exceeding the standard.

[0058] The personnel monitoring method provided in this embodiment utilizes the inherent attenuation characteristics of wireless signals in confined spaces to establish a low-cost positioning mechanism that requires no additional hardware inside the well. This effectively overcomes the shortcomings of traditional solutions, such as high cost and inability to provide remote positioning. By fusing and analyzing real-time coarse distance data with various sensor data, it achieves a leap from static threshold alarms to dynamic trend warnings, significantly improving the predictability of potential risks and the timeliness of warnings, thus providing more intelligent and reliable safety protection for personnel working in confined spaces.

[0059] In some implementations, the signal strength-distance mapping is defined by a machine learning model, which is pre-trained from at least training data pairs of signal strength-distance data. The step of determining the distance data of the portable terminal entering the confined space corresponding to the signal strength based on a preset signal strength-distance mapping relationship includes: Step S164: Based on the signal strength and the machine learning model, the machine learning model outputs a predicted value of the current depth distance to obtain the distance data.

[0060] In this embodiment, the machine learning model can be pre-trained using training data pairs of signal strength-distance data. Specifically, pre-training can be performed using the following method: First, collect training data.

[0061] Within a confined space of the target type (communication wells of various typical structures), workers carrying a transmitting device of the same model as a portable terminal can transmit signals at known distance points (each sampling point is spaced 1 or 2 meters apart). At the space entrance, a receiving device of the same model as a monitoring terminal records the signal strength (RSSI) value at each distance point. Each sample data is recorded as a (distance value, RSSI value) pair. To improve the model's generalization ability, each sample can also be associated with its corresponding environmental feature identifier, such as the type of communication well (e.g., "fiber optic well," "power well"), metal density level identifier (e.g., "high," "medium," "low"), etc., thus forming an extended training sample: (distance value, RSSI value, well type, metal density).

[0062] Then, the collected raw training data is processed to suit model training.

[0063] Outliers caused by obvious interference (transient occlusion, communication errors) can be removed. Alternatively, adding random Gaussian noise can slightly perturb the original RSSI value, generating more samples to improve the model's robustness. Environmental features (well type, metal density) can also be digitally encoded (e.g., one-hot encoding, numerical grading) and used together with the RSSI value as input features for the machine learning model.

[0064] Finally, the processed training data is used to train the model to be trained, resulting in a prediction model for the distance data.

[0065] Random Forest or Gradient Boosting Decision Tree (GBDT / XGBoost) models can be used as the training model. During training, the processed training data can be divided into training and test sets in a certain ratio (8:2). The selected model can be trained using the training set data. The essence of training is to allow the model to continuously adjust its internal parameters and learn a function F such that for input features X (including RSSI, environmental features, etc.), its predicted value F(X) is as close as possible to the true distance label Y. This optimization process can be accomplished by minimizing the error between the predicted distance and the actual distance (mean squared error MSE).

[0066] In this embodiment, the signal strength can be directly input into the machine learning model to obtain the predicted value of the current depth distance, and the predicted value can be used as the distance data.

[0067] In this embodiment, the machine learning model can be pre-trained not only with training data pairs of signal strength-distance data, but also with training data pairs of environmental feature identifiers-signal strength-distance data.

[0068] This implementation uses a machine learning model to define the signal strength-distance mapping relationship, which can adaptively learn complex signal attenuation patterns in confined spaces. This effectively overcomes the limitations of traditional fixed mapping models when faced with interference such as multipath effects and environmental changes, and significantly improves the accuracy and robustness of distance estimation. At the same time, this data-driven method reduces the dependence on prior environmental knowledge, enhances the adaptability and scalability of the system in confined spaces with different structures, and provides a more reliable location information foundation for security monitoring.

[0069] In some implementations, the training data input to the machine learning model during pre-training also includes at least one of a communication well type identifier and / or a well metal density identifier.

[0070] In this embodiment, the communication well type identifier can be represented as coded information indicating the category to which the current communication well belongs. It is understood that different categories of wells, due to differences in their structural dimensions, internal layout, and typical cable counts, will produce different modes of attenuation to the wireless signal. For example, the numerical code 1 can represent a "straight-line deep well" (simple structure, large depth), code 2 can represent a "well with a branch structure," code 3 can represent a "large underground equipment room," and so on. Similarly, the string code "Power" can represent a power cable well, the code "Fiber" can represent an optical fiber communication well, and the code "Broadcast" can represent a broadcasting line well. Within wells of different functions, the material, diameter, and quantity density of typical cables vary significantly, resulting in different signal attenuation models.

[0071] In this embodiment, the in-well metal density indicator can be represented as a quantitative encoding of the total amount of metal conductors inside the communication well. For example, the numbers 1, 2, and 3 can represent "low metal density," "medium metal density," and "high metal density," respectively. This level can be determined by a professional through visual assessment based on experience.

[0072] This implementation integrates environmental features such as communication well type identifiers and metal density identifiers within the well into the training process of the machine learning model. This enables the model to learn the differentiated impact of different structural features on signal propagation, effectively improving the model's generalization ability and the universality and accuracy of distance prediction across different types of communication wells. It significantly reduces positioning deviations caused by differences in well structure, achieving the technical effect of "one-time training, applicable to multiple wells," and greatly improving the system's deployment convenience and reliability in complex real-world scenarios.

[0073] like Figure 2 As shown, in some embodiments, the step of obtaining the distance data by having the machine learning model output a predicted value of the current depth distance based on the signal strength and the machine learning model includes: Step S1642: Obtain the environmental feature identifier of the current communication well; wherein the environmental feature identifier includes the communication well type identifier and / or the metal density identifier inside the well.

[0074] In this embodiment, the human-machine interface (touchscreen) of the monitoring terminal can provide predefined drop-down menus or selection buttons. Before the operation begins, the monitoring personnel outside the well can manually select the type of the current communication well (e.g., from options such as "electricity well," "telecommunications well," and "integrated utility tunnel") and the metal density level (e.g., from options such as "high," "medium," and "low") based on the markings on the manhole cover or their own experience. The processor of the monitoring terminal can convert the selected option into the corresponding digital code (e.g., converting "electricity well" to code 1 and "metal density - high" to code 3) and temporarily store it in memory for later use.

[0075] Step S1644: Input the signal strength and the environmental feature identifier into the machine learning model, and output the distance data from the machine learning model.

[0076] This implementation method introduces specific environmental feature identifiers of the current communication well during the actual positioning process, enabling the machine learning model to perform adaptive calculations for different well structures and metal densities. This effectively solves the problem of signal attenuation pattern changes caused by differences in different communication well structures. As a result, without the need to retrain the model, the distance prediction accuracy and applicability of a single model in different environments are significantly improved, achieving a precise positioning effect that adapts one model to multiple well types.

[0077] In some embodiments, the sensing data package includes gas concentration data collected by a gas sensor and / or vital sign data collected by a vital sign sensor. The step of performing security monitoring operations based on the distance data and the sensor data included in the sensor data packet includes: Step S182: Perform a fusion analysis based on the changing trends of the distance data, the gas concentration data, and / or the vital signs data; Step S184: When the analysis results indicate a security risk, trigger the generation and sending of an early warning command.

[0078] In this embodiment, the gas concentration data may include: methane concentration, hydrogen sulfide concentration, carbon monoxide concentration, oxygen volume fraction, and inhalable particulate matter concentration, etc.

[0079] In this embodiment, the vital signs data may include: heart rate, body surface temperature, blood oxygen saturation, and exercise status information, etc.

[0080] In one specific and possible implementation, a fusion analysis can be performed based on the changing trends of the distance data and the changing trends of the gas concentration data.

[0081] Specifically, the fusion analysis of distance and gas concentration data trends can capture the spatial distribution characteristics of environmental risks. This means tracking the spatiotemporal correlation between personnel movement and environmental changes. More specifically, the monitoring terminal can analyze whether, within a preset time window, distance data shows a continuously increasing first trend (indicating personnel are venturing deeper into confined spaces), and simultaneously, whether gas concentration data shows a continuously rising second trend. When these two trends are temporally synchronized and statistically significant, the monitoring terminal determines that a high-risk pattern exists where "the deeper one goes, the more dangerous the environment becomes."

[0082] In one specific and possible implementation, a fusion analysis can be performed based on the changing trends of the distance data and the changing trends of the vital signs data.

[0083] Specifically, the fusion analysis of distance data and vital sign data trends can assess anomalies in a person's physiological state within a spatial context. The monitoring terminal not only focuses on whether vital sign data exceeds absolute thresholds but also on abnormal trends that occur as the person's location changes. For example, if a person enters a deep area and their heart rate shows a sustained upward trend, or their blood oxygen saturation shows a sustained downward trend, even if the gas concentration is normal, the monitoring terminal may still classify it as a risk. This can effectively identify physiological stress responses caused by hypoxia, stress, or inhalation of low concentrations of harmful gases, compensating for the limitations of single-environment monitoring and providing more comprehensive physiological safety monitoring for personnel.

[0084] In a specific and possible implementation, a fusion analysis can be performed based on the changing trends of the distance data, the changing trends of the gas concentration data, and the changing trends of the vital signs data.

[0085] Specifically, the fusion analysis based on the changing trends of distance, gas concentration, and vital signs data can integrate the logic of the first two analyses to identify the most urgent and dangerous composite accident signs. The monitoring terminal analyzes the changing trends of the three data points in parallel: continuously increasing distance (deepening investigation), continuously rising specific gas concentration (environmental deterioration), and simultaneous abnormal vital signs (personnel experiencing adverse reactions). When all three trends are met simultaneously, it indicates that an accident may be occurring and has already caused physiological effects on personnel, allowing the monitoring terminal to trigger the highest level of warning. This cross-validation mechanism minimizes false alarms and missed alarms, ensuring the fastest possible emergency response in the most critical situations.

[0086] In this embodiment, the warning instruction can be represented as a structured set of digital commands generated by the processor of the monitoring terminal after determining a safety risk. It may include the following fields: risk event identifier (used to uniquely identify this warning event), risk level (characterizing the severity of the risk), risk type (indicating the cause of the risk, such as "synergistic increase in methane concentration," "abnormal signs of hypoxia," "compound emergency event," etc.), portable terminal ID (identifying which worker is facing the risk), location data (current estimated depth distance), and timestamp (time the warning was triggered).

[0087] In this embodiment, the warning command can be sent to the alarm module on the local monitoring terminal, to the portable terminal, or to the backend server of the remote monitoring center.

[0088] This implementation method achieves a leap from single-parameter threshold judgment to dynamic risk trend assessment by integrating and analyzing personnel location change trends with environmental gas concentration and vital sign data from multiple dimensions. It can identify complex risk patterns such as "dangerous gas concentration increases synchronously as personnel go deeper" or "personnel exhibit physiological abnormalities in dangerous areas" earlier, significantly improving the timeliness and accuracy of early warning and providing more intelligent and forward-looking protection for confined space operations.

[0089] In some embodiments, the step of fusing and analyzing the trend of change in the distance data with the trend of change in the gas concentration data and / or the trend of change in the vital signs data includes: Step S1822: Determine whether the distance data shows a first trend of continuous increase and the gas concentration data shows a second trend of continuous increase within a preset time window.

[0090] In this embodiment, the first trend can be represented as the distance data exhibiting a statistically significant monotonically increasing characteristic within a preset time window. Specific determination methods include, but are not limited to: obtaining a positive slope and a slope value greater than a preset threshold after performing linear regression on the distance data sequence; calculating the difference between consecutive sampling points to obtain a continuously positive difference sequence; or using non-parametric statistical methods to verify the significance of its increasing trend, etc.

[0091] In this embodiment, the second trend can be represented as the gas concentration data also exhibiting a statistically significant monotonically increasing characteristic within the same time period. The judgment criteria can correspond to the first trend. The significance of the trend can be confirmed by the positive slope of the linear regression of the concentration data sequence, the concentration difference sequence that is continuously positive, or statistical hypothesis testing. Furthermore, the time benchmark for trend judgment is synchronized with the first trend to ensure spatiotemporal correlation.

[0092] Step S1824: If the first trend and the second trend are both true, then the analysis result indicates a security risk.

[0093] This implementation method establishes a dual trend correlation judgment mechanism between increasing distance and rising gas concentration, achieving a technological leap from independent parameter threshold monitoring to multi-dimensional dynamic risk perception. It can keenly identify the core danger mode of personnel continuously entering and simultaneously aggravating environmental hazards, effectively overcoming the problem of delayed early warning caused by the lack of location information in traditional monitoring, and significantly improving the predictability and reliability of confined space safety monitoring.

[0094] In some implementations, the step of determining whether the distance data shows a first trend of continuous increase and whether the gas concentration data shows a second trend of continuous increase within a preset time window includes: Step S18222: Within the preset time window, the distance data is sampled to generate a first sequence, and the gas concentration data is sampled to generate a second sequence.

[0095] In this embodiment, instantaneous values ​​of distance data and gas concentration data can be captured at a fixed sampling frequency (e.g., once per second) within a preset time window (e.g., the past 30 seconds). The continuous and flowing data stream can be transformed into a discrete set of data points with equal time intervals, namely, a first sequence (distance sequence) and a second sequence (concentration sequence).

[0096] Step S18224: Calculate the difference between adjacent sampled values ​​in the first sequence to obtain the first difference sequence, and calculate the difference between adjacent sampled values ​​in the second sequence to obtain the second difference sequence.

[0097] In this embodiment, a first-order difference operation can be performed on the first and second sequences generated above, that is, the previous value is subtracted from the next value in the sequence to obtain two new sequences: a first difference sequence (distance difference) and a second difference sequence (concentration difference).

[0098] Each value (positive, negative, or zero) in the difference sequence directly reflects whether the parameter increased, decreased, or remained unchanged between two adjacent sampling times. The advantage of this data processing is that it shifts the focus of analysis from the absolute level of the data to the direction of change, greatly reducing the dependence on the absolute precision of the data itself and highlighting the method's adaptability to coarse distance data.

[0099] In a specific implementation scheme, the first sequence is: [D1, D2, D3, ..., Dn] (where Dn is the latest data); the second sequence is: [C1, C2, C3, ..., Cn] (where Cn is the concentration corresponding to time Dn).

[0100] Perform first-order difference operations on the first and second sequences respectively, that is: For the first sequence: calculate ΔD1=D2-D1, ΔD2=D3-D2, ..., ΔDn-1=Dn-Dn-1 to obtain the first difference sequence [ΔD1, ΔD2, ..., ΔDn-1].

[0101] For the second sequence: the same calculation can be performed to obtain the second difference sequence [ΔC1, ΔC2, ..., ΔCn-1].

[0102] Step S18226: Count the number of consecutive positive values ​​in the first difference sequence to obtain the first consecutive positive difference count, and count the number of consecutive positive values ​​in the second difference sequence to obtain the second consecutive positive difference count.

[0103] In this embodiment, this step does not simply count the total number of positive values ​​in the entire difference sequence, but rather finds and calculates the longest segment with consecutive positive values. For example, a difference sequence of [+1, +2, -1, +3, +4, +5] has 3 consecutive positive differences (composed of the last three +3, +4, +5). The resulting first and second consecutive positive differences measure the inertia of the "uninterrupted and continuous growth" of distance and concentration. This effectively filters out random fluctuations and short-term pullbacks in the data, ensuring that the identified trend is stable and strong.

[0104] Step S18228: If the first consecutive positive difference count is greater than or equal to the first number threshold, and the second consecutive positive difference count is greater than or equal to the second number threshold, then it is determined that the first trend and the second trend are both valid.

[0105] In this embodiment, the two statistics obtained in the previous step (the first and second consecutive positive deviation counts) can be compared with preset first and second count thresholds. These first and second count thresholds can be pre-set; for example, they can be set to require five consecutive increasing sample values. Only when both conditions are met simultaneously is a synergistic trend of "personnel continuing to penetrate deeper" and "gas concentration continuing to rise" ultimately determined.

[0106] This implementation addresses the technical challenge of reliable trend identification in scenarios with limited data precision by introducing a lightweight trend judgment mechanism based on difference sequences and the number of consecutive positive differences. This method transforms complex trend analysis into a continuous statistical analysis of the direction of data change. Through the construction of dual difference sequences and threshold comparison of the number of consecutive positive differences, it achieves the collaborative identification of trends of increasing distance and increasing concentration. This design not only effectively overcomes the failure problem of traditional precise algorithms due to coarse distance data, but also significantly improves the system's anti-interference capability and robustness by focusing on the continuity of the direction of change rather than absolute numerical precision. Furthermore, the computational complexity of this scheme is extremely low, perfectly suited to the resource constraints of embedded devices.

[0107] In some implementations, after determining that the first trend and the second trend are simultaneously established, the method further includes: Continue to count the number of consecutive positive differences and the number of consecutive positive differences for the sampling data of subsequent preset rounds; In this embodiment, the subsequent preset rounds can be represented as the number of times sampling continues after the initial trend is determined to be established (e.g., sampling 5 more times) or a fixed period of time (e.g., monitoring for another 10 seconds). During the period of the subsequent preset rounds, the processor of the monitoring terminal continues to operate and continuously executes the aforementioned sampling, difference calculation, and consecutive positive difference count statistics process.

[0108] If, within the subsequent preset rounds, the number of consecutive positive errors obtained is consistently not lower than their respective thresholds, a risk confirmation is triggered, and an upgraded warning operation is executed. The upgraded warning operation includes at least one of the following: increasing the level of the warning instruction, shortening the sending interval of the warning instruction, or activating mandatory reporting to the remote monitoring center.

[0109] In this embodiment, "consistently not less than" can be expressed as follows: in each round of statistics during the confirmation period (i.e., the period of subsequent preset rounds), the number of consecutive positive differences is greater than or equal to the first threshold, and the number of consecutive positive differences is greater than or equal to the second threshold. Once the above conditions are met, a risk confirmation event flag is generated.

[0110] In this embodiment, in response to this confirmation event, the monitoring terminal can initiate a series of pre-set upgrade operations, including at least one of the following: The level of the warning instruction can be upgraded. Specifically, for example, from a suggestive "Level 1 Warning" to a more serious "Level 2 Warning," or from "Level 2 Warning" to the highest level of emergency, "Level 3 Warning." Different levels correspond to different sound and light effects (e.g., higher pitch and more rapid flashing lights).

[0111] This can shorten the interval between sending early warning commands. Specifically, for example, the interval for reporting the status to the remote monitoring center can be changed from once every 30 seconds to once every 5 seconds, achieving "screen-scrolling" alarms and ensuring that the information is noticed immediately.

[0112] Forced reporting to the remote monitoring center can be activated. Specifically, regardless of whether the reporting function has been enabled before, the complete alarm information packet will be sent to the remote monitoring center unconditionally and immediately through all available network channels (e.g., 4G / 5G) and a receipt will be requested to ensure that the information is delivered.

[0113] This implementation method introduces a risk confirmation and early warning escalation mechanism, establishing a multi-level safety barrier based on the initial trend judgment. When a continuous risk trend is detected, the stability of the trend is continuously verified in subsequent preset rounds, effectively filtering false alarms caused by instantaneous interference and significantly improving the reliability of early warning. At the same time, the dynamic early warning escalation strategy based on the persistence of risk realizes a gradient response from initial warning to mandatory intervention, ensuring that higher-level protective measures can be automatically triggered when the dangerous situation continues to develop, providing a progressive and adaptive important guarantee for the safety of confined space operations.

[0114] like Figure 3 As shown, this embodiment provides a personnel monitoring system for confined spaces, which may include: A portable terminal is configured to be carried into the confined space by a person, collect sensor data within the confined space, and then transmit the sensor data wirelessly.

[0115] In one specific implementation, the portable terminal may include: The sensor module is used to collect gas concentration data and vital sign data of personnel in confined spaces; The first wireless communication module is used to transmit the gas concentration data and the vital signs data of the personnel via wireless signals.

[0116] A first processor, electrically connected to the sensor module and the first wireless communication module, is configured to control the acquisition operation of the sensor module and process the acquired gas concentration data and vital sign data of the personnel before transmitting them through the first wireless communication module.

[0117] A monitoring terminal is configured to receive wireless signals transmitted by a portable terminal carried by a person via a wireless communication module; wherein the wireless signals carry sensor data packets; measure the signal strength of the received wireless signals; determine the distance data of the portable terminal penetrating the confined space corresponding to the signal strength based on a preset signal strength-distance mapping relationship; wherein the signal strength-distance mapping relationship characterizes the attenuation law of the wireless signals by the confined space structure; and perform security monitoring operations based on the distance data and the sensor data included in the sensor data packets.

[0118] In this embodiment, the monitoring terminal is also configured to execute any of the methods described above in the method for monitoring personnel in a confined space.

[0119] In one specific implementation scheme, the monitoring terminal may include: The second wireless communication module is used to receive wireless signals from the portable terminal; Signal strength measurement unit, used to measure the signal strength of received wireless signals; The second processor is electrically connected to the second wireless communication module and the signal strength measurement unit, and it stores a preset signal strength-distance mapping relationship inside. The alarm module, electrically connected to the second processor, is configured to perform an alarm operation in response to a warning command issued by the second processor.

[0120] According to an embodiment of the present invention, an electronic device is provided; please refer to... Figure 4 The electronic device in this embodiment may include one or more of the following components: a processor, a network interface, memory, non-volatile memory, and one or more application programs, wherein the one or more application programs may be stored in non-volatile memory and configured to be executed by one or more processors, and the one or more programs are configured to perform the methods as described in the foregoing method embodiments.

[0121] According to embodiments of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a computer, causes the computer to perform the method described in any of the above embodiments.

[0122] According to embodiments of the present invention, a computer program product comprising instructions is also provided, which, when executed by a computer, cause the computer to perform a method in any of the above embodiments.

[0123] As is known from common technical knowledge, this invention can be implemented through other embodiments that do not depart from its spirit or essential characteristics. Therefore, the disclosed embodiments described above are merely illustrative in all respects and are not the only ones. All modifications within the scope of this invention or its equivalents are included in this invention.

[0124] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0125] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0126] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0127] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1The steps of the function specified in one or more boxes.

[0128] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for personnel monitoring in confined spaces, characterized in that, The method is executed by a monitoring terminal located outside the confined space, and the method includes: Receives wireless signals transmitted by a portable terminal carried by a person via a wireless communication module; wherein the wireless signals carry sensor data packets; Measure the signal strength of the received wireless signal; Based on a preset signal strength-distance mapping relationship, the distance data of the portable terminal entering the confined space corresponding to the signal strength is determined; wherein, the signal strength-distance mapping relationship characterizes the attenuation law of the wireless signal by the confined space structure; Based on the distance data and the sensor data included in the sensor data packet, a security monitoring operation is performed.

2. The method according to claim 1, characterized in that, The signal strength-distance mapping relationship is defined by a machine learning model, which is pre-trained from at least training data pairs of signal strength-distance data. The step of determining the distance data of the portable terminal entering the confined space corresponding to the signal strength based on a preset signal strength-distance mapping relationship includes: Based on the signal strength and the machine learning model, the machine learning model outputs a predicted value of the current depth distance to obtain the distance data.

3. The method according to claim 2, characterized in that, The machine learning model is pre-trained, and the training data input to it also includes at least one of communication well type identifier and / or well metal density identifier.

4. The method according to claim 3, characterized in that, The step of obtaining the distance data by outputting a predicted value of the current depth distance from the machine learning model based on the signal strength and the machine learning model includes: Obtain the environmental feature identifier of the current communication well; wherein, the environmental feature identifier includes a communication well type identifier and / or a metal density identifier within the well; The signal strength and the environmental feature identifier are input into the machine learning model, and the machine learning model outputs the distance data.

5. The method according to claim 1, characterized in that, The sensor data package includes gas concentration data collected by a gas sensor and / or vital sign data collected by a vital sign sensor. The step of performing security monitoring operations based on the distance data and the sensor data included in the sensor data packet includes: A fusion analysis is performed based on the changing trends of the distance data, the gas concentration data, and / or the vital signs data. When the analysis results indicate a security risk, an alert command is generated and sent.

6. The method according to claim 5, characterized in that, The step of fusing and analyzing the trend of the distance data with the trend of the gas concentration data and / or the trend of the vital signs data includes: Determine whether, within a preset time window, the distance data shows a first trend of continuous increase, and whether the gas concentration data shows a second trend of continuous increase; If both the first trend and the second trend are true, then the analysis results indicate a security risk.

7. The method according to claim 6, characterized in that, The step of determining whether the distance data shows a first trend of continuous increase and whether the gas concentration data shows a second trend of continuous increase within a preset time window includes: Within the preset time window, the distance data is sampled to generate a first sequence, and the gas concentration data is sampled to generate a second sequence; Calculate the difference between adjacent sample values ​​in the first sequence to obtain the first difference sequence, and calculate the difference between adjacent sample values ​​in the second sequence to obtain the second difference sequence; The number of consecutive positive values ​​in the first difference sequence is counted to obtain the first consecutive positive difference count, and the number of consecutive positive values ​​in the second difference sequence is counted to obtain the second consecutive positive difference count. If the first consecutive positive difference count is greater than or equal to the first threshold, and the second consecutive positive difference count is greater than or equal to the second threshold, then it is determined that the first trend and the second trend are both valid.

8. The method according to claim 7, characterized in that, After determining that the first trend and the second trend are both true, the method further includes: Continue to count the number of consecutive positive differences and the number of consecutive positive differences for the sampling data of subsequent preset rounds; If, within the subsequent preset rounds, the number of consecutive positive errors obtained is consistently not lower than their respective thresholds, a risk confirmation is triggered, and an upgraded warning operation is executed. The upgraded warning operation includes at least one of the following: increasing the level of the warning instruction, shortening the sending interval of the warning instruction, or activating mandatory reporting to the remote monitoring center.

9. A personnel monitoring system for confined spaces, characterized in that, The personnel monitoring system includes: A portable terminal is configured to be carried into the confined space by a person, collect sensor data within the confined space, and then transmit the sensor data wirelessly. A monitoring terminal is configured to receive wireless signals transmitted by a portable terminal carried by a person via a wireless communication module; wherein the wireless signals carry sensor data packets; measure the signal strength of the received wireless signals; determine the distance data of the portable terminal penetrating the confined space corresponding to the signal strength based on a preset signal strength-distance mapping relationship; wherein the signal strength-distance mapping relationship characterizes the attenuation law of the wireless signals by the confined space structure; and perform security monitoring operations based on the distance data and the sensor data included in the sensor data packets.

10. An electronic device, characterized in that, include: A memory, and one or more processors communicatively connected to the memory; The memory stores instructions that can be executed by the one or more processors to cause the one or more processors to implement the method as described in any one of claims 1 to 8.

Images