Limited space-based operation risk early warning method and system, medium and product

Abstract

The invention discloses an operation risk early warning method and system based on a limited space, a medium and a product, and relates to the field of risk analysis. The method comprises the following steps: firstly, acquiring operation state parameters of operation equipment, and automatically identifying current operation stages with different inherent risk attributes according to statistical characteristics of the operation state parameters; then, calling a preset monitoring strategy matched with the operation stage; then, acquiring current and historical environment monitoring data according to the monitoring strategy, and calculating change trend characteristics of the current and historical environment monitoring data; and finally, performing comprehensive risk judgment by integrating the current monitoring data, the change trend characteristics and the identified operation stage to obtain a current risk state, and sending early warning information when the risk state is early warning or dangerous. According to the method, the operation stage is automatically identified, and the monitoring strategy and the risk model are dynamically adjusted, so that intelligent and prospective early warning of the limited space operation risk is realized, and the effectiveness and the accuracy of safety management are improved.

Description

Technical Field

[0001] This application relates to the field of risk analysis, and in particular to a method, system, medium, and product for early warning of operational risks based on confined spaces. Background Technology

[0002] Work in confined spaces, such as construction inside chemical storage tanks, poses a continuous threat to personnel safety due to the enclosed environment and the variability of hazards. Traditional safety measures mainly rely on static environmental monitoring and external manual supervision before work begins. This approach has limitations in capturing dynamic risks arising from construction activities in real time, which may lead to untimely warnings and thus affect the overall effectiveness of safety assurance.

[0003] To enhance dynamic monitoring capabilities, a dynamic monitoring scheme based on fixed or mobile sensor networks is commonly employed in related technologies. This scheme deploys multiple sensor nodes within a limited space to continuously collect environmental parameters and transmit them in real time. When the monitored values ​​exceed a preset safety threshold, the system automatically triggers an alarm. Compared to traditional methods, this dynamic monitoring approach significantly improves the timeliness of risk detection.

[0004] However, the aforementioned dynamic monitoring solutions still require optimization in practical applications. For example, in operations such as tank corrosion prevention, the core risks differ significantly between different stages, such as rust removal, spraying, and drying. Because the monitoring strategies of related technical solutions are typically static and lack the ability to adaptively adjust based on the operational stage, the system may increase unnecessary energy consumption at one stage due to continuous monitoring of non-core risk indicators, while at another critical stage, the fixed sampling frequency or a single risk model may reduce the sensitivity to early warning of core risks, thus lowering the overall effectiveness and intelligence level of operational risk early warning. Summary of the Invention

[0005] This application provides a method, system, medium, and product for early warning of operational risks in confined spaces, which can improve the accuracy and reliability of early warning of operational risks in confined work spaces.

[0006] Firstly, this application provides a method for early warning of operational risks based on limited space, applied to a server of a risk warning system. The risk warning system includes at least a server and environmental monitoring equipment, comprising: acquiring operational status parameters of the operational equipment; identifying the current operational stage based on the statistical characteristics of the operational status parameters; the operational stage being a pre-divided work process unit with different inherent risk attributes; invoking a preset monitoring strategy corresponding to the operational stage; the monitoring strategy including at least the types of environmental monitoring data to be collected and the data collection frequency under the operational stage; acquiring environmental monitoring data collected by the environmental monitoring equipment at the current moment according to the monitoring strategy; the environmental monitoring data including at least the concentration of harmful gases and oxygen concentration; extracting historical environmental monitoring data within a preset time period and the environmental monitoring data at the current moment, and calculating the trend characteristics of the environmental monitoring data; performing a comprehensive risk judgment based on the environmental monitoring data, the trend characteristics, and the operational stage to obtain the risk status at the current moment; the risk status being the result of determining the safety level of the current operational environment, including safe, warning, and dangerous; and generating and sending warning information to a preset receiving terminal when the risk status is warning or dangerous.

[0007] By adopting the above technical solution, the server first automatically identifies the current operation stage by analyzing the operating status parameters of the equipment. Next, the server can invoke a monitoring strategy perfectly matched to that specific stage. Then, by further analyzing environmental monitoring data and its changing trends, combined with the inherent risk attributes of the operation stage, the system can conduct a more accurate risk assessment, thereby issuing timely warnings before a dangerous situation develops. In summary, this solution improves the accuracy and timeliness of risk warnings, transforming the safety management of confined space operations from a passive response to proactive prevention, and enhancing the safety level of operators.

[0008] In conjunction with some embodiments of the first aspect, in some embodiments, a preset monitoring strategy corresponding to the operation stage is invoked, specifically including: obtaining a preset initial monitoring strategy according to the operation stage; extracting chemical property parameters of the operation materials from a preset material safety database based on the obtained identification information of the operation materials, wherein the chemical property parameters include at least the percentage of volatile organic compounds, flash point temperature value, and toxicity level label; determining the risk level of the operation materials according to the chemical property parameters and the usage amount of the operation materials, wherein the risk level includes low risk, medium risk, and high risk; and adjusting the initial monitoring strategy according to the risk level of the operation materials to obtain an adjusted monitoring strategy.

[0009] By adopting the above technical solution, the server obtains an initial monitoring strategy based on the operational stage and further incorporates adjustments to the initial monitoring strategy based on the operational materials, resulting in a monitoring strategy more adapted to the operational stage. By identifying material information and querying its chemical properties, the server can quantitatively assess the risk level constituted by the material itself (e.g., highly volatile coatings) and its usage. The server dynamically adjusts the initial monitoring strategy based on this risk level, making the formulation of monitoring strategies more refined and personalized. This ensures that monitoring resources can be precisely focused on the core risks generated by specific materials, improving the efficiency and targeting of the risk warning system.

[0010] In conjunction with some embodiments of the first aspect, in some embodiments, the risk warning system further includes a wearable device; after the step of adjusting the initial monitoring strategy according to the risk level of the work materials to obtain the adjusted monitoring strategy, the method further includes: acquiring the current location data and physiological data of the workers collected by the wearable device according to the monitoring strategy.

[0011] By adopting the above technical solution, the server, through the introduction of wearable devices, acquires the location and physiological data of workers, overcoming the limitation of traditional risk assessments that treat workers as passive recipients of environmental influences. The server uses location data to combine risk assessment with the real-time location of personnel, accurately calculating their exposure risks; and uses physiological data to identify the actual impact of risks on the human body. In summary, this solution lays the data foundation for achieving accurate and personalized risk warnings.

[0012] In conjunction with some embodiments of the first aspect, in some embodiments, a comprehensive risk assessment is performed based on environmental monitoring data, trend characteristics, and the stage of operation to obtain the current risk status. Specifically, this includes: calculating the environmental risk level based on environmental monitoring data and trend characteristics, where the environmental risk level is a quantitative value of the overall environmental hazard level of the confined space; calculating the exposure risk level based on the location data of the workers and the environmental risk level, where the exposure risk level is a quantitative value of the environmental hazard level of the workers' current location; calculating the physiological risk level based on physiological data, where the physiological risk level is a quantitative value of the degree of abnormality in the workers' physiological state; weighting and fusing the exposure risk level and the physiological risk level according to the stage of operation to obtain a comprehensive risk score; and comparing the comprehensive risk score with a preset risk status threshold to determine the current risk status.

[0013] By adopting the above technical solution, the server can comprehensively consider environmental monitoring data, trend characteristics, and operational stages to conduct multi-dimensional risk assessments. The server first calculates the environmental risk level, quantifying the overall hazardous degree of the confined space environment. Then, the server combines the location data of the workers to calculate their actual exposure risk level; simultaneously, it analyzes physiological data to assess the degree of abnormality in the workers' physical condition. Next, the server dynamically adjusts the weights of each risk factor according to the characteristics of different operational stages, achieving stage-adaptive risk assessment. In summary, this solution enables risk judgment to effectively identify complex and potential risks, improving the accuracy and reliability of operational risk early warning.

[0014] In conjunction with some embodiments of the first aspect, in some embodiments, the operation status parameters of the working equipment are acquired, and the current operation stage of the work process is identified based on the statistical characteristics of the operation status parameters. Specifically, this includes: acquiring the operation status parameters of the working equipment in real time, the operation status parameters including at least the operating current or power consumption of the working equipment; calculating the statistical characteristics of the operation status parameters, the statistical characteristics including at least the mean and variance; comparing the statistical characteristics of the operation status parameters in adjacent time windows through a preset time window, and determining that the operation stage has switched when the change in the statistical characteristics is greater than a preset change threshold and the duration is greater than a preset duration threshold; after determining that the operation stage has switched, inputting the statistical characteristics in the current time window into a preset operation stage identification model to obtain the current operation stage of the work process.

[0015] By adopting the above technical solution, the server can acquire the operating status parameters of the equipment in real time and identify changes in the work stage by calculating statistical characteristics. The server uses a time window analysis method; a work stage switch is only determined when the statistical characteristic change of adjacent windows exceeds a preset threshold and persists for a certain period, avoiding misjudgments caused by instantaneous fluctuations in equipment load. By inputting the statistical characteristics of the current window into a preset work stage identification model, the server can accurately identify the current work stage without manual intervention. In summary, this solution improves the intelligence level of the risk warning system and solves the problems of lag and inaccuracy caused by reliance on manual reporting in traditional methods.

[0016] In conjunction with some embodiments of the first aspect, in some embodiments, after inputting the statistical features within the current time window into a preset work stage identification model to obtain the work stage of the current work process, the method further includes: acquiring the location data of the workers and identifying the work area of ​​the workers based on the location data; acquiring a preset work stage sequence for the work area, wherein the work stage sequence is a set of multiple work stages arranged sequentially for a specific work area; if the work stage output by the work stage identification model does not belong to the work stage sequence, calculating the feature similarity between the statistical features within the current time window and the preset operating state parameter feature range of each work stage in the work stage sequence; and determining the work stage with the highest feature similarity as the corrected work stage of the current work process.

[0017] By adopting the above technical solution, after identifying the work stage, the server can also identify the work area where the worker is located through the worker's location data and obtain the preset work stage sequence for that work area. When the work stage output by the model does not match the expected sequence, the server will calculate the similarity between the current statistical features and the feature range of each expected stage, and select the work stage with the highest similarity as the correction result. Therefore, this solution solves the problem of stage misjudgment caused by similar equipment operating states. For example, when the current characteristics of rust removal equipment and spraying equipment are similar, the system can use the work area and process sequence for auxiliary judgment to improve the identification accuracy. In summary, this solution improves the robustness and adaptability of the risk warning system, and improves the overall safety management efficiency by optimizing for the inherent risk characteristics of specific work stages.

[0018] In conjunction with some embodiments of the first aspect, in some embodiments, sending early warning information to operators and safety managers based on the risk status specifically includes: when the risk status is a warning, triggering a first-level early warning response, the first-level early warning response including sending a prompt signal to a wearable device worn by the operator; when the risk status is a danger, triggering a second-level early warning response, the second-level early warning response including sending an alarm signal to the wearable device and simultaneously activating an audible and visual alarm device at the entrance of the confined space.

[0019] By adopting the above technical solution, the server can trigger tiered early warning responses based on the severity of the risk. When the risk status is at the warning level, the server triggers a first-level response, sending a notification signal to the wearable devices of workers to attract their attention through slight vibration or low-volume alerts, without interfering with normal operations. When the risk status escalates to danger, the server triggers a second-level response, sending a strong alarm signal to the wearable devices and simultaneously activating the audible and visual alarm devices at the entrance to the confined space, ensuring that both inside and outside personnel are promptly informed of the dangerous situation. In summary, this solution's multi-channel early warning information transmission enhances the reliability and redundancy of the risk early warning system.

[0020] Secondly, this application provides a server for a risk warning system, the server of the risk warning system comprising: one or more processors and a memory; the memory is coupled to one or more processors, the memory is used to store computer program code, the computer program code including computer instructions, and the one or more processors call the computer instructions to cause the server of the risk warning system to perform the method as described in the first aspect and any possible implementation thereof.

[0021] Thirdly, this application provides a computer-readable storage medium including instructions that, when executed on a server of a risk warning system, cause the server of the risk warning system to perform the method described in the first aspect and any possible implementation thereof.

[0022] Fourthly, this application provides a computer program product that, when run on a risk warning system server, causes the risk warning system server to execute the method described in the first aspect and any possible implementation thereof.

[0023] Understandably, the server of the risk warning system provided in the second aspect, the computer program product provided in the third aspect, and the computer storage medium provided in the fourth aspect are all used to execute the method provided in the embodiments of this application. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods, and will not be repeated here.

[0024] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0025] 1. Because it adopts a confined space-based operation risk early warning method, the system can automatically identify the operation stage, dynamically adjust the monitoring strategy, analyze the trend of environmental data changes, and make a comprehensive risk judgment. This effectively solves the problem that static monitoring strategies in related technologies cannot adapt to risk changes in different operation stages, thereby achieving more accurate and timely risk early warning and improving the safety assurance level of confined space operations.

[0026] 2. Due to the adoption of a monitoring strategy adjustment mechanism based on material risk level, the system can dynamically optimize the allocation of monitoring resources according to the chemical property parameters and usage of the materials being worked on. This effectively solves the problems of resource waste and insufficient monitoring of key risks caused by the lack of targeted monitoring strategies in related technologies, thereby achieving more refined and efficient environmental monitoring and providing accurate safety assurance for confined space operations.

[0027] 3. By adopting a technical solution that uses wearable devices to collect the location and physiological data of workers, the system has built a dual monitoring system for both the environment and personnel. This effectively solves the limitation of single monitoring that only focuses on environmental parameters while ignoring the status of personnel in related technologies. As a result, it has achieved a comprehensive and multi-dimensional risk perception capability, enabling the system to simultaneously assess environmental hazards and abnormal physiological conditions of personnel, and providing more comprehensive and humane safety protection. Attached Figure Description

[0028] Figure 1 This is a flowchart illustrating a method for early warning of operational risks based on limited space in an embodiment of this application;

[0029] Figure 2 This is another flowchart illustrating the operation risk warning method based on limited space in the embodiments of this application;

[0030] Figure 3 This is a schematic diagram of the physical device structure of a server for a risk warning system in an embodiment of this application. Detailed Implementation

[0031] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification of this application, the singular expressions “a,” “an,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to any or all possible combinations including one or more of the listed items.

[0032] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0033] To facilitate understanding, the application scenarios of the embodiments of this application are described below.

[0034] In related technologies, dynamic risk monitoring for confined space operations can be achieved by employing static monitoring strategies. This method typically involves pre-setting a fixed network of environmental monitoring sensors within the confined space. For example, during corrosion protection work inside a storage tank, whether in the rust removal stage where dust risks are predominant or the spraying stage where organic gas poisoning risks are predominant, the same set of monitoring parameters and acquisition frequency are used. When any monitored value (such as the concentration of harmful gases) exceeds a single, pre-set, fixed threshold, the system triggers an alarm.

[0035] The confined space-based operational risk early warning method described in this application identifies the current operational stage, enabling intelligent and precise risk warning. It adjusts the monitoring focus based on the specific risks of each stage. For example, in tank corrosion protection operations, the system first acquires the operating status parameters of the equipment and analyzes their statistical characteristics to automatically identify the current operational stage as rust removal. Then, the system calls a preset monitoring strategy corresponding to the operational stage, which focuses on high-frequency monitoring of dust and oxygen concentrations. When the operation switches to the spraying stage, the monitoring strategy automatically adjusts to high-frequency monitoring of volatile organic compound (VOC) concentrations. Simultaneously, the system continuously calculates the changing trend characteristics of environmental monitoring data and performs a comprehensive risk assessment based on the environmental monitoring data, changing trend characteristics, and operational stage. For instance, even if the VOC concentration has not yet exceeded the standard, if its changing trend shows a rapid increase, combined with the current high-risk spraying stage, the system can predict the risk status in advance and send a warning message to a preset receiving terminal.

[0036] As can be seen, the dynamic risk warning method based on the operation stage in this application embodiment can not only achieve real-time monitoring of the risks of confined space operations, but also effectively solve the problems of rigid monitoring strategies, single risk assessment dimensions, and untimely warning of dynamic risk evolution in related technologies. In this way, the accuracy, foresight and intelligence of risk warning are achieved, and the level of operational safety assurance is improved.

[0037] To facilitate understanding, the method provided in this implementation will be described in detail below, using the above scenario as an example. Please refer to [link / reference]. Figure 1 This is a flowchart illustrating a method for early warning of operational risks based on limited space in an embodiment of this application.

[0038] S101. Obtain the operating status parameters of the operating equipment, and identify the current operation stage of the operation process based on the statistical characteristics of the operating status parameters. The operation stage is a process unit with different inherent risk attributes that is pre-divided according to the operation process.

[0039] Among them, "operating equipment" refers to various mechanical equipment that performs operations within a confined space, such as sprayers, grinders, welding machines, and exhaust fans, used to complete specific work tasks; "operating status parameters" refers to data indicators reflecting the working status of the operating equipment, including but not limited to measurable physical quantities such as operating current, power consumption, rotational speed, vibration frequency, and temperature; "statistical characteristics" refers to the characteristic values ​​obtained after mathematical statistical processing of the operating status parameter sequence, such as mean, variance, peak value, kurtosis, and skewness, used to characterize the central tendency and dispersion of the data distribution; "operating process" refers to a set of continuous operation steps required to complete a certain work task within a confined space, such as the tank corrosion prevention operation process which may include steps such as preparation, rust removal, spraying, drying, and cleaning; "operating stage" refers to a relatively independent process unit in the operating process, with each operating stage having different inherent risk attributes, such as the main risks of poisoning and explosion caused by the volatilization of organic solvents in the spraying stage, while the main risks of dust and noise in the rust removal stage.

[0040] The risk warning system server (hereinafter referred to as the server) executes this step after the risk warning system is activated and receives a signal indicating the start of confined space operations: identifying the specific stage of the current operation process. Specifically, the server collects real-time data streams of the operating status parameters of the operating equipment through a sensor network or equipment control system interface connected to the operating equipment. The server preprocesses the collected raw parameter data, including noise reduction, smoothing, and standardization, to improve data quality. Then, within a preset time window (e.g., 30 seconds), the server calculates statistical features of the processed parameter data, such as the average and standard deviation of the current, and the spectral distribution characteristics of the vibration signal. The server inputs the extracted statistical features into a pre-trained operation stage identification model, which can be a rule-based expert system or a machine learning classification model. The identification model matches the input features with preset stage feature patterns and outputs the most likely current operation stage. The server compares the identification result with the stage judgment result from the previous moment. When the judgment results are consistent multiple times (e.g., 3 times), it confirms that the operation stage has changed and updates the current operation stage status in the system.

[0041] S102. Invoke the preset monitoring strategy corresponding to the operation phase. The monitoring strategy shall at least include the types of environmental monitoring data to be collected and the data collection frequency under the operation phase.

[0042] The preset monitoring strategy refers to the environmental parameter monitoring scheme customized for a specific operation phase, which is used to guide the working mode of environmental monitoring equipment. The scheme is set in advance by safety experts based on operation characteristics and risk analysis and stored in the system during the system configuration phase. The environmental monitoring data type refers to the types of environmental parameters that need to be collected, such as oxygen concentration, carbon monoxide concentration, hydrogen sulfide concentration, combustible gas concentration, volatile organic compound concentration, temperature, humidity, etc. The data acquisition frequency refers to the time interval or number of samples per second for sampling specific environmental parameters, such as 0.2Hz (sampling once every 5 seconds), 1Hz (sampling once per second), etc.

[0043] After successfully identifying the current operation stage, the server executes this step, dynamically adjusting the focus and intensity of environmental monitoring based on the inherent risk characteristics of different operation stages. Specifically, based on the current operation stage identifier identified in step S101, the server queries the pre-configured operation stage-monitoring strategy mapping table in the system database (e.g., as shown in Table 1) to obtain the monitoring strategy configuration corresponding to the current operation stage. The server parses the obtained monitoring strategy configuration information, extracting the list of environmental parameter types that need to be monitored and the collection frequency settings for each parameter. The server checks whether the current monitoring strategy is consistent with the monitoring strategy executed at the previous moment; if not, the monitoring configuration needs to be updated. The server sends the parsed monitoring strategy parameters to the edge computing gateway or directly to the environmental monitoring equipment via a network communication interface (such as the MQTT protocol).

[0044] Work phase Inherent risk attributes Core monitoring items sampling frequency Preparation stage Oxygen deficiency / oxygen enrichment residual harmful gases <![CDATA[Oxygen concentration (O2) Total volatile organic compounds (TVOC)]]> Oxygen: 0.2Hz (5 seconds / cycle) TVOC: 0.1Hz (10 seconds / cycle) Rust removal stage Dust hazards (high) Mechanical injury / Noise and oxygen deficiency <![CDATA[Dust concentration (PM10 / TSP), Oxygen concentration (O2)]]> Dust: 1Hz (1 second / cycle) Oxygen: 0.5Hz (2 seconds / cycle) Spraying stage Toxic gas poisoning (high), fire / explosion (high), oxygen deficiency <![CDATA[Volatile Organic Compounds (VOCs), Combustible Gas Concentration (LEL), Oxygen Concentration (O2)]]> VOCs / LEL: 2Hz (0.5 seconds / cycle) Oxygen: 1Hz (1 second / cycle) Drying / curing stage Fire risk due to continuous volatilization of toxic gases Volatile organic compounds (VOCs) temperature (Temp) VOCs: 0.2Hz (5 seconds / cycle) Temperature: 0.1Hz (10 seconds / cycle) Cleaning and Inspection Phase Residual harmful gases and oxygen deficiency <![CDATA[Oxygen concentration (O2) Total Volatile Organic Compounds (TVOC)]]> Oxygen: 0.5Hz (2 seconds / cycle) TVOC: 0.2Hz (5 seconds / cycle)

[0045] Table 1

[0046] S103. According to the monitoring strategy, obtain the environmental monitoring data collected by the environmental monitoring equipment at the current moment. The environmental monitoring data shall include at least the concentration of harmful gases and the oxygen concentration.

[0047] Among them, environmental monitoring equipment refers to various sensor devices deployed in a confined space to collect environmental parameters, such as gas detectors, temperature and humidity sensors, dust sensors, etc.; environmental monitoring data refers to the numerical values ​​of various parameters reflecting the environmental conditions within a confined space; harmful gas concentration refers to the content of gases that are potentially harmful to human health in the air, such as the concentration values ​​of carbon monoxide (CO), hydrogen sulfide (H2S), volatile organic compounds (VOCs), etc., usually expressed in ppm (parts per million) or mg / m³; oxygen concentration refers to the proportion of oxygen in the air, usually expressed as volume percentage (Vol%). The normal atmospheric oxygen concentration is about 20.9%. When it is below 19.5%, the human body will feel hypoxia, and when it is below 16%, there will be serious health risks.

[0048] After the monitoring strategy is determined and issued, the server executes this step, acquiring real-time environmental data collected by environmental monitoring equipment according to the monitoring strategy optimized for the current operational phase. Specifically, the server continuously listens for data reports from environmental monitoring equipment or edge gateways via a network communication interface. Based on the received monitoring strategy instructions, the environmental monitoring equipment adjusts its internal sampling configuration, including enabling or disabling specific sensors and setting the sampling frequency of different sensors. The environmental monitoring equipment collects environmental parameters at the configured frequency, packages the collected data, adds metadata such as timestamps and device IDs, and sends it to the server via a preset communication protocol (such as MQTT, HTTP, etc.). Upon receiving the data packet, the server performs data parsing and preliminary verification, checking the data integrity, timestamp validity, and data value rationality. The server performs necessary unit conversions and standardization on the verified data to ensure data format consistency. The server associates the processed environmental monitoring data with the current timestamp and stores it in the system's time-series database, while maintaining a cached copy of the current data for immediate risk assessment calculations.

[0049] S104. Extract historical environmental monitoring data within a preset time period and environmental monitoring data at the current moment, and calculate the changing trend characteristics of the environmental monitoring data.

[0050] The preset time period refers to a time window pre-set for trend analysis. It is usually set according to the changing characteristics of different environmental parameters and the risk characteristics of the operation stage. For example, it can be set to 3-5 minutes for rapidly changing volatile organic compound concentrations and 5-10 minutes for relatively slowly changing oxygen concentrations. Historical environmental monitoring data refers to the sequence of environmental parameter values ​​collected and stored before the preset time period. The trend characteristics represent the changing patterns and characteristics of environmental monitoring data in the time dimension, including but not limited to the rate of change (first derivative), the acceleration of change (second derivative), and volatility indicators (such as standard deviation and coefficient of variation), which are used to predict the future trend of environmental parameters and assess the rate of risk accumulation.

[0051] After acquiring the environmental monitoring data for the current moment, the server performs this step to analyze the historical trends of environmental parameters. Specifically, the server calculates the starting time of the historical data based on the current moment and the preset time window length. The server queries and extracts historical environmental monitoring data from the time-series database and preprocesses the extracted historical data. The server merges the preprocessed historical data with the newly acquired environmental monitoring data for the current moment to form a complete time series. The server applies mathematical models and statistical methods to the merged time series data to calculate the trend characteristics of each environmental parameter. For key parameters (such as the concentration of harmful gases), the server calculates its first derivative (rate of change) and obtains the rate of change of the parameter over time using the difference method or linear regression method, with the unit being "parameter unit / time" (e.g., ppm / min). The server further calculates the second derivative (acceleration of change) to reflect the change in the rate of change of the parameter, used to determine whether the risk is accumulating rapidly or tending to stabilize. The server also calculates the volatility indicators of the parameter within the time window, such as standard deviation or coefficient of variation, to assess the stability of the environmental state. Finally, the server organizes the calculated trend characteristic values ​​into structured data as input for subsequent risk assessment.

[0052] S105. Based on environmental monitoring data, trend characteristics, and the stage of operation, a comprehensive risk assessment is conducted to obtain the current risk status. The risk status is the result of the judgment of the safety level of the current working environment, including safety, warning, and danger.

[0053] Among them, comprehensive risk assessment refers to the integrated analysis of multiple risk factors and information sources to obtain a comprehensive evaluation result of the current safety status of the working environment; risk status refers to the judgment result of the current safety level of the working environment, which is divided into three levels: safe status means that the current environmental parameters are within the normal range and the trend of change is stable, and the operation can be carried out normally; warning status means that the environmental parameters are close to but have not yet exceeded the danger threshold, or although they are within the safe range, the trend of change indicates that they may reach the danger level in the short term, requiring attention and preventive measures; danger status means that the environmental parameters have exceeded the safety threshold, or although they have not exceeded it, the trend of change is extremely unfavorable, and it is expected to reach the danger level in a very short time, requiring immediate emergency measures.

[0054] After acquiring environmental monitoring data and its changing trend characteristics, the server executes this step. By comprehensively considering current environmental parameter values, parameter changing trends, and the inherent risk characteristics of the work phase, it conducts a comprehensive assessment of the safety status of the current work environment. Specifically, the server first loads the risk assessment model parameters for the currently identified work phase from the configuration library, including the safety threshold, warning threshold, and danger threshold for each environmental parameter, as well as the weighting coefficient of each parameter in that phase. The server compares the current environmental monitoring data with the corresponding safety thresholds and calculates the absolute risk score for each parameter. For hazardous gas concentrations, the server uses a piecewise function to calculate its absolute risk score: a lower score when the concentration is within the safe range, a medium score when approaching the warning threshold, and the highest score when exceeding the danger threshold. The server evaluates the trend risk score for each parameter based on the changing trend characteristics calculated in step S104. For parameters with a positive and large rate of change (such as a rapid increase in hazardous gas concentration), the server assigns a higher trend risk score; for parameters with a negative rate of change (such as a decrease in concentration), the server assigns a lower trend risk score. The server considers the inherent risk level of the current work phase and sets a basic risk score for the risk assessment. For example, the spraying stage involves flammable and explosive substances, so its basic risk score is higher than that of the preparation stage. The server adds the basic risk score, the absolute value risk score of each parameter (multiplied by its corresponding weight), and the trend risk score (multiplied by its corresponding weight) to obtain a comprehensive risk score. Based on a preset risk grading standard, the server maps the comprehensive risk score to three risk levels: safe, warning, or dangerous. For example, a comprehensive risk score below 40 is considered safe, 40-70 is considered warning, and above 70 is considered dangerous.

[0055] Optionally, in some embodiments, the server may employ a risk assessment method based on weighted scoring. The server first sets a base risk score for the current operational stage, reflecting the inherent risk level of that stage. The server calculates an absolute risk score for each environmental monitoring parameter, using a piecewise function to map the actual parameter value to a score range of 0-100. The server calculates a trend risk score for the changing trend characteristics of each parameter; parameters with greater change rates and accelerations receive higher trend risk scores. Based on the characteristics of the current operational stage, the server assigns different weight coefficients to different parameters, focusing on the core risk indicators of that stage. The server calculates the total risk score: Total Risk Score = Base Risk Score + Σ(Absolute Risk Score of Parameter i × Weight i) + Σ(Trend Risk Score of Parameter j × Weight j).

[0056] S106. When the risk status is warning or danger, generate and send warning information to the preset receiving terminal.

[0057] Among them, the preset receiving terminal refers to various devices or systems pre-configured to receive early warning information, including but not limited to smart wearable devices worn by workers (such as smart safety helmets and smart bracelets), mobile terminals used by safety management personnel (such as smartphones and tablets), on-site sound and light alarm devices, and displays in the safety monitoring center; early warning information refers to structured information generated by the system that includes risk status, risk source, risk level, and response suggestions, used to notify relevant personnel of the current risk situation and guide them to take corresponding measures.

[0058] The server executes this step after determining the risk status as a warning or danger, promptly transmitting risk information to relevant personnel, triggering the corresponding level of emergency response, preventing further escalation of the risk, and ensuring the safety of workers.

[0059] In this embodiment, a confined space-based operation risk early warning method is adopted, including a complete technical solution that acquires the operating status parameters of the operating equipment to identify the operation stage, calls the corresponding monitoring strategy, acquires environmental monitoring data, calculates the trend characteristics of change, makes a comprehensive risk judgment, and sends early warning information. Therefore, the system can realize automatic identification of the operation stage, dynamic adjustment of the monitoring strategy, trend prediction of environmental risks, and multi-dimensional risk assessment. This effectively solves the problems of static and fixed monitoring strategies, single-dimensional risk assessment, and untimely early warning in related technologies, thereby realizing intelligent, accurate, and forward-looking risk early warning, and improving the safety assurance level and emergency response efficiency of confined space operations.

[0060] Based on the above embodiments, the method provided in this embodiment will be described in further detail below. Please refer to... Figure 2 This is another flowchart illustrating the operation risk warning method based on limited space in this application embodiment.

[0061] S201. Obtain the operating status parameters of the operating equipment, and identify the current operation stage based on the statistical characteristics of the operating status parameters. The operation stage is a process unit with different inherent risk attributes that is pre-divided according to the operation process.

[0062] This step specifically includes:

[0063] Real-time acquisition of operating status parameters of the work equipment, including at least the operating current or power consumption of the work equipment;

[0064] Calculate the statistical characteristics of the operating status parameters, including at least the mean and variance;

[0065] By comparing the statistical characteristics of the operating status parameters of adjacent time windows through a preset time window, when the change in the statistical characteristics is greater than the preset change threshold and the duration is greater than the preset duration threshold, it is determined that the operation phase has switched.

[0066] After determining that a job stage has switched, the statistical features within the current time window are input into the preset job stage identification model to obtain the job stage in which the current job process is located.

[0067] Among them, the change in statistical features refers to the degree of difference in statistical feature values ​​between adjacent time windows; the preset change threshold represents the critical value for judging that the statistical feature has changed significantly. Its setting scheme is determined through historical data analysis or expert experience, and can be set to specific values ​​such as the mean change exceeding 30% or the variance change exceeding 50%, to filter normal fluctuations and significant changes; the preset duration threshold represents the minimum duration for which the statistical feature change needs to continue. Its setting scheme is determined according to the start-up and shutdown characteristics and working modes of different operating equipment, and is usually set to 3-5 consecutive time windows to avoid misjudgments caused by instantaneous fluctuations; the operation stage identification model refers to the algorithm model used to map the statistical features of equipment operating status parameters to predefined operation stages, such as machine learning models such as support vector machines (SVM), random forests, or neural networks.

[0068] After the risk warning system is activated and a signal indicating the start of confined space operations is received, the server executes this step to identify the current stage of the operation. Specifically, the server first continuously collects the operating current or power consumption data of the operating equipment at a preset sampling frequency (e.g., 10 times per second) through a current sensor, power meter, or equipment control system interface connected to the operating equipment. The server preprocesses the collected raw data, dividing the preprocessed data into segments according to preset time windows (e.g., 30 seconds), and calculates statistical characteristics for the data within each time window. These characteristics mainly include the mean (reflecting the average level of equipment load) and variance (reflecting the degree of fluctuation in equipment load), and may also include auxiliary characteristics such as maximum, minimum, and kurtosis. The server compares the statistical characteristics of the current time window with those of the previous time window to calculate the change in characteristics. The server determines whether the calculated change in characteristics exceeds a preset threshold and records the number of time windows that consecutively exceed the threshold. When the number of time windows that consecutively exceed the preset threshold reaches a preset duration threshold (e.g., 3 consecutive windows), the server determines that the operation stage has switched. After confirming the switch to a new work phase, the server inputs the statistical feature vector calculated within the current time window into a pre-trained work phase recognition model. This model analyzes the patterns in the feature vectors, matches them to the most likely work phase category (such as preparation, rust removal, or painting), and outputs the recognition result.

[0069] Optionally, in some embodiments, the training process of the job phase identification model includes:

[0070] The server first collects a large amount of historical job data, including time series of operating status parameters for different equipment at various known job stages, along with corresponding manually labeled job stage tags. The server cleans and preprocesses the collected raw data to ensure data quality. The server then segments the preprocessed data according to preset time windows and calculates statistical features for each time window, constructing feature label pairs. Finally, the server divides the constructed dataset into training and testing sets in approximately a 7:3 ratio, with the training set used for model learning and the testing set used to evaluate model performance.

[0071] The server performs feature engineering on the training set, including feature extraction, feature selection, and feature transformation, to improve the model's discriminative ability. The server selects multiple candidate machine learning algorithms, such as Support Vector Machine (SVM), Random Forest, Gradient Boosting Decision Tree (GBDT), and Multilayer Perceptron (MLP), and trains an initial model on the training set. The server uses cross-validation to optimize the hyperparameters of each candidate model to find the optimal parameter configuration. The server evaluates the performance of each optimized candidate model on the test set, calculating evaluation metrics such as accuracy, precision, recall, and F1 score. The server selects the best-performing model as the final recognition model for the job stage and serializes and saves it.

[0072] The server deploys the trained model to the production environment and sets up a regular evaluation and update mechanism to ensure the model's continued effectiveness.

[0073] Optionally, to further improve the accuracy of work stage identification, when the work stage output by the identification model has uncertainty or does not conform to the preset process, this application embodiment also provides a correction mechanism based on work area and process sequence, specifically including:

[0074] S202. Obtain the location data of the workers and identify their work areas based on the location data.

[0075] The work area refers to a specific functional zone or work area within a limited space, such as the bottom area, side wall area, and top area of ​​a storage tank. Different areas may have different environmental characteristics and risk factors.

[0076] After acquiring the operator's location data, the server performs this step to determine the specific work area where the operator is currently located. Specifically, the server first receives real-time location data from the wearable device and matches the location coordinates with a pre-established finite-space 3D model in the system. This finite-space 3D model includes the boundary definitions and attribute information of each functional area. The server uses a spatial inclusion relationship judgment algorithm to determine the specific work area to which the operator's current location coordinates belong. For example, if the coordinate values ​​meet the spatial range constraints of the bottom area of ​​the storage tank, the operator is determined to be located in the bottom area of ​​the storage tank.

[0077] Optionally, the server can also perform smoothing processing by combining historical trajectory data to avoid frequent switching of area judgments due to instantaneous fluctuations in location data. The server associates the recognition results with timestamps to record the time that workers stay in different work areas.

[0078] S203. Obtain the preset work stage sequence for the work area. The work stage sequence is a set of multiple work stages arranged in sequence for a specific work area.

[0079] The operation phase sequence represents a set of multiple operation phases predefined for a specific operation area and arranged in the normal working process order, reflecting which work contents should be performed in what order within the area; the setting scheme for this operation phase sequence can be achieved by analyzing historical operation records and process flow diagrams, extracting typical operation phases and their normal execution order for each operation area, forming structured sequence data and storing it in the system configuration library.

[0080] After identifying the work area where the operator is located, the server executes this step to obtain the standard sequence of stages in the normal work process for that work area. Specifically, based on the work area identifier identified in step S207, the server queries the system configuration database to obtain the preset work stage sequence associated with the current work area. This work stage sequence is usually stored in the form of an ordered list, with each list item containing information such as the work stage identifier, stage name, stage description, expected duration range, and dependencies of preceding stages. The server parses the obtained work stage sequence data, extracts the set of all work stage identifiers allowed to be executed in the current work area, and the normal execution order of these stages. For example, for the bottom area of ​​a storage tank, the normal work stage sequence might be "Preparation stage → Rust removal stage → Bottom anti-corrosion coating application stage → Coating curing stage → Inspection stage → Cleaning stage".

[0081] Optionally, the server can also extract the timing constraints between each job stage, such as certain stages must start within a specific time window after the previous stage is completed, or the duration of certain stages must meet minimum or maximum limits.

[0082] S204. If the work stage output by the work stage identification model does not belong to the work stage sequence, calculate the feature similarity between the statistical features within the current time window and the preset operating state parameter feature range of each work stage in the work stage sequence.

[0083] Among them, feature similarity represents the degree of matching between the statistical characteristics of the currently observed equipment operating status parameters and the preset feature range; the preset operating status parameter feature range refers to the effective value range of the statistical characteristics of the equipment operating status parameters predefined for each operation stage. The setting scheme is to analyze historical operation data, extract the statistical distribution characteristics of the equipment operating status parameters in each operation stage, and determine the normal fluctuation range of statistical characteristics such as mean and variance. It is usually set as mean ± 2 times standard deviation or 5%~95th percentile interval.

[0084] This step is executed when the server detects a mismatch between the output of the job phase identification model and the allowed phase sequence for the job area. Feature similarity analysis is used to find the most likely actual job phase. Specifically, the server first checks whether the job phase identifier output by the job phase identification model in step S201 is included in the job phase sequence obtained in step S203. When the job phase output by the model is found to be outside the allowed sequence, the server initiates the feature similarity analysis process. The server loads the preset operating state parameter feature ranges for each phase in the job phase sequence from the system configuration library. These ranges are typically stored as multi-dimensional feature vectors, containing expected intervals for multiple statistical features such as mean, variance, and peak value. The server extracts the statistical feature vector of the equipment operating state parameters calculated within the current time window. This feature vector is identical to the feature vector input to the job phase identification model in step S201. The server calculates the similarity or distance metric between the current feature vector and the preset feature ranges for each phase in the job phase sequence. Commonly used metric methods include Euclidean distance, Mahalanobis distance, and cosine similarity. The server normalizes the calculated similarity or distance values, converting them into similarity scores in the range of 0-1, with higher values ​​indicating higher similarity.

[0085] Optionally, the setting scheme for the preset operating state parameter characteristic range can be based on probabilistic modeling using a Gaussian mixture model, including:

[0086] First, a large amount of historical operational status data containing accurate operational stage labels is collected, and statistical features (mean, variance, etc.) are extracted for each time window to form a multidimensional feature space.

[0087] Then, for all feature points belonging to the same work phase, GMM is used for fitting. GMM can treat these data points as a mixture of one or more Gaussian distributions, thereby capturing the complex patterns of equipment operation in that phase (for example, there may be two sub-patterns in the same spraying phase: "high-power spraying" and "low-power touch-up").

[0088] Finally, the preset operating state parameter feature range is no longer a simple rectangular interval (such as mean ± 2 standard deviations), but is defined as the region in the GMM model where the posterior probability of a new feature point belonging to the model is greater than a certain confidence threshold (such as 0.8). When calculating feature similarity, the server directly calculates the posterior probability of the statistical feature vector of the current time window generated by the GMM models of different job stages, and the job stage corresponding to the model with the highest probability value is the stage with the highest similarity.

[0089] S205. The task stage with the highest feature similarity is determined as the task stage in the current corrected task process.

[0090] Specifically, the server sorts the similarity scores of each job stage in descending order and identifies the job stage with the highest similarity score as a candidate result for subsequent stage corrections. The server may also consider the time factor, checking the duration from the last confirmed job stage to the current time, and assessing the time rationality of transitioning to the stage with the highest feature similarity. The server determines the job stage with the highest feature similarity score as the job stage in the corrected current job flow and updates the job stage state variable in the system.

[0091] S206. Invoke the preset monitoring strategy corresponding to the operation phase. The monitoring strategy shall at least include the types of environmental monitoring data to be collected and the data collection frequency under the operation phase.

[0092] This step specifically includes:

[0093] The initial monitoring strategy is pre-set based on the operation phase;

[0094] Based on the identification information of the work materials, the chemical property parameters of the work materials are extracted from the preset material safety database. The chemical property parameters include at least the percentage of volatile organic compounds, flash point temperature value, and toxicity level label.

[0095] The risk level of the work materials is determined based on the chemical property parameters and the amount of work materials used. The risk levels include low risk, medium risk, and high risk.

[0096] Based on the risk level of the materials being handled, the initial monitoring strategy is adjusted to obtain the adjusted monitoring strategy.

[0097] The preset initial monitoring strategy refers to the basic environmental monitoring plan pre-configured for a specific operation phase, including the initial settings of the types of environmental parameters to be monitored and the sampling frequency; the operation materials refer to the chemical substances or materials used or handled during the operation, such as paints, solvents, cleaning agents, and preservatives; the material identification information refers to the code or label used to uniquely identify the operation materials, usually contained on the material packaging or in electronic tags, such as barcodes, QR codes, and RFID tags; the preset material safety database refers to a structured data set storing safety information of various operation materials, including information such as the chemical composition, physical properties, hazardous characteristics, and safe operation guidelines of the materials; the flash point temperature value represents the lowest temperature at which a material can generate enough vapor to form a flammable mixture with air and be ignited under standard test conditions, usually in degrees Celsius, and is an important indicator for assessing the fire hazard of materials; the toxicity level label refers to the classification mark according to the degree of harm of materials to human health, usually divided into low toxicity, moderate toxicity, and high toxicity levels; the risk level of the operation materials refers to the degree of danger determined by comprehensively considering the chemical properties and usage of the materials, used to guide the strength of safety protection measures and the strictness of monitoring strategies.

[0098] After successfully identifying the current work stage, the server executes this step, dynamically adjusting the focus and intensity of environmental monitoring based on the work stage and the characteristics of the materials used. Specifically, the server first queries the system database based on the current work stage identifier identified in step S201 to obtain the initial monitoring strategy configuration corresponding to the current work stage. The initial monitoring strategy typically includes a list of environmental parameter types that need to be monitored in this stage and the default sampling frequency for each parameter. The server receives the identification information of the work materials from the field. This information can be obtained by scanning the barcodes or QR codes on the material packaging with a handheld scanning device, automatically collected by an RFID reader, or manually entered by the operator through a terminal. The server uses the obtained material identification information as a search key to query a preset material safety database (such as a hazardous chemicals database or the company's internal material safety data sheet) to extract detailed chemical property parameters of the material. The extracted parameters include at least the percentage of volatile organic compounds (VOCs) content (e.g., 45% VOC content in paint), flash point temperature (e.g., 23°C flash point for solvents), and toxicity level (e.g., moderate toxicity). The server also obtains the material usage information declared for the current work, such as a planned use of 20 kg of paint. The server calculates the overall risk level of the material based on the extracted chemical property parameters and usage amount, applying preset risk assessment rules. The assessment rules typically employ a scoring system or decision tree method. For example, a material is classified as high-risk if its VOC content is >40%, flash point <30℃, and usage amount >10 kg; medium-risk if its VOC content is between 20% and 40% or its flash point is between 30℃ and 60℃, and usage amount between 5 and 10 kg; and low-risk if all other conditions are met.

[0099] Based on the calculated material risk level, the server modifies the initial monitoring strategy according to preset adjustment rules. These rules may include: for high-risk materials, adding monitoring parameter types (e.g., adding monitoring of specific hazardous gases), increasing sampling frequency (e.g., increasing VOCs sampling frequency from once every 10 seconds to once every 5 seconds), and lowering warning thresholds (e.g., lowering the VOCs warning threshold from 50 ppm to 30 ppm); for medium-risk materials, moderately adjusting monitoring parameters and frequencies; and for low-risk materials, maintaining the initial monitoring strategy. The server generates the final adjusted monitoring strategy, including a complete list of environmental parameters to be monitored and the sampling frequency settings for each parameter, and saves it as the currently effective monitoring strategy configuration.

[0100] Optionally, the preset adjustment rules can be set by establishing a parameterized strategy adjustment function. This strategy adjustment function takes the material risk level (low, medium, high) as input and outputs adjustment coefficients for various parameters in the initial monitoring strategy (such as collection frequency and early warning threshold).

[0101] Let the baseline value of a certain parameter in the initial monitoring strategy be P0 (for example, the initial acquisition frequency is 0.1Hz), and the adjusted value be P_adj.

[0102] The adjustment rule function can be designed as: P_adj = P0 × f(RiskLevel), where f(RiskLevel) is a piecewise function, including:

[0103] If the material risk level is high risk

[0104] The adjustment coefficient f(High) = k_high (for example, k_high = 2 indicates that the frequency is doubled). For the warning threshold, it is P_adj = P0 × (1 / k_high), that is, the threshold is halved.

[0105] If the material risk level is medium risk

[0106] Then the adjustment factor f(Medium) = k_medium (for example, k_medium = 1.5).

[0107] If the material risk level is low risk

[0108] Then the adjustment factor f(Low) = k_low (for example, k_low = 1, that is, no adjustment).

[0109] The specific value of the adjustment coefficient k can be determined through safety specifications or simulation. The goal is to ensure that the warning time (Time-to-Alarm) of the risk warning system can be maintained within an acceptable safety margin under different risk levels.

[0110] S207. According to the monitoring strategy, obtain the current location data and physiological data of the workers collected by the wearable device;

[0111] Wearable devices refer to smart equipment worn by workers that have data acquisition and communication functions, such as smart safety helmets, smart bracelets, and smart vests; location data refers to coordinate information that characterizes the spatial position of workers in a limited space, usually including three-dimensional coordinates (x, y, z) or distance and direction relative to a specific reference point, used to determine the precise location of workers in hazardous areas; physiological data refers to biomedical parameters that reflect the physical condition of workers, including but not limited to heart rate, body temperature, blood oxygen saturation, respiratory rate, and skin conductance, used to monitor the health status and physiological stress response of workers.

[0112] After adjusting the monitoring strategy, the server executes this step to obtain the real-time location and physiological status information of the workers. Specifically, the server first parses the adjusted monitoring strategy configuration to determine the data types (such as location, heart rate, body temperature, etc.) and sampling frequency to be collected from the wearable device (e.g., location data updated every 2 seconds, heart rate data updated every 5 seconds). The server establishes a data connection with the smart device worn by the workers through a preset communication protocol (such as BLE, Wi-Fi, ZigBee, or a dedicated low-power wide-area network). The wearable device has multiple built-in sensors, including a UWB (ultra-wideband) module or Bluetooth beacon receiver for positioning, and photoplethysmography (PPG) sensors, temperature sensors, accelerometers, etc., for physiological monitoring. Based on the received sampling configuration, the wearable device activates the corresponding sensors and sets the sampling frequency to begin data collection. The wearable device packages the collected location and physiological data, adds metadata such as device ID and timestamp, and sends it to the server via a wireless communication network. The server receives and parses data packets, verifies the integrity and timeliness of the data, stores valid data associated with the current timestamp, and retains a cached copy of the current data for immediate risk assessment calculations.

[0113] S208. According to the monitoring strategy, acquire the environmental monitoring data collected by the environmental monitoring equipment at the current moment. The environmental monitoring data shall include at least the concentration of harmful gases and the concentration of oxygen.

[0114] S209. Extract historical environmental monitoring data within a preset time period and environmental monitoring data at the current moment, and calculate the changing trend characteristics of the environmental monitoring data;

[0115] Steps S208, S209 and Figure 1 The steps S103 and S104 of the above embodiments are described similarly and will not be repeated here.

[0116] S210. Based on environmental monitoring data, trend characteristics, and the stage of operation, a comprehensive risk assessment is conducted to obtain the current risk status. The risk status is the result of determining the safety level of the current working environment, including safety, warning, and danger.

[0117] This step specifically includes:

[0118] Based on environmental monitoring data and changing trend characteristics, the environmental risk level is calculated. The environmental risk level is a quantitative value of the overall environmental hazard level of a confined space.

[0119] Based on the location data of the workers and the environmental risk level, the exposure risk level is calculated. The exposure risk level is a quantitative value of the degree of environmental hazard at the current location of the workers.

[0120] Based on physiological data, the physiological risk level is calculated, which is a quantitative value of the degree of abnormality in the physiological state of the worker.

[0121] Based on the operational phase, the exposure risk level and physiological risk level are weighted and integrated to obtain a comprehensive risk score;

[0122] The overall risk score is compared with the preset risk status threshold to determine the risk status at the current moment.

[0123] Among them, the environmental risk level represents the quantitative value of the overall environmental hazard level of a confined space, usually expressed as a score from 0 to 100, with higher values ​​indicating a higher degree of environmental hazard; the exposure risk level refers to the quantitative value of the environmental hazard level of the worker's current location, reflecting the risk level of the worker's actual exposure to the hazardous environment; the physiological risk level represents the quantitative value of the degree of abnormality in the worker's physiological state, used to quantify the extent to which the worker's physical condition deviates from the normal range; the comprehensive risk score is the final risk assessment value obtained by weighted integration of multiple risk factors, used to comprehensively reflect the overall safety status of the current working environment; the preset risk state threshold represents the critical value used to distinguish different risk states, and its setting scheme is determined based on industry safety standards and historical accident data analysis, usually setting two thresholds: a warning threshold (e.g., 60 points) and a danger threshold (e.g., 80 points), used to map the comprehensive risk score to the three-level risk state.

[0124] After acquiring environmental monitoring data, trend characteristics, and the location and physiological data of workers, the server performs this step to comprehensively assess the safety status of the current working environment and determine the corresponding risk level. Specifically, the server first calculates the environmental risk level based on the environmental monitoring data and trend characteristics. The server compares the current values ​​of various environmental parameters (such as oxygen concentration and hazardous gas concentration) with their safety thresholds and calculates the normalized deviation. Considering the trend characteristics of environmental parameters, the server increases the risk weight of parameters with a positive and large rate of change (such as a rapid increase in hazardous gas concentration) and appropriately decreases the risk weight of parameters with a negative rate of change (such as a decrease in concentration). The server then performs a weighted average of the risk scores of each environmental parameter according to preset weights to obtain the overall environmental risk level, with a value ranging from 0 to 100.

[0125] The server calculates the exposure risk level based on the location data of the workers and the environmental risk level. First, the server determines the precise coordinates of the workers within a confined space based on the location data. Then, combining this with the location of environmental monitoring equipment, the server estimates the environmental parameter values ​​at the workers' location using a spatial interpolation algorithm (such as Kriging interpolation). Based on the estimated environmental parameter values ​​and the risk characteristics of the current work phase, the server calculates the exposure risk level of the workers' location, with a value ranging from 0 to 100.

[0126] The server calculates the physiological risk level based on physiological data. It compares the collected physiological data (such as heart rate, body temperature, and blood oxygen saturation) with preset normal ranges to calculate the degree of abnormality for each indicator. The server considers the correlation and overall performance between different physiological indicators; for example, an elevated heart rate accompanied by a decrease in blood oxygen may indicate a more serious health risk. Based on the degree of abnormality and correlation analysis of each physiological indicator, the server calculates the overall physiological risk level, with a value ranging from 0 to 100.

[0127] The server performs a weighted fusion of exposure risk level and physiological risk level based on the current work stage. The server loads the risk weight configuration for the current work stage from the configuration library; different work stages may assign different weights to exposure risk and physiological risk. For example, during the spraying stage, exposure risk may be given more weight (0.7), while the physiological risk weight may be relatively reduced (0.3); whereas during high-intensity physical labor stages, the physiological risk weight may be increased. Based on the loaded weight configuration, the server calculates the weighted fusion comprehensive risk score: Comprehensive Risk Score = Exposure Risk Level × Exposure Weight + Physiological Risk Level × Physiological Weight.

[0128] The server compares the calculated comprehensive risk score with the preset risk status threshold. When the comprehensive risk score is lower than the warning threshold (e.g., 60 points), the current risk status is determined to be safe; when the comprehensive risk score is greater than or equal to the warning threshold but lower than the danger threshold (e.g., 60-80 points), the current risk status is determined to be warning; when the comprehensive risk score is greater than or equal to the danger threshold (e.g., 80 points), the current risk status is determined to be dangerous.

[0129] S211. When the risk status is warning or danger, generate and send warning information to the preset receiving terminal.

[0130] This step specifically includes:

[0131] When the risk status is warning, the first-level warning response is triggered, which includes sending a prompt signal to the wearable devices worn by the workers.

[0132] When the risk status is dangerous, a Level 2 warning response is triggered. The Level 2 warning response includes sending an alarm signal to the wearable device and simultaneously activating the audible and visual alarm device at the entrance of the confined space.

[0133] The first-level early warning response refers to the lower-level emergency measures triggered by the early warning status, which mainly focus on reminders and warnings and do not interrupt the work process; the second-level early warning response refers to the higher-level emergency measures triggered by the dangerous status, which include stronger alarm signals and possible mandatory intervention measures.

[0134] After determining the current risk status, the server executes this step to trigger corresponding early warning responses based on different risk levels, promptly notifying relevant personnel and initiating necessary safety measures. Specifically, the server first checks the current risk status category (safe, warning, or hazard) determined in step S210. When the risk status is safe, the server does not trigger an early warning response and continues the normal monitoring process. When the risk status is warning, the server triggers a first-level early warning response. The server generates an early warning information packet containing structured information such as the warning time, the reason for the warning (e.g., abnormal oxygen concentration decline trend), the risk level, and recommended measures. The server sends the early warning information to the wearable device worn by the operator via a wireless communication network. Upon receiving the early warning information, the wearable device alerts the operator in an appropriate manner, such as by generating a slight vibration, emitting a low-volume but clear alert tone, or displaying the warning information on the screen. This alert intensity is sufficient to attract the operator's attention without causing panic or interfering with normal operations.

[0135] When the risk status is deemed hazardous, the server triggers a Level 2 warning response. The server generates a high-priority hazard alarm information package, containing detailed hazard source information, risk assessment results, and necessary emergency response guidance. The server sends alarm information in parallel through multiple communication channels to ensure reliable information transmission. The server sends an alarm signal to the wearable devices of the workers. Upon receiving the signal, the devices vibrate strongly, emit a high-decibel alarm, and display a prominent red warning message, forcibly attracting the workers' attention and prompting them to take immediate action (such as evacuating the hazardous area). Simultaneously, the server sends an emergency notification to the mobile terminals of safety management personnel, containing a detailed description of the hazard and the workers' location information. The server activates the audible and visual alarm device at the entrance to the confined space through a control interface, emitting sound and flashing alarms to alert external monitoring personnel and those in the surrounding area to the hazardous situation. Optionally, in some embodiments, the server may also automatically activate emergency ventilation equipment, cut off specific power supplies, or execute other preset emergency measures. The server continuously monitors changes in the hazardous status. When the risk decreases to a safe level, it sends an all-clear signal, stops the audible and visual alarms, and notifies relevant personnel that the danger has been eliminated.

[0136] In this embodiment, a confined space-based operation risk early warning method is adopted, including a complete technical solution that identifies operation stages based on statistical features of equipment operating status parameters, performs stage correction in conjunction with the operation area, adjusts monitoring strategies according to material risk levels, acquires personnel location and physiological data collected by wearable devices, calculates environmental monitoring data change trends, performs multi-dimensional comprehensive risk judgment, and implements graded early warning responses. Therefore, the system can achieve accurate identification of operation stages, optimized allocation of monitoring resources, dual monitoring of personnel and environment, and graded risk responses. This effectively solves the problems in related technologies, such as reliance on manual stage identification, lack of targeted monitoring strategies, single risk assessment dimensions, and "one-size-fits-all" early warning responses. As a result, it achieves comprehensiveness, accuracy, and intelligence in risk early warning, and improves the safety management efficiency and accident prevention capabilities of confined space operations.

[0137] The server of the risk warning system in the embodiments of this invention is described below from the perspective of hardware processing. Please refer to [link / reference needed]. Figure 3 This is a schematic diagram of the physical device structure of a server for a risk warning system in this application embodiment.

[0138] It should be noted that, Figure 3 The server structure of the risk warning system shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0139] like Figure 3As shown, the risk warning system's server includes a CPU 301, which can perform various appropriate actions and processes based on a program stored in the read-only memory ROM 302 or a program loaded from the storage section 308 into the random access memory RAM 303, such as executing the methods described in the above embodiments. The RAM 303 also stores various programs and data required for system operation. The CPU 301, ROM 302, and RAM 303 are interconnected via a bus 304. An I / O interface 305 is also connected to the bus 304.

[0140] The following components are connected to I / O interface 305: input section 306 including audio input devices, push-button switches, etc.; output section 307 including a liquid crystal display (LCD) and audio output devices, indicator lights, etc.; storage section 308 including a hard disk, etc.; and communication section 309 including a network interface card such as a LAN (Local Area Network) card, modem, etc. Communication section 309 performs communication processing via a network such as the Internet. Drive 310 is also connected to I / O interface 305 as needed. Removable media 311, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 310 as needed so that computer programs read from them can be installed into storage section 308 as needed.

[0141] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing computer programs for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 309, and / or installed from removable medium 311. When the computer program is executed by CPU 301, it performs the various functions defined in the present invention.

[0142] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0143] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. Each block in a flowchart or block diagram may represent a module, program segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those shown in the drawings.

[0144] Specifically, the server of the risk warning system in this embodiment includes a processor and a memory. The memory stores a computer program. When the computer program is executed by the processor, it implements the operation risk warning method based on limited space provided in the above embodiment.

[0145] In another aspect, the present invention also provides a computer-readable storage medium, which may be included in the server of the risk warning system described in the above embodiments; or it may exist independently and not assembled into the server of the risk warning system. The storage medium carries one or more computer programs, which, when executed by the processor of a server of the risk warning system, cause the server of the risk warning system to implement the space-constrained operational risk warning method provided in the above embodiments.

[0146] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application 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 scope of the technical solutions of the embodiments of this application.

[0147] As used in the above embodiments, depending on the context, the term "when..." can be interpreted as meaning "if...", "after...", "in response to determining...", or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if (the stated condition or event) is interpreted as meaning "if determining...", "in response to determining...", "when (the stated condition or event) is detected", or "in response to detecting (the stated condition or event)".

[0148] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A method for early warning of operational risks based on confined space, applied to the server of a risk warning system, wherein the risk warning system includes at least a server and environmental monitoring equipment, characterized in that, include: The operating status parameters of the operating equipment are obtained, and the current operation stage of the operation process is identified based on the statistical characteristics of the operating status parameters. The operation stage is a process unit with different inherent risk attributes that is pre-divided according to the operation process. Invoke a preset monitoring strategy corresponding to the operation phase, the monitoring strategy including at least the types of environmental monitoring data to be collected and the data collection frequency under the operation phase; According to the monitoring strategy, environmental monitoring data collected by environmental monitoring equipment at the current moment is obtained, and the environmental monitoring data includes at least the concentration of harmful gases and the oxygen concentration; Extract historical environmental monitoring data within a preset time period and environmental monitoring data at the current moment, and calculate the trend characteristics of the environmental monitoring data. Based on the environmental monitoring data, the changing trend characteristics, and the operation stage, a comprehensive risk assessment is performed to obtain the current risk status. The risk status is the result of determining the safety level of the current operation environment, including safe, warning, and dangerous. When the risk status is warning or danger, a warning message is generated and sent to a preset receiving terminal.

2. The method according to claim 1, characterized in that, Invoking a preset monitoring strategy corresponding to the operation stage, specifically including: The preset initial monitoring strategy is obtained according to the operation stage; Based on the identification information of the work materials, the chemical property parameters of the work materials are extracted from the preset material safety database. The chemical property parameters include at least the percentage of volatile organic compounds, flash point temperature value, and toxicity level identification. Based on the chemical property parameters and the amount of the work material used, the risk level of the work material is determined, and the risk level includes low risk, medium risk, and high risk. Based on the risk level of the materials being worked on, the initial monitoring strategy is adjusted to obtain the adjusted monitoring strategy.

3. The method according to claim 2, characterized in that, The risk warning system also includes wearable devices; After adjusting the initial monitoring strategy according to the risk level of the work materials to obtain the adjusted monitoring strategy, the method further includes: According to the monitoring strategy, the current location data and physiological data of the workers are acquired by the wearable device.

4. The method according to claim 3, characterized in that, Based on the environmental monitoring data, the changing trend characteristics, and the operational stage, a comprehensive risk assessment is performed to determine the current risk status, specifically including: Based on the environmental monitoring data and the changing trend characteristics, the environmental risk level is calculated, whereby the environmental risk level is a quantitative value of the overall environmental hazard level of the confined space. Based on the location data of the workers and the environmental risk level, the exposure risk level is calculated, whereby the exposure risk level is a quantitative value of the environmental hazard level of the current location of the workers. Based on the physiological data, a physiological risk level is calculated, whereby the physiological risk level is a quantitative value representing the degree of abnormality in the worker's physiological state. Based on the operational phase, the exposure risk level and the physiological risk level are weighted and fused to obtain a comprehensive risk score; The comprehensive risk score is compared with a preset risk status threshold to determine the risk status at the current moment.

5. The method according to claim 1, characterized in that, The process of acquiring the operating status parameters of the operating equipment and identifying the current operating stage of the work process based on the statistical characteristics of the operating status parameters specifically includes: Real-time acquisition of operating status parameters of the working equipment, wherein the operating status parameters include at least the operating current or power consumption of the working equipment; Calculate the statistical characteristics of the operating state parameters, wherein the statistical characteristics include at least the mean and variance; By comparing the statistical characteristics of the operating status parameters of adjacent time windows through a preset time window, when the change in the statistical characteristics is greater than a preset change threshold and the duration is greater than a preset duration threshold, it is determined that the operation stage has switched. After determining that a job stage has switched, the statistical features within the current time window are input into a preset job stage identification model to obtain the job stage in which the current job process is located.

6. The method according to claim 5, characterized in that, After inputting the statistical features within the current time window into a preset job stage identification model to obtain the job stage of the current job process, the method further includes: Acquire the location data of the workers, and identify the work area of ​​the workers based on the location data; Obtain a preset sequence of work stages for the work area, wherein the sequence of work stages is a set of work stages arranged in sequence for a specific work area; If the work stage output by the work stage identification model does not belong to the work stage sequence, then calculate the feature similarity between the statistical features within the current time window and the preset operating state parameter feature range of each work stage in the work stage sequence. The stage with the highest feature similarity is determined as the stage in the current modified workflow.

7. The method according to any one of claims 1-6, characterized in that, Based on the aforementioned risk status, early warning information is sent to operators and safety management personnel, specifically including: When the risk status is a warning, a first-level warning response is triggered, which includes sending a prompt signal to the wearable device worn by the worker. When the risk status is dangerous, a second-level warning response is triggered. The second-level warning response includes sending an alarm signal to the wearable device and simultaneously activating the audible and visual alarm device at the entrance of the confined space.

8. A server for a risk early warning system, characterized in that, The server of the risk warning system includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to cause the server of the risk warning system to perform the method as described in any one of claims 1-7.

9. A computer-readable storage medium comprising instructions, characterized in that, When the instruction is executed on the server of the risk warning system, the server of the risk warning system performs the method as described in any one of claims 1-7.

10. A computer program product, characterized in that, When the computer program product is run on the server of the risk warning system, the server of the risk warning system performs the method as described in any one of claims 1-7.

Images