Oxygen respirator monitoring alarm method and system

Abstract

The present application provides an oxygen respirator monitoring and alarm method and system, relating to the technical field of oxygen respirator monitoring and alarm technology. The key technical aspects of the method are: obtaining a firefighter's response to a preset status confirmation request and monitoring parameters related to the firefighter's physiological discomfort; determining whether there is a preset correlation between the obtained response result and abnormal fluctuations in the monitoring parameters, indicating that the firefighter failed to effectively respond to the status confirmation request due to physiological discomfort; and when the determined correlation meets the preset correlation analysis rules, determining that the firefighter is at risk of early disability due to non-hypoxia reasons, where oxygen supply is sufficient but mobility may be impaired, and issuing a warning or alarm. The oxygen respirator monitoring and alarm method and system provided by the present application have the advantages of improving alarm accuracy and compensating for the shortcomings of traditional oxygen respirator alarm systems that focus primarily on oxygen supply safety.

Description

Technical Field

[0001] The present application relates to the technical field of oxygen respirator monitoring and alarm technology, and in particular to an oxygen respirator monitoring and alarm method and system. Background Art

[0002] Firefighters face extremely high risks during firefighting and rescue operations, particularly in environments like underground commercial complexes with low visibility, complex internal structures, and narrow passageways. To ensure firefighter safety, an integrated firefighter safety assurance system is typically deployed. This system integrates multiple monitoring and communication modules, including firefighting oxygen respirators for real-time monitoring of cylinder pressure, valve status, and estimated remaining operating time; firefighting uniforms with integrated sensors for ambient temperature and humidity, multiple gas concentrations, and physiological parameters such as heart rate and skin temperature; firefighting helmets with integrated communications, information display, and lighting; and modules that provide indoor positioning information. The data collected by these modules is aggregated into a personal processing unit worn by the firefighter, which communicates with the command center at the rear.

[0003] However, in the actual rescue process, firefighters may encounter a variety of complex situations, which may pose potential threats to their safety, and the existing monitoring and alarm systems may not be able to identify these risks in a timely and effective manner. For example, when passing through a narrow passage, firefighters may inhale a small amount of irritating gas that has not been completely removed by the filtration system, causing severe coughing and short-term dizziness, resulting in body shaking and slow movement. In this case, even if key indicators such as the cylinder pressure of the oxygen respirator are still within a safe range and the hypoxia alarm has not been triggered, the firefighter's physiological state and ability to move may be affected. At the same time, the posture sensor on the fire suit may only record short-term irregular activities, which does not reach the preset alarm conditions that trigger a distress state (such as falling or being still for a long time).

[0004] Existing integrated safety assurance systems typically include a mechanism for proactively confirming firefighter status, sending status confirmation requests to firefighters via voice assistants or face mask displays. However, in situations where physical discomfort (such as dizziness or coughing) restricts movement, firefighters may not be able to effectively respond to status confirmation requests within the preset response window, resulting in a "status confirmation timeout."

[0005] Furthermore, although the oxygen respirator cylinder pressure remains normal, the firefighter's coughing and discomfort may cause brief, irregular fluctuations in breathing rate, resulting in a slightly higher-than-normal instantaneous oxygen consumption rate. However, this fluctuation is not large enough to cause the estimated remaining working time to drop rapidly to the alarm threshold. At the same time, the gas sensor on the firefighter's suit may detect a slight upward trend in the concentration of a certain irritant gas in the environment, but it also does not reach the concentration threshold that would trigger an independent gas alarm.

[0006] Existing monitoring and alarm methods for firefighting oxygen respirators primarily rely on single or independent thresholds, such as oxygen pressure. This traditional alarm logic exhibits significant flaws in situations where multiple indicators (such as status confirmation response, respiratory parameter fluctuations, subtle changes in ambient gas concentrations, and irregular posture movements) exhibit abnormal fluctuations, and the combination of these abnormalities could indicate that a firefighter is unable to effectively respond to status confirmation requests due to physiological discomfort. Relying solely on the oxygen respirator's own cylinder pressure threshold alarm would not trigger any warning in this situation, as oxygen reserves are still sufficient. If a "status confirmation timeout" event is simply equated with a most urgent distress signal and immediately notified to the command center and teammates, false alarms could occur when a firefighter is only temporarily unable to respond or communications are temporarily disrupted, disrupting rescue operations. However, the lack of an effective identification and early warning mechanism for this type of firefighter's potentially incapacitating condition, which is caused by multiple physiological indicator abnormalities and environmental factors that do not meet clear alarm criteria, could delay early attention and necessary assistance to firefighters. If a firefighter's condition deteriorates further, such as from dizziness, a fall, or loss of consciousness, even the normal oxygen supply from the oxygen respirator cannot guarantee their safety. A clear alarm may not be triggered until oxygen depletion or a more serious secondary incident occurs, at which point the optimal opportunity for rescue may have been missed. In particular, the alarm system of an oxygen respirator, whose core goal is to ensure oxygen supply safety, does not directly stem from the oxygen supply itself, but rather from the abnormal state of its user. This requires that oxygen respirator monitoring and alarm methods be integrated with a wider range of firefighter status monitoring information. These risks, which do not directly affect the oxygen supply but can lead to oxygen safety issues (due to user disability and inability to effectively utilize oxygen), can be identified early in the development of an incident, and appropriate warnings or alarms can be issued based on the level of risk.

[0007] In view of the above problems, the existing technology is in urgent need of improvement. Summary of the Invention

[0008] The purpose of this application is to provide an oxygen respirator monitoring alarm method and system, which has the advantages of improving the accuracy of the alarm and making up for the shortcomings of traditional oxygen respirator alarm systems that mainly focus on oxygen supply safety.

[0009] In the first aspect, the present application provides an oxygen respirator monitoring and alarm method, the technical solution is as follows:

[0010] include:

[0011] Obtain the firefighter's response to the preset status confirmation request;

[0012] obtaining monitoring parameters from a plurality of different sources related to the physiological discomfort state of the firefighter;

[0013] determining, based on a preset correlation analysis rule, whether there is a preset correlation between the acquired response result and the abnormal fluctuations of the monitoring parameters acquired from the multiple different sources, indicating that the firefighter failed to effectively respond to the status confirmation request due to the physiological discomfort;

[0014] When the judged correlation meets the preset correlation analysis rules, it is determined that the firefighter is at risk of early disability due to reasons other than hypoxia, where the oxygen supply is still sufficient but the mobility may be impaired, and based on the determined early disability risk, a warning or alarm for the early disability risk is triggered.

[0015] Furthermore, in the present application, the step of determining, based on a preset correlation analysis rule, whether there is a preset correlation between the acquired response result and the abnormal fluctuations of the monitoring parameters acquired from the multiple different sources, indicating that the firefighter failed to effectively respond to the status confirmation request due to the physiological discomfort, includes:

[0016] Acquiring real-time physiological data indicating current individual physiological characteristics of the firefighter and real-time environmental parameters indicating current environmental conditions at the rescue site;

[0017] Determining one or more rule adjustment instructions for the preset association analysis rule based on a difference between the real-time physiological data and the baseline physiological data of the firefighter, or based on a change characteristic of the real-time environmental parameter;

[0018] Applying the one or more rule adjustment instructions to adjust at least one parameter or logical condition in the preset association analysis rule to generate an adjusted association analysis rule;

[0019] Using the adjusted association analysis rules, in combination with the acquired response result and the abnormal fluctuations of the monitoring parameters acquired from the multiple different sources, it is determined whether there is a preset correlation between the response result and the abnormal fluctuations of the monitoring parameters from the multiple different sources, indicating that the firefighter failed to effectively respond to the status confirmation request due to the physiological discomfort.

[0020] Furthermore, in the present application, the step of determining one or more rule adjustment instructions for the preset association analysis rule based on the difference between the real-time physiological data and the baseline physiological data of the firefighter, or based on the change characteristics of the real-time environmental parameters, includes:

[0021] obtaining a first set of candidate rule adjustment parameters determined by a difference between the real-time physiological data and a baseline physiological data of the firefighter;

[0022] Obtaining a second set of candidate rule adjustment parameters determined by the change characteristics of the real-time environmental parameters;

[0023] determining, according to a preset conflict determination rule, whether there is an indication conflict between the first set of candidate rule adjustment parameters and the second set of candidate rule adjustment parameters;

[0024] When it is determined that there is an indication conflict, determining the one or more rule adjustment instructions for the preset association analysis rule in combination with the first set of candidate rule adjustment parameters and the second set of candidate rule adjustment parameters according to a preset conflict handling mechanism;

[0025] When it is determined that there is no indication conflict, the first group of candidate rule adjustment parameters determined based on the difference between the real-time physiological data and the baseline physiological data of the firefighter, or the second group of candidate rule adjustment parameters determined based on the changing characteristics of the real-time environmental parameters, or the first group of candidate rule adjustment parameters and the second group of candidate rule adjustment parameters are combined to determine the one or more rule adjustment instructions for the preset association analysis rules.

[0026] Further, in the present application, when the preset conflict handling mechanism includes at least two conflict handling strategies, and the current conflict situation formed by the first set of candidate rule adjustment parameters and the second set of candidate rule adjustment parameters simultaneously satisfies the conditions for activating the at least two conflict handling strategies, or the conflict situation does not fully satisfy the activation conditions of any preset conflict handling strategy, the step of determining the one or more rule adjustment instructions based on the preset conflict handling mechanism in combination with the first set of candidate rule adjustment parameters and the second set of candidate rule adjustment parameters includes:

[0027] obtaining quantitative characteristics of the conflict situation;

[0028] Obtaining a time series variation trend of the first set of candidate rule adjustment parameters and a time series variation trend of the second set of candidate rule adjustment parameters;

[0029] Obtain feedback data on the execution effects of historical rule adjustment instructions related to historical conflict situations, historical parameter trends, and historically adopted conflict resolution strategies;

[0030] selecting, based on a preset set of meta-rules and in combination with the obtained quantitative characteristics of the conflict situation, the obtained time series change trend, and the obtained feedback data, a conflict handling strategy from at least two conflict handling strategies included in the preset conflict handling mechanism, or combining the at least two conflict handling strategies to generate a current conflict handling solution;

[0031] The one or more rule adjustment instructions are determined by applying the current conflict handling solution and combining the first set of candidate rule adjustment parameters and the second set of candidate rule adjustment parameters.

[0032] Furthermore, in the present application, when the meta-rule indicates that the at least two conflict handling strategies need to be combined to generate the current conflict handling solution, the step of combining the at least two conflict handling strategies based on the preset meta-rule set and in combination with the obtained quantitative characteristics of the conflict situation, the obtained time series change trend, and the obtained feedback data to generate the current conflict handling solution includes:

[0033] Identifying a combination parameter of the at least two conflict handling strategies, the number of which is greater than one and which can be adjusted, the combination parameter being used to characterize a combination mode of the at least two conflict handling strategies, the combination mode including at least one of weight distribution, execution order, and fusion logic;

[0034] Based on the indication of the meta-rule, or with reference to the valid combination parameters recorded in the feedback data that are similar to the quantitative characteristics of the conflict situation at that moment and the time series change trend, setting a set of combination parameter values ​​as a starting point for the obtained combination parameters that are greater than one and can be adjusted;

[0035] Generating a set of combination parameter values ​​greater than one and different for selection by adjusting the value of at least one parameter among the combination parameters greater than one and adjustable;

[0036] Using a preset combination effect evaluation logic, for each generated set of selectable combination parameter values, and in combination with the quantitative characteristics of the conflict situation at that moment, the time series change trend, and the feedback data, a numerical evaluation index of the corresponding expected effect is calculated. The combination effect evaluation logic is used to output a numerical evaluation index of the expected effect of combining the selectable combination parameter values ​​based on the input quantitative characteristics of the conflict situation at that moment, the time series change trend, the feedback data, and the set of selectable combination parameter values;

[0037] From the calculated numerical evaluation indicators of each set of selectable combination parameter values, select a set of combination parameter values ​​that makes the numerical evaluation indicators meet a pre-set optimization goal as the final determined combination parameter;

[0038] The at least two conflict handling strategies are combined by applying the selected finally determined combination parameter to generate the current conflict handling solution.

[0039] Furthermore, in the present application, the monitoring parameters from the multiple different sources include at least two of: environmental gas concentration parameters, firefighter breathing parameters, firefighter posture parameters, firefighter heart rate parameters, and firefighter body surface temperature parameters.

[0040] Furthermore, in the present application, the real-time environmental parameter includes at least one of ambient temperature and gas concentration.

[0041] Furthermore, in the present application, the meta-rule set is a set of rules that defines the logic for selecting one of the at least two conflict handling strategies included in the preset conflict handling mechanism, or determining the combination of the at least two conflict handling strategies based on the quantitative characteristics of the conflict situation, the time series change trend, and the feedback data.

[0042] Furthermore, in this application, the preset combination effect evaluation logic includes at least:

[0043] Mapping the input quantitative characteristics of the conflict situation, the time series change trend, the key elements in the feedback data, and the currently selected combined parameter values ​​to be evaluated into a set of predefined scoring items;

[0044] Set corresponding weight coefficients and scoring rules for each scoring item;

[0045] The numerical evaluation index is obtained by calculating the score of each scoring item and performing weighted summation or logical reasoning.

[0046] In a second aspect, the present application also proposes an oxygen respirator monitoring and alarm system, which includes:

[0047] A response result acquisition module is used to obtain the firefighter's response result to the preset status confirmation request;

[0048] a monitoring parameter acquisition module, configured to acquire monitoring parameters from a plurality of different sources related to the firefighter's physiological discomfort, wherein each of the monitoring parameters from the plurality of different sources does not reach a preset independent alarm threshold, and the monitoring parameters from the plurality of different sources exhibit abnormal fluctuations indicative of the physiological discomfort;

[0049] a correlation determination module, configured to determine, based on preset correlation analysis rules, whether there is a preset correlation between the response result obtained by the response result obtaining module and abnormal fluctuations of the monitoring parameters from multiple different sources obtained by the monitoring parameter obtaining module, indicating that the firefighter failed to effectively respond to the status confirmation request due to the physiological discomfort;

[0050] The risk assessment and early warning module is used to determine that the firefighter is at risk of early disability due to reasons other than hypoxia, where the oxygen supply is sufficient but the ability to move may be impaired, when the correlation determined by the correlation assessment module satisfies the preset correlation analysis rules, and trigger a warning or alarm for the early disability risk based on the determined early disability risk.

[0051] As can be seen from the above, the oxygen respirator monitoring alarm method and system provided by the present application obtains the response results of the firefighters to the status confirmation request and the abnormal fluctuations of multiple sub-threshold monitoring parameters, and judges whether there is a correlation of failure to respond effectively due to physiological discomfort based on correlation analysis, thereby identifying the early disability risk caused by non-hypoxia reasons and triggering an early warning or alarm. It has the ability to effectively identify the early disability risk of firefighters caused by multiple abnormal combinations that are difficult to detect with existing technologies, and provide early warnings or alarms at the early stage when the firefighters have sufficient oxygen supply but their mobility may be impaired, thereby improving the accuracy of the alarm and making up for the shortcomings of traditional oxygen respirator alarm systems that mainly focus on oxygen supply safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 A flow chart of an oxygen respirator monitoring and alarm method provided in this application.

[0053] Figure 2 This is a structural diagram of an oxygen respirator monitoring and alarm system provided in this application.

[0054] In the figure: 1. First acquisition module; 2. Second acquisition module; 3. Judgment module; 4. Early warning module. DETAILED DESCRIPTION

[0055] The technical solutions in this application will be clearly and completely described below in conjunction with the drawings in this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all of the embodiments. The components of the present application generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application for which protection is claimed, but merely represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of this application.

[0056] It should be noted that similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and should not be understood as indicating or implying relative importance.

[0057] In traditional existing methods, when firefighters have sufficient oxygen supply but fail to effectively respond to status confirmation requests due to physiological discomfort, and when multiple threshold monitoring parameters fluctuate abnormally, it is impossible to effectively identify early disability risks and issue appropriate warnings or alarms. This limitation stems from the fact that existing systems mainly rely on single or independent monitoring parameter thresholds for alarm judgment. When multiple related parameters do not reach their respective independent alarm thresholds, even if the combined abnormal fluctuations of these parameters and the user's non-responsive behavior jointly indicate potential physiological discomfort and impaired mobility risks, the system cannot identify them as events that require attention. As a result, the system may not be able to perceive the early deterioration of the user's status in a timely manner, affecting the accurate assessment of the user's safety status and the classification of risk levels, thereby delaying necessary intervention measures.

[0058] For example, during a rescue operation in a complex environment, a firefighter wears a personal safety system that integrates multiple monitoring functions. This system includes an oxygen respirator monitoring module, a physiological status monitoring module, and an environmental monitoring module. The system sends a status confirmation request to the user according to a preset strategy. However, due to temporary physiological discomfort caused by inhaling a trace amount of irritant, the user fails to respond promptly through the system interface or voice command, resulting in a response timeout event. Simultaneously, the oxygen respirator monitors fluctuations in the user's breathing rate and instantaneous oxygen consumption rate, but these fluctuations do not reach the thresholds required to trigger oxygen-related alarms. The physiological status monitoring module detects slight irregularities in the user's posture, but these do not meet the preset distress posture alarm conditions, such as a fall or prolonged inactivity. The environmental monitoring module detects a slight increase in the concentration of a specific gas in the environment, but this concentration is far below the threshold required to trigger a gas alarm. In this case, despite the user's non-response and the abnormal fluctuations in multiple monitoring parameters, existing systems lack the ability to conduct comprehensive correlation analysis on these discrete signals, making it impossible to determine whether there is a correlation between these events that indicates the user's mobility may be impaired due to physiological discomfort. As a result, the system will not trigger any alarms or will only trigger low-level, unclear prompts, which fail to effectively reflect the specific early risks faced by users.

[0059] If the above issues are not addressed, the system will not be able to issue effective warnings when the user's oxygen supply is still sufficient but they are already showing early signs of physiological discomfort and impaired mobility. This may cause the user's condition to continue to deteriorate unnoticed until traditional alarms (such as oxygen depletion alarms or distress posture alarms) that are usually associated with more serious situations are triggered. By then, the user may have completely lost their ability to move or consciousness, missing the optimal opportunity for early intervention and rescue. This delayed risk identification and inappropriate alarm mechanism directly affects the user's safety level in complex and dangerous environments and increases the risk of rescue operations.

[0060] In this regard, refer to Figure 1, this application proposes an oxygen respirator monitoring and alarm method, comprising:

[0061] S110, obtaining a firefighter's response to the preset status confirmation request;

[0062] S120, obtaining monitoring parameters from multiple different sources related to the firefighter's physiological discomfort state;

[0063] S130. Determine, based on a preset correlation analysis rule, whether there is a preset correlation between the acquired response result and abnormal fluctuations of the monitoring parameters acquired from multiple different sources, indicating that the firefighter failed to effectively respond to the status confirmation request due to physiological discomfort;

[0064] S140. When the judged correlation meets the preset correlation analysis rules, it is determined that the firefighter is at risk of early disability due to reasons other than hypoxia, where the oxygen supply is sufficient but the mobility may be impaired, and based on the determined early disability risk, a warning or alarm for the early disability risk is triggered.

[0065] Obtaining the firefighter's response to the preset status confirmation request refers to the firefighter's feedback received after the system sends a signal to inquire about the firefighter's current status. This can be achieved through voice recognition, button confirmation, mask display interaction, etc.

[0066] Acquiring monitoring parameters related to a firefighter's physiological discomfort from multiple sources involves collecting data from various sensors on the firefighter's equipment. This data reflects the firefighter's physiological condition or the environment they are in. Specifically, this data can be acquired using devices such as respiratory rate sensors, gas concentration sensors, posture sensors, and heart rate sensors.

[0067] Determining whether there is a predetermined correlation between the acquired response result and abnormal fluctuations in monitoring parameters acquired from multiple different sources, based on preset correlation analysis rules, indicating that the firefighter failed to effectively respond to the status confirmation request due to physiological discomfort, involves analyzing, using pre-set logic or models, whether there is a specific correlation pattern between the response result and sub-threshold abnormal fluctuations in the multi-source monitoring parameters. This pattern indicates that the firefighter's failure to respond was due to physiological discomfort.

[0068] When the correlation judgment satisfies the pre-set correlation analysis rules, the system determines that the firefighter faces a risk of early disability due to non-hypoxia reasons, where oxygen supply is sufficient but mobility may be impaired. This means that if the previous correlation judgment result is true, the system concludes that the firefighter faces a specific danger. This danger is not due to insufficient oxygen, but rather to other factors that may cause a decrease in mobility despite sufficient oxygen supply. This can be achieved through logical judgment, state machine transitions, and other methods.

[0069] Triggering an early-stage disability risk warning or alarm based on a previously identified early-stage disability risk means the system issues a corresponding prompt based on the specific early-stage risk identified. This prompt is tailored to the nature of the specific risk and can be implemented through audible alarms, visual prompts, or by sending a message to a command center.

[0070] The core innovation of this application lies in the comprehensive analysis of the correlation between the response results of firefighters to status confirmation requests and abnormal fluctuations in monitoring parameters from multiple different sources, thereby identifying the risk of firefighters facing early disability due to reasons other than hypoxia when there is sufficient oxygen supply, and triggering early warnings or alarms based on this risk, which is different from existing methods that only rely on single or independent thresholds for alarms.

[0071] The working process and principle of this application is that this method aims to identify the risk of early disability when firefighters fail to effectively respond to system status confirmation requests due to physiological discomfort when there is sufficient oxygen supply, and when multiple monitoring parameters that do not reach independent alarm thresholds show abnormal fluctuations.

[0072] Specifically, the system first obtains the firefighter's response to a preset status confirmation request. This request is used to proactively detect the firefighter's current status and interactive capabilities. The response can be a proactive confirmation signal from the firefighter or a failure to receive a valid response within a preset time window. Failure to respond indicates that the firefighter may be experiencing some abnormality.

[0073] Simultaneously, monitoring parameters related to the firefighter's physiological discomfort are collected from multiple different sources. These parameters can include respiratory rate, oxygen consumption rate, heart rate, body surface temperature, posture changes, and ambient gas concentrations. These parameters are collected by different sensors or modules integrated into the firefighter's equipment. The purpose of acquiring parameters from multiple different sources is to provide more comprehensive information, reflecting subtle changes in the firefighter's physiological state and the environment they are in. The fluctuations in these parameters individually may not be sufficient to trigger independent alarm thresholds, but their combined changes may indicate potential problems.

[0074] Furthermore, based on pre-determined association analysis rules, it is determined whether there is a specific correlation between the obtained response results and the abnormal fluctuations of monitoring parameters obtained from multiple different sources. This correlation is defined as indicating that the firefighter failed to effectively respond to the status confirmation request due to physiological discomfort. The association analysis rules are used to identify specific patterns. For example, when the firefighter fails to respond to the request, if it is accompanied by fluctuations in breathing rate and a slight increase in ambient gas concentration, there may be a correlation indicating physiological discomfort. Through this correlation judgment, the method is able to distinguish between real risks caused by physiological discomfort and non-response caused by other reasons (such as temporary communication interruption).

[0075] Therefore, when the determined correlation satisfies pre-determined correlation analysis rules, the method determines that the firefighter is at risk of early-stage disability due to non-hypoxia factors, where oxygen supply is sufficient but mobility may be impaired. This determination clarifies the nature (non-oxygen causes) and stage (early stage, potential impairment of mobility), unlike traditional oxygen depletion alarms. This early risk identification enables the system to issue warnings before the problem escalates.

[0076] Finally, based on the identified early incapacitation risk, the method triggers a warning or alarm specific to that risk. This warning or alarm can be graded based on the severity of the risk. For example, a local alert can be sent to the firefighter, or a specific type of alarm signal can be sent to the command center, indicating that the risk is "early incapacitation risk due to reasons other than hypoxia," rather than the traditional oxygen depletion or distress posture alarms. This allows the command center to understand the specific nature of the risk and implement targeted support measures.

[0077] As a preferred embodiment, the solution of this application is specifically implemented as follows:

[0078] A firefighter's personal safety system integrates an oxygen respirator monitoring module, a physiological status monitoring module, and an environmental monitoring module. Periodically or upon detecting a specific event (such as entering a new area), the system issues a "Please confirm status" request to the firefighter via a voice prompt within the helmet. The system sets a response time window, such as a few seconds. The response result acquisition module records whether the firefighter confirms the situation within this time window, either through voice commands or the buttons on the helmet. If no valid confirmation is received, it is recorded as a "response timeout."

[0079] Meanwhile, the monitoring parameter acquisition module continuously collects data from various sources. The oxygen respirator monitoring module provides respiratory rate and instantaneous oxygen consumption rate data. The physiological status monitoring module provides heart rate and posture sensor data. The environmental monitoring module provides ambient temperature, humidity, and various gas concentration data. Real-time values ​​of these parameters are acquired.

[0080] The correlation determination module receives a response result (e.g., "response timeout") and real-time data on multiple monitoring parameters provided by the monitoring parameter acquisition module. This module internally stores a set of pre-determined correlation analysis rules. For example, a rule might state: "If the response timeout occurs, and the respiratory rate fluctuates by more than a preset threshold A within the past 10 seconds, and the concentration of ambient gas X increases by more than a preset threshold B within the past 30 seconds, then a correlation indicating physiological discomfort is determined to exist." The correlation determination module then matches the acquired response result and monitoring parameters with the set of rules.

[0081] When the correlation judgment module determines that the current situation meets one or more correlation analysis rules, the risk assessment and warning module is activated. Based on this assessment, this module determines that the firefighter is at risk of early-stage disability due to non-hypoxia reasons, where oxygen supply is sufficient but mobility may be impaired. For example, the system determines that the current situation meets the risk model of "physiological discomfort leading to impaired mobility."

[0082] Based on this identified early disability risk, the risk assessment and warning module triggers a corresponding warning or alarm. For example, the system can send a text prompt to the display screen inside the firefighter's helmet, "Caution: Possible risk of early physiological discomfort," and simultaneously send a specific type of alarm signal to the rear command center, indicating that the risk is "risk of early disability due to reasons other than hypoxia," rather than the traditional "oxygen depletion alarm" or "distress posture alarm." This allows the command center to understand the specific nature of the risk and take targeted support measures.

[0083] Through the above scheme, this application solves the problem that existing methods cannot effectively identify early disability risks and issue appropriate warnings or alarms when firefighters have sufficient oxygen supply but are unable to effectively respond to status confirmation requests due to physiological discomfort, and when multiple monitoring parameters fluctuate abnormally. This method can identify early disability risks caused by potential physiological discomfort and non-oxygen reasons by comprehensively analyzing the correlation between the firefighters' response status and abnormal fluctuations in various physiological and environmental monitoring data. As a result, the system can detect potential dangers in advance when the problem has not yet seriously deteriorated and the oxygen supply is still sufficient, and trigger warnings or alarms for specific risk properties, thereby achieving more timely and appropriate intervention and buying time to ensure the safety of firefighters.

[0084] Specifically, some of the aforementioned solutions in this application propose using preset correlation analysis rules to determine whether there is a correlation between a firefighter's failure to effectively respond to a status confirmation request and abnormal fluctuations in monitoring parameters related to physiological discomfort, thereby identifying firefighters' early disability risk. However, because firefighters' individual physiological characteristics and the environmental conditions at the rescue site are dynamically changing, a fixed, preset correlation analysis rule may not be able to fully adapt to these changes, resulting in inaccurate correlation judgments, thereby affecting the reliability of early disability risk identification and potentially leading to false positives or missed positives.

[0085] In this regard, the present application further proposes a step of determining, based on a preset correlation analysis rule, whether there is a preset correlation between the acquired response result and abnormal fluctuations in monitoring parameters acquired from multiple different sources, indicating that the firefighter failed to effectively respond to the status confirmation request due to physiological discomfort, including:

[0086] Acquiring real-time physiological data indicating the current individual physiological characteristics of firefighters, and real-time environmental parameters indicating the current environmental status of the rescue scene;

[0087] Determining one or more rule adjustment instructions for preset association analysis rules based on differences between the real-time physiological data and the firefighter's baseline physiological data, or based on change characteristics of real-time environmental parameters;

[0088] Applying one or more rule adjustment instructions to adjust at least one parameter or logical condition in a preset association analysis rule to generate an adjusted association analysis rule;

[0089] Using the adjusted association analysis rules, the obtained response results are combined with abnormal fluctuations of monitoring parameters obtained from multiple different sources to determine whether there is a preset correlation between the response results and the abnormal fluctuations of monitoring parameters from multiple different sources, indicating that the firefighter failed to effectively respond to the status confirmation request due to physiological discomfort.

[0090] This method aims to address the problem of inaccurate judgments when using fixed, pre-set association analysis rules, which can fail to account for individual differences among firefighters and real-time environmental changes. Its core approach is to incorporate real-time dynamic information and adaptively adjust the association analysis rules based on this information, thereby improving the accuracy and reliability of early disability risk assessments.

[0091] Specifically, the solution first acquires real-time physiological data indicating the firefighter's current individual physiological characteristics, as well as real-time environmental parameters indicating the current state of the rescue scene. Real-time physiological data, such as heart rate, respiratory rate, and body temperature, can reflect the firefighter's current physical condition. This data is affected by various factors, including individual differences, fatigue level, and health status. Real-time environmental parameters, such as temperature, humidity, and specific gas concentrations, can reflect the firefighter's external environmental conditions. These factors can also affect the firefighter's physiological state and monitored parameters. Acquiring this real-time, dynamic data provides foundational information for subsequent rule adjustments. Real-time physiological data can be acquired using physiological sensors worn by firefighters, such as heart rate monitors, thermometers, and respiratory sensors. Real-time environmental parameters can be acquired using environmental sensors integrated into firefighter uniforms or helmets, such as gas sensors and temperature and humidity sensors. This data is transmitted to a personal processing unit or a command center for processing.

[0092] Next, based on the differences between the real-time physiological data and the firefighter's baseline physiological data, or based on the changing characteristics of real-time environmental parameters, one or more rule adjustment instructions for the preset association analysis rules are determined. Comparing real-time physiological data with the firefighter's baseline physiological data can quantify the degree of deviation of the current physiological state from their normal or baseline state. This difference may indicate the firefighter's current physiological load or potential abnormality. For example, if the real-time heart rate is significantly higher than the firefighter's baseline resting heart rate or average heart rate during the mission, it may indicate increased physiological load. Simultaneously, analyzing the time series of real-time environmental parameters can assess the potential impact of environmental factors on the firefighter's physiological state and parameter performance. For example, a rapid increase in ambient temperature can cause an increase in body temperature and heart rate. Based on these physiological differences or environmental change characteristics, the system can intelligently generate instructions for adjusting the preset association analysis rules. For example, in high temperature and high humidity environments, the threshold for determining abnormal fluctuations in physiological parameters may need to be adjusted. Or, when detecting trace amounts of irritating gases, the correlation between respiratory system-related indicators and the unresponsive state may need to be more sensitively monitored. Determining rule adjustment instructions is a key step in achieving dynamic rule adaptability. The rule adjustment instruction can be a numerical value indicating that a threshold value needs to be increased or decreased by a certain amount, or it can be a logical flag indicating that a logical condition needs to be activated or disabled.

[0093] Then, one or more rule adjustment instructions are applied to adjust at least one parameter or logical condition in the preset association analysis rule, generating an adjusted association analysis rule. This step specifically implements the adjustment instructions determined in the previous step onto the preset association analysis rule. Adjustment can involve modifying the numerical parameters used to determine association in the rule, such as adjusting the threshold for determining abnormal fluctuations in a physiological parameter from a preset value of X to X + ΔX, where ΔX is determined by the rule adjustment instruction. Adjustment can also involve adding, deleting, or modifying the weight of logical conditions in the rule, such as adding a judgment condition in a specific environment that requires both abnormal fluctuations in physiological parameter A and abnormal changes in environmental parameter B to determine the existence of an association, or increasing the weight of abnormal fluctuations in a parameter in the association judgment. Through this adjustment, a general preset rule is transformed into a more targeted, adjusted rule that better suits the current individual physiological state of the firefighter and the characteristics of the environment in which they are located.

[0094] Finally, using the adjusted association analysis rules, the obtained response results are combined with abnormal fluctuations in monitoring parameters from multiple different sources to determine whether there is a pre-set correlation between the response results and the abnormal fluctuations in the monitoring parameters from these multiple sources, indicating that the firefighter failed to effectively respond to the status confirmation request due to physiological discomfort. Unlike directly using fixed, preset rules for judgment, this method uses dynamically adjusted rules. This adjusted rule more accurately assesses the strength and nature of the association between the firefighter's failure to respond to the status confirmation request and abnormal fluctuations in monitoring parameters under the current individual physiological state and environmental conditions, thereby more reliably determining whether there is a risk of early disability due to physiological discomfort. For example, when the ambient temperature is high, the adjusted rule may allow for a wider range of heart rate fluctuations. When detecting trace amounts of toxic gas, the adjusted rule may be more sensitive to small fluctuations in respiratory rate. By using adjusted rules for judgment, misjudgments caused by individual differences or environmental changes can be reduced, improving the accuracy of early disability risk identification. The judgment result is then used to determine whether the firefighter is at risk of early disability and trigger a corresponding warning or alarm.

[0095] Through the above steps, the scheme realizes dynamic adaptive adjustment of association analysis rules, enabling it to better adapt to individual differences among firefighters and changes in the rescue site environment, improving the accuracy and robustness of early identification of disability risks, thereby enabling more timely and reliable warnings or alarms, and providing more effective technical means to ensure the safety of firefighters.

[0096] Specifically, in some of the above-mentioned solutions of the present application, it is proposed to determine one or more rule adjustment instructions for pre-set association analysis rules based on the difference between real-time physiological data and the baseline physiological data of firefighters, or based on the changing characteristics of real-time environmental parameters, so as to make the association analysis rules more adaptable to individual differences and environmental changes. However, in this process, real-time physiological data and real-time environmental parameters may change at the same time, and the direction or degree of rule adjustment indicated by them may be inconsistent or even conflicting. Simply describing the method of determining rule adjustment instructions fails to address the potential conflict between the adjustment requirements from these two different sources or to comprehensively utilize the information of both according to a specific goal. This may result in the determined rule adjustment instructions failing to accurately reflect the actual situation or consider all relevant factors, affecting the credibility of subsequent association judgments based on the changed rules, and may increase the possibility of erroneous reports or missed reports.

[0097] In this regard, the present application further proposes that the steps of determining one or more rule adjustment instructions for pre-set association analysis rules based on the difference between the real-time physiological data and the firefighter's baseline physiological data, or based on the changing characteristics of the real-time environmental parameters, include:

[0098] obtaining a first set of candidate rule adjustment parameters determined by differences between the real-time physiological data and a baseline physiological data of the firefighter;

[0099] obtaining a second set of candidate rule adjustment parameters determined by the change characteristics of the real-time environmental parameters;

[0100] Determining whether there is an indication conflict between the first set of candidate rule adjustment parameters and the second set of candidate rule adjustment parameters according to a pre-set conflict determination rule;

[0101] When it is determined that there is an indication conflict, one or more rule adjustment instructions for the pre-set association analysis rule are determined based on a pre-set conflict handling mechanism and in combination with the first set of candidate rule adjustment parameters and the second set of candidate rule adjustment parameters;

[0102] When it is determined that there is no indication conflict, the first set of candidate rule adjustment parameters determined based on the difference between the real-time physiological data and the firefighter's baseline physiological data, or the second set of candidate rule adjustment parameters determined based on the changing characteristics of the real-time environmental parameters, or the combination of the first set of candidate rule adjustment parameters and the second set of candidate rule adjustment parameters, are used to determine one or more rule adjustment instructions for the pre-set association analysis rules.

[0103] Among them, this solution provides detailed regulations and improvements for the steps of determining rule adjustment instructions, aiming to accurately and stably generate rule adjustment instructions and be able to handle situations where there are conflicts in adjustment requirements from different data sources.

[0104] First, by obtaining a first set of candidate rule adjustment parameters based on the difference between real-time physiological data and the firefighter's baseline physiological data, and a second set of candidate rule adjustment parameters based on the changing characteristics of real-time environmental parameters, this solution distinguishes between adjustment suggestions derived from physiological and environmental states, providing a basis for resolving potential conflicts. For example, the first set of candidate rule adjustment parameters might include rule weight adjustment suggestions calculated based on the degree of deviation of heart rate and respiratory rate from baseline values, while the second set of candidate rule adjustment parameters might include rule threshold adjustment suggestions calculated based on the rate of change of ambient temperature and toxic gas concentration.

[0105] Next, based on pre-defined conflict resolution rules, a determination is made as to whether there is a conflict between the two sets of candidate rule adjustment parameters. This step identifies situations requiring special handling. The conflict resolution rules can be a set of logical conditions. For example, if the first set of parameters indicates a lowering of a threshold, while the second set indicates an increase, a conflict is determined. When a conflict is determined, this solution no longer selects a single source. Instead, it determines one or more final rule adjustment instructions based on a pre-defined conflict resolution mechanism, combining the first and second sets of candidate rule adjustment parameters. This means that the system integrates, coordinates, or selectively applies adjustment suggestions from different sources to generate a logically consistent and objectively targeted adjustment instruction, thus resolving conflicting adjustment requests from different data sources. For example, the conflict resolution mechanism can use a weighted average approach, assigning weights based on the current credibility of physiological and environmental data to calculate the final adjustment parameters. Alternatively, it can prioritize physiological or environmental data recommendations in specific conflict scenarios based on pre-defined priority rules.

[0106] When it is determined that no conflicting instructions exist, this solution determines one or more rule adjustment instructions based on a first set of candidate rule adjustment parameters determined by the difference between the real-time physiological data and the firefighter's baseline physiological data, or a second set of candidate rule adjustment parameters determined by the changing characteristics of real-time environmental parameters, or a combination of the first and second sets of candidate rule adjustment parameters. This provides the possibility of selecting an adjustment source that meets specific objectives or considering both sources in a comprehensive manner when there is no conflict. For example, when there is no conflict, the first or second set of parameters can be directly adopted, or the two sets of parameters can be simply superimposed or averaged. By introducing candidate parameters, conflict determination, and conflict resolution mechanisms, this solution can more thoroughly address adjustment requirements from different monitoring sources. When these requirements conflict, they can be coordinated according to a pre-defined plan to ensure that the numerical indicators meet specific targets, thereby generating more accurate and stable rule adjustment instructions.

[0107] As a result, the modified association analysis rules are more capable of achieving their intended functions, helping to more accurately identify firefighters' early disability risk. By explicitly identifying and addressing potential conflicts between physiological and environmental data in determining rule adjustment instructions, this solution resolves complex situations that cannot be handled by simple "or" logic, improving the accuracy and stability of rule adjustments, thereby enhancing the credibility of early disability risk assessments and reducing the likelihood of erroneous or missed reports.

[0108] Specifically, some of the aforementioned solutions in this application propose determining one or more rule adjustment instructions for a preset association analysis rule based on a preset conflict resolution mechanism, combining a first set of candidate rule adjustment parameters determined by the difference between real-time physiological data and baseline physiological data, and a second set of candidate rule adjustment parameters determined by the changing characteristics of real-time environmental parameters. This mechanism is used to process and generate adjustment instructions when there is a conflict between the direction or degree of rule adjustment indicated by the physiological data and the environmental parameters. However, in actual applications, the preset conflict resolution mechanism may include multiple conflict resolution strategies. When the conflict situation formed by the first set of candidate rule adjustment parameters and the second set of candidate rule adjustment parameters is complex, such as when the conditions for activating multiple conflict resolution strategies are simultaneously met, or when the conditions for activating any of the preset conflict resolution strategies are not fully met, how to select the most appropriate one from these strategies, or how to effectively combine these strategies to generate rule adjustment instructions that can more accurately and robustly reflect the current complex situation, is a problem that existing solutions have not fully addressed. Simply applying fixed strategies or lacking a flexible strategy selection / combination mechanism may result in inaccurate rule adjustment instructions, affecting the effectiveness of subsequent association analysis rule adjustments, and thus the accuracy and timeliness of early disability risk identification.

[0109] In this regard, the present application further proposes that when the preset conflict handling mechanism includes at least two conflict handling strategies, and the current conflict situation formed by the first set of candidate rule adjustment parameters and the second set of candidate rule adjustment parameters simultaneously meets the conditions for activating the at least two conflict handling strategies, or the conflict situation does not fully meet the activation conditions of any preset conflict handling strategy, the steps of determining one or more rule adjustment instructions based on the preset conflict handling mechanism and in combination with the first set of candidate rule adjustment parameters and the second set of candidate rule adjustment parameters include:

[0110] Obtaining quantitative characteristics of conflict situations;

[0111] Obtaining a time series variation trend of adjustment parameters of the first set of candidate rules and a time series variation trend of adjustment parameters of the second set of candidate rules;

[0112] Obtain feedback data on the execution effects of historical rule adjustment instructions related to historical conflict situations, historical parameter trends, and historically adopted conflict resolution strategies;

[0113] Based on a preset set of meta-rules and in combination with the obtained quantitative characteristics of the conflict situation, the obtained time series change trend, and the obtained feedback data, a conflict resolution strategy is selected from at least two conflict resolution strategies included in a preset conflict resolution mechanism, or at least two conflict resolution strategies are combined to generate a current conflict resolution solution;

[0114] The current conflict handling solution is applied, and one or more rule adjustment instructions are determined in combination with the first set of candidate rule adjustment parameters and the second set of candidate rule adjustment parameters.

[0115] Among them, when the preset conflict handling mechanism includes at least two conflict handling strategies, and the current conflict situation composed of the first set of candidate rule adjustment parameters and the second set of candidate rule adjustment parameters is complex or ambiguous, the quantitative characteristics of the conflict situation are first obtained. The quantitative characteristics of the conflict situation can be a numerical or classified representation of the degree of conflict, the difference in conflict direction, the number of parameters involved, etc. For example, the angular difference between the adjustment directions (such as increasing or decreasing the rule weight) indicated by the first and second sets of parameters can be calculated, or the ratio or difference between the adjustment amplitudes indicated by them can be calculated. These quantitative characteristics provide an objective description of the current conflict state.

[0116] Furthermore, the time series change trends of the adjustment parameters of the first set of candidate rules and the second set of candidate rule adjustment parameters are obtained. Time series change trends can include the rate of change of parameter values ​​over time, acceleration, volatility, periodicity, etc. For example, it can be calculated whether the adjustment parameters indicated by physiological parameters have been continuously increasing or fluctuating significantly over a recent period of time, and whether the adjustment parameters indicated by environmental parameters are stable or changing rapidly. This trend information reflects the dynamic evolution of the conflict and its potential development direction.

[0117] In addition, feedback data on the effectiveness of historical rule adjustment instructions related to historical conflict situations, historical parameter trends, and historically adopted conflict resolution strategies is obtained. This feedback data can include the conflict resolution strategies used when similar conflict situations and parameter trends occurred in the past, as well as an assessment of the effectiveness of the rule adjustment instructions generated by that strategy on subsequent association analysis rules (for example, whether they successfully identified early risks or reduced false positives or negatives). This historical data constitutes an empirical knowledge base that guides decision-making in current complex situations.

[0118] Based on a preset set of meta-rules, combined with the acquired quantitative characteristics of the conflict situation, the acquired time series trends, and the acquired feedback data, a conflict resolution strategy is selected from at least two conflict resolution strategies included in the preset conflict resolution mechanism, or at least two conflict resolution strategies are combined to generate the current conflict resolution solution. A meta-rule set is a set of higher-level rules that guide how to select or combine basic conflict resolution strategies based on the current specific situation (described by quantitative characteristics and time trends) and historical experience (provided by feedback data). For example, a meta-rule might stipulate: If the conflict level is high and the physiological parameter trend shows rapid deterioration, a conservative strategy (i.e., more likely to trigger an alert) is preferred; if the conflict level is moderate and historical data shows that a certain strategy is most effective under similar trends, that strategy is selected; if the current situation shares similarities with multiple historical scenarios, or if a single strategy is insufficient, the meta-rule might indicate that multiple strategies should be combined, such as by weighted averaging or sequential execution. The meta-rules can be based on the experience of domain experts.

[0119] That is, the meta-rule set is a logical set, the content of which is the rules guiding how to select or combine strategies, and defines the logic for selecting one of the at least two conflict handling strategies included in the preset conflict handling mechanism, or determining the combination of the at least two conflict handling strategies, based on the quantitative characteristics of the conflict situation, the time series change trend, and the feedback data.

[0120] Apply the current conflict resolution scheme, and in combination with the first set of candidate rule adjustment parameters and the second set of candidate rule adjustment parameters, determine one or more rule adjustment instructions. Once the current optimal conflict resolution scheme (whether a single strategy or a combination of strategies) is determined, the scheme is applied to the candidate adjustment parameters generated by the physiological data and environmental data, and the specific instructions for adjusting the association analysis rules are finally calculated. For example, if the selected strategy is to take the average of the two, the final instruction may be the average of the adjustment values ​​indicated by the first and second sets of parameters; if the selected strategy is a weighted combination, the weighted average is calculated based on the determined weights; if the selected strategy is to prioritize physiological data, the final instruction is mainly determined based on the first set of parameters.

[0121] By incorporating quantitative characteristics of conflict situations, time series trends of parameters, and historical feedback data, and utilizing meta-rules to select or combine strategies, the system no longer simply executes fixed conflict handling logic but instead learns and adapts, dynamically adjusting its approach based on specific situations. This dynamic, experience-based strategy decision-making process enables the generated rule adjustment instructions to more accurately reflect the current complex physiological and environmental conditions, thereby improving the precision and robustness of association analysis rule adjustments.

[0122] Specifically, in some of the above-mentioned schemes of the present application, it is proposed to select a conflict handling strategy from at least two conflict handling strategies included in the preset conflict handling mechanism based on a preset meta-rule set and in combination with the obtained quantitative characteristics of the conflict situation, the obtained time series change trend, and the obtained feedback data, or to combine at least two conflict handling strategies to generate a current conflict handling solution for handling the indicated conflict between a first set of candidate rule adjustment parameters determined by the difference between the real-time physiological data and the firefighter's baseline physiological data and a second set of candidate rule adjustment parameters determined by the change characteristics of the real-time environmental parameters. However, when the meta-rule indicates that at least two conflict handling strategies need to be combined, how to determine the specific combination method (such as weight, order, fusion logic, etc.) so that the generated current conflict handling solution can most effectively resolve the current conflict situation and ultimately generate the optimal rule adjustment instructions is an issue that has not been elaborated in detail or may face challenges in the above-mentioned schemes. Simply combining strategies may not fully utilize the current conflict characteristics, trends, and historical experience, resulting in suboptimal rule adjustment instructions, thereby affecting the accuracy of subsequent association analysis rule adjustments and the timeliness and accuracy of early disability risk determination.

[0123] In this regard, the present application further proposes that when a meta-rule indicates that at least two conflict handling strategies need to be combined to generate a current conflict handling solution, the steps of combining the at least two conflict handling strategies to generate the current conflict handling solution based on a preset meta-rule set and in combination with the obtained quantitative characteristics of the conflict situation, the obtained time series change trend, and the obtained feedback data include:

[0124] Identifying a combination parameter of at least two conflict handling strategies, the number of which is greater than one and which can be adjusted, where the combination parameter is used to characterize a combination mode of the at least two conflict handling strategies, the combination mode including at least one of weight distribution, execution order, and fusion logic;

[0125] Based on the indication of the meta-rule, or with reference to the valid combination parameters recorded in the feedback data similar to the quantitative characteristics and time series change trend of the conflict situation at the moment, a set of combination parameter values ​​as a starting point is set for the combination parameters that are greater than one and can be adjusted;

[0126] Generating a set of combination parameter values ​​greater than one and different for selection by adjusting the value of at least one parameter among the combination parameters greater than one and adjustable;

[0127] Using the preset combination effect evaluation logic, for each generated set of optional combination parameter values, and in combination with the quantitative characteristics of the conflict situation at that moment, the time series change trend, and the feedback data, a numerical evaluation index of the corresponding expected effect is calculated. The combination effect evaluation logic is used to output a numerical evaluation index of the expected effect of combining the optional combination parameter values ​​based on the input quantitative characteristics of the conflict situation at that moment, the time series change trend, the feedback data, and the set of optional combination parameter values;

[0128] From the calculated numerical evaluation indicators of each set of selectable combination parameter values, a set of combination parameter values ​​that makes the numerical evaluation indicators meet the pre-set optimization goal is selected as the final determined combination parameter;

[0129] The finally determined combination parameter selected is applied to combine at least two conflict handling strategies to generate a current conflict handling solution.

[0130] Specifically, this solution provides a systematic, data-based optimization method for determining the optimal combination when meta-rules indicate the need to combine multiple conflict resolution strategies.

[0131] First, identify the combination parameters of at least two conflict handling strategies, the number of which is greater than one and can be adjusted. These parameters characterize the combination of strategies, such as the weights of different strategies, the order of execution, or the fusion logic of the results. This step is the basis for combinatorial optimization and clarifies the variables that can be adjusted. Specifically, for two strategies A and B, the combination parameters may include the weight of strategy A (for example, a value between 0 and 1), the weight of strategy B, a Boolean value indicating whether strategy A is executed before strategy B, or specific function parameters used to fuse the output results of strategies A and B. The number of these parameters is greater than one and can be adjusted within a certain range.

[0132] Next, based on the meta-rules' indications or reference to effective combination parameters in similar situations in historical feedback data, a set of starting values ​​is set for these adjustable combination parameters. This provides a reasonable starting point, helping to improve the efficiency of subsequent optimization searches. For example, if the meta-rules indicate a preference for a certain strategy combination based on the current conflict characteristics, or if historical feedback data shows that a specific weight distribution works well in similar conflict situations, these indications or historical data can be used as initial combination parameter values. This step leverages the meta-rules and feedback data obtained from previous solutions, providing useful prior information for the subsequent optimization process.

[0133] Then, by adjusting the value of at least one of these combined parameters, a number greater than one, distinct, and selectable set of combined parameter values ​​is generated. This step is intended to explore different possible combinations and provide alternatives for subsequent selection. For example, if the initial weight combination is (0.5, 0.5), by increasing or decreasing one of the weights, multiple different weight combination sets can be generated, such as (0.6, 0.4), (0.4, 0.6), and so on. If the combination method includes the execution order, different order sets can be generated, such as executing strategy A first and strategy B first.

[0134] Subsequently, using the preset combination effect evaluation logic, for each generated set of optional combination parameter values, the numerical evaluation index of the expected effect of using that combination parameter value is calculated, combining the current quantitative characteristics of the conflict situation, time series change trends, and historical feedback data. The combination effect evaluation logic is the core, capable of predicting the possible effects of the combination scheme based on the current specific situation and the set combination parameters, converting qualitative combination methods into quantifiable evaluation indicators. This evaluation logic receives the quantitative characteristics and trends of the current conflict (obtained from previous schemes), as well as historical feedback data, and outputs a numerical value based on the combination parameter value set to be evaluated, such as the accuracy of the predicted rule adjustment instructions, the timeliness of early disability risk assessment, or the false alarm rate. By combining this contextual information (characteristics, trends, feedback) with the combination parameters to be evaluated for evaluation, this solution can dynamically and context-awarely predict the effects of different combination methods. This is different from methods that simply fix the combination method or select strategies based on a single factor, reflecting its adaptability to complex situations.

[0135] Then, from the multiple numerical evaluation indicators calculated, the combination parameter value set that ensures the evaluation indicator meets the pre-set optimization goal (for example, maximizing the expected effect or minimizing the expected risk) is selected and determined as the final combination parameter to be adopted. This step makes decisions based on quantitative evaluation and ensures that the selected combination solution is the optimal one. For example, if the optimization goal is to maximize the expected accuracy of rule adjustments, the combination parameter value set corresponding to the highest evaluation indicator is selected.

[0136] Finally, the final combination parameters selected are applied to combine at least two conflict handling strategies to generate the optimal conflict handling solution for the current specific conflict situation. This step converts the optimization results into a practical executable solution to guide the subsequent generation of rule adjustment instructions. Through the above steps, this solution can dynamically and data-drivenly determine the optimal combination of multiple conflict handling strategies based on the current specific situation and historical experience, thereby improving the effectiveness and accuracy of conflict handling, and then optimizing the adjustment of association analysis rules, and ultimately improving the reliability of early disability risk warning. By utilizing the conflict characteristics, trends and feedback data obtained from previous solutions to drive the optimization evaluation of the combination method, this solution can generate a conflict handling solution that is more suitable for the current situation than a fixed combination or simple selection, thereby more effectively solving the rule adjustment problem caused by the conflict of physiological and environmental parameters, and improving the performance of the overall warning system.

[0137] As a preferred embodiment, the solution of this application is specifically implemented as follows:

[0138] This solution is applied when the meta-rule indicates that at least two conflict handling strategies need to be combined to generate the current conflict handling solution, for example, when the rule adjustment direction indicated by the physiological data conflicts with the rule adjustment direction indicated by the environmental parameters, and the meta-rule determines that multiple processing methods need to be integrated based on the current conflict intensity and historical patterns. First, the combination parameters of the conflict handling strategies to be combined are identified. For example, strategy one may focus on adjustments based on physiological trends, and strategy two may focus on adjustments based on the magnitude of environmental changes. Their combination method can be characterized by weight parameters, for example, assigning a weight W1 to strategy one and a weight W2 to strategy two, where W1+W2=1. These weight parameters are combination parameters that are greater than one and can be adjusted.

[0139] Next, based on the meta-rule's indications, or by referencing effective combination parameters recorded in historical feedback data that are similar to the current conflict situation's quantitative characteristics (e.g., physiological fluctuation amplitude, environmental change rate) and time series change trends (e.g., physiological indicators continue to deteriorate, environmental indicators tend to stabilize), a set of starting values ​​is set for the identified combination parameters. For example, if historical data shows that strategies focusing on physiological trends are more effective in similar situations, the starting weight value set can be set to {W1 = 0.7, W2 = 0.3}.

[0140] Then, by adjusting the value of at least one of these combined parameters, a greater than one set of distinct, selectable combined parameter value sets is generated. For example, starting from the starting value {W1=0.7, W2=0.3}, a series of selectable sets can be generated, such as {W1=0.6, W2=0.4}, {W1=0.8, W2=0.2}, {W1=0.5, W2=0.5}, etc.

[0141] Then, using the preset combination effect evaluation logic, for each generated set of selectable combination parameter values, combined with the quantitative characteristics of the conflict situation at that moment, the time series trend, and feedback data, a numerical evaluation index of the corresponding expected effect is calculated. Based on this combination effect evaluation logic, the current situation data (conflict characteristics, trends, feedback) and the set of combination parameter values ​​to be evaluated are input, and a numerical value is output, such as the accuracy score of the predicted rule adjustment instruction or the comprehensive score of the predicted false positive / missing negative risk.

[0142] Then, from the calculated numerical evaluation indicators for each set of selectable combined parameter values, a set of combined parameter values ​​is selected that satisfies the pre-set optimization goal. For example, if the optimization goal is to maximize the accuracy score of the rule adjustment instruction, the set of combined parameter values ​​with the highest score is selected.

[0143] Finally, the final combination parameters are applied to combine at least two conflict resolution strategies to generate the current conflict resolution solution. For example, if the final combination parameters are {W1 = 0.75, W2 = 0.25}, the outputs of strategy 1 and strategy 2 are merged in a ratio of 75% and 25% to form the final rule adjustment instruction.

[0144] Through the above technical solution, this application solves the problem of how to determine the specific combination method when the meta-rule indicates that multiple conflict handling strategies need to be combined so that the generated current conflict handling solution can most effectively solve the current conflict situation and ultimately generate the optimal rule adjustment instructions. By identifying adjustable combination parameters, setting a starting point based on historical experience and the current situation, systematically exploring different combination possibilities, and using data-driven evaluation logic to quantitatively predict the effect of each combination solution, the optimal combination parameters are finally selected. As a result, the generated rule adjustment instructions can more accurately reflect the current complex situation, improve the effectiveness of subsequent association analysis rule adjustments, and thereby improve the timeliness and accuracy of early disability risk determination.

[0145] Specifically, the preset combination effect evaluation logic includes at least:

[0146] Mapping the input conflict situation's quantitative characteristics, time series change trends, key elements in the feedback data, and the currently selected combined parameter values ​​to be evaluated into a set of predefined scoring items;

[0147] Set corresponding weight coefficients and scoring rules for each scoring item;

[0148] Numerical evaluation indicators are obtained by calculating the scores of each scoring item and performing weighted summation or logical reasoning.

[0149] The input quantitative characteristics of the conflict situation, the time series change trend, the key elements in the feedback data, and the optional combined parameter values ​​to be evaluated are mapped into a set of predefined scoring items:

[0150] The purpose of this step is to convert the complex input information of the combination effect evaluation logic (including the quantitative characteristics of the current conflict, the time series change trend of the parameters, the key information in the historical feedback data, and the specific set of combination parameter values ​​being evaluated) into a series of standardized, quantifiable evaluation scoring items. The predefined scoring items are determined in advance based on fire rescue expertise and system design goals, and can reflect the evaluation dimensions of different aspects of the combination effect. For example, one scoring item can correspond to the urgency of the current conflict, another scoring item can correspond to the success rate of the selected combination parameters in similar high-risk scenarios in history, and another scoring item can correspond to the impact of the combination parameters on the system response time. By mapping diverse input information into these standardized scoring items, the foundation is laid for subsequent quantitative evaluation and comprehensive calculation, so that information from different sources and of different natures can be considered within a unified framework.

[0151] Set the corresponding weight coefficient and scoring rules for each scoring item:

[0152] The purpose of this step is to assign relative importance to each scoring item defined in the previous step in the overall evaluation, and to clarify how to give specific scores based on the actual situation of the scoring items. The weight coefficient reflects the degree of influence of different scoring items on the final combined effect evaluation results. For example, in some scenarios, the weight of the scoring item with a high success rate indicated in the historical feedback data may be higher than the calculation complexity of the combination parameters. The scoring rules define how to calculate the score of each scoring item based on the input value of each scoring item (for example, the specific value of the conflict urgency, the specific percentage of the historical success rate). It can be a simple linear mapping or a more complex nonlinear function or logical judgment. By setting weight coefficients and scoring rules, the system can carry out focused and standardized quantification of various influencing factors based on preset evaluation strategies and professional knowledge.

[0153] By calculating the scores of each scoring item and performing weighted summation or logical reasoning, numerical evaluation indicators are obtained;

[0154] The purpose of this step is to combine the quantitative scores of each scoring item to form a final numerical evaluation index that can represent the expected effect of the current optional combination parameter value to be evaluated. The score of each scoring item is calculated by scoring the actual input value of each scoring item according to the scoring rules set in the previous step. Then, by performing a weighted summation, that is, multiplying the score of each scoring item by its corresponding weight coefficient and adding them together, a comprehensive numerical value is obtained. Alternatively, through logical reasoning, for example, based on a set of "if-then" rules, the final evaluation index value is inferred based on the combination of the scores of each scoring item. Regardless of whether weighted summation or logical reasoning is used, the purpose is to aggregate multi-dimensional scoring information into a single numerical evaluation index that can intuitively reflect the quality of the combination effect, and compare and select between multiple sets of optional combination parameter values ​​to determine the best combination of conflict resolution strategies.

[0155] In the solution of the present application, the monitoring parameters from multiple different sources include at least two of: environmental gas concentration parameters, firefighter breathing parameters, firefighter posture parameters, firefighter heart rate parameters, and firefighter body surface temperature parameters.

[0156] Ambient gas concentration parameters refer to the concentration of specific or unknown gases in the surrounding environment as monitored by gas sensors on firefighter suits. Even slight increases in trace irritant gas concentrations can be a direct environmental factor that can cause physiological discomfort in firefighters. Even if these increases do not reach the independent gas alarm threshold, these changes can still provide important information indicating potential risks.

[0157] Firefighter breathing parameters refer to data such as respiratory rate and instantaneous oxygen consumption rate, as monitored by oxygen respirators. Coughing and discomfort can cause fluctuations in respiratory rate and instantaneous oxygen consumption rate. Even if these fluctuations do not trigger a hypoxia alarm, they are important indicators of physiological abnormalities.

[0158] Firefighter posture parameters refer to data such as body sway and activity levels, as monitored by posture sensors on firefighter uniforms. Dizziness can cause body sway and slow movement. Even if these posture changes don't reach the alarm conditions of a fall or prolonged inactivity, they can still be important indicators of potential impairment in mobility.

[0159] Firefighter heart rate and skin temperature parameters are physiological indicators monitored by firefighting uniforms. These parameters can reflect firefighters' physiological load and stress state, and abnormal fluctuations in these parameters may also be associated with physiological discomfort.

[0160] Using these parameters from different sensors and categories as multiple monitoring sources ensures the system can capture sub-threshold anomaly information related to firefighters' physiological discomfort and environmental factors from multiple dimensions. The combination and correlation analysis of these parameters is fundamental to identifying early, hidden incapacitation risks that are difficult to detect using traditional single-threshold alarm methods.

[0161] In the solution of the present application, the real-time environmental parameter includes at least one of the ambient temperature and the gas concentration.

[0162] Ambient temperature and gas concentration are key environmental factors that firefighters may directly perceive and influence their physiological state in complex rescue environments. These specific parameters serve as inputs for determining rule adjustment instructions. Their contribution lies in providing objective data representing the state of the firefighter's microenvironment. This allows the system to adjust association analysis rules based on environmental changes, improving the consistency of risk assessments.

[0163] Secondly, refer to Figure 2 , the present application further proposes an oxygen respirator monitoring and alarm system, the system comprising:

[0164] The first acquisition module 1 is used to obtain the response result of the firefighter to the preset status confirmation request;

[0165] a second acquisition module 2 for acquiring monitoring parameters from multiple different sources related to the firefighter's physiological discomfort, wherein the monitoring parameters from the multiple different sources each fail to reach a preset independent alarm threshold, and the monitoring parameters from the multiple different sources exhibit abnormal fluctuations indicative of physiological discomfort;

[0166] a judgment module 3, configured to judge, based on a preset correlation analysis rule, whether there is a preset correlation between the response result obtained by the response result obtaining module and abnormal fluctuations of the monitoring parameters from multiple different sources obtained by the monitoring parameter obtaining module, indicating that the firefighter failed to effectively respond to the status confirmation request due to physiological discomfort;

[0167] The early warning module 4 is used to determine that the firefighter is at risk of early disability due to reasons other than hypoxia, where the oxygen supply is sufficient but the ability to move may be impaired, when the correlation determined by the correlation judgment module meets the preset correlation analysis rules, and trigger a warning or alarm for the early disability risk based on the determined early disability risk.

[0168] By obtaining the firefighters' response results to status confirmation requests and abnormal fluctuations in multiple sub-threshold monitoring parameters, and judging whether there is a correlation of failure to respond effectively due to physiological discomfort based on correlation analysis, the early disability risks caused by non-hypoxia reasons can be identified and early warnings or alarms can be triggered. It has the ability to effectively identify the early disability risks of firefighters caused by multiple abnormal combinations that are difficult to detect with existing technologies, and provide early warnings or alarms at the early stage when the firefighters' oxygen supply is sufficient but their mobility may be impaired, thereby improving the accuracy of the alarm and making up for the shortcomings of traditional oxygen respirator alarm systems that mainly focus on oxygen supply safety.

[0169] In addition, in some preferred embodiments, an oxygen respirator monitoring and alarm system proposed in this application can perform any one of the steps in the above method.

[0170] The above description is merely an embodiment of the present application and is not intended to limit the scope of protection of the present application. For those skilled in the art, various modifications and variations of the present application are possible. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A monitoring and alarm method for an oxygen respirator, characterized in that: include: Obtain the firefighter's response to the preset status confirmation request; obtaining monitoring parameters from a plurality of different sources related to the physiological discomfort state of the firefighter; determining, based on a preset correlation analysis rule, whether there is a preset correlation between the acquired response result and the abnormal fluctuations of the monitoring parameters acquired from the multiple different sources, indicating that the firefighter failed to effectively respond to the status confirmation request due to the physiological discomfort; When the determined correlation satisfies the preset correlation analysis rule, determining that the firefighter is at risk of early disability due to reasons other than hypoxia, where oxygen supply is sufficient but mobility may be impaired, and triggering a warning or alarm for the early disability risk based on the determined early disability risk; The step of determining, based on a preset correlation analysis rule, whether there is a preset correlation between the acquired response result and the abnormal fluctuations of the monitoring parameters acquired from the multiple different sources, indicating that the firefighter failed to effectively respond to the status confirmation request due to the physiological discomfort, includes: Acquiring real-time physiological data indicating current individual physiological characteristics of the firefighter and real-time environmental parameters indicating current environmental conditions at the rescue site; Determining one or more rule adjustment instructions for the preset association analysis rule based on a difference between the real-time physiological data and the baseline physiological data of the firefighter, or based on a change characteristic of the real-time environmental parameter; Applying the one or more rule adjustment instructions to adjust at least one parameter or logical condition in the preset association analysis rule to generate an adjusted association analysis rule; Using the adjusted association analysis rule, in combination with the acquired response result and the abnormal fluctuations of the monitoring parameters acquired from the multiple different sources, determining whether there is a preset correlation between the response result and the abnormal fluctuations of the monitoring parameters acquired from the multiple different sources, indicating that the firefighter failed to effectively respond to the status confirmation request due to the physiological discomfort; The step of determining one or more rule adjustment instructions for the preset association analysis rule based on the difference between the real-time physiological data and the baseline physiological data of the firefighter, or based on the change characteristics of the real-time environmental parameters, includes: obtaining a first set of candidate rule adjustment parameters determined by a difference between the real-time physiological data and a baseline physiological data of the firefighter; Obtaining a second set of candidate rule adjustment parameters determined by the change characteristics of the real-time environmental parameters; determining, according to a preset conflict determination rule, whether there is an indication conflict between the first set of candidate rule adjustment parameters and the second set of candidate rule adjustment parameters; When it is determined that there is an indication conflict, determining the one or more rule adjustment instructions for the preset association analysis rule in combination with the first set of candidate rule adjustment parameters and the second set of candidate rule adjustment parameters according to a preset conflict handling mechanism; When it is determined that there is no indication conflict, the first group of candidate rule adjustment parameters determined based on the difference between the real-time physiological data and the baseline physiological data of the firefighter, or the second group of candidate rule adjustment parameters determined based on the changing characteristics of the real-time environmental parameters, or the first group of candidate rule adjustment parameters and the second group of candidate rule adjustment parameters are combined to determine the one or more rule adjustment instructions for the preset association analysis rules.

2. The oxygen respirator monitoring and alarm method according to claim 1, characterized in that: When the preset conflict handling mechanism includes at least two conflict handling strategies, and the current conflict situation formed by the first set of candidate rule adjustment parameters and the second set of candidate rule adjustment parameters simultaneously satisfies conditions for activating the at least two conflict handling strategies, or the conflict situation does not fully satisfy the activation conditions of any preset conflict handling strategy, the step of determining the one or more rule adjustment instructions based on the preset conflict handling mechanism and in combination with the first set of candidate rule adjustment parameters and the second set of candidate rule adjustment parameters includes: obtaining quantitative characteristics of the conflict situation; Obtaining a time series variation trend of the first set of candidate rule adjustment parameters and a time series variation trend of the second set of candidate rule adjustment parameters; Obtain feedback data on the execution effects of historical rule adjustment instructions related to historical conflict situations, historical parameter trends, and historically adopted conflict resolution strategies; selecting, based on a preset set of meta-rules and in combination with the obtained quantitative characteristics of the conflict situation, the obtained time series change trend, and the obtained feedback data, a conflict handling strategy from at least two conflict handling strategies included in the preset conflict handling mechanism, or combining the at least two conflict handling strategies to generate a current conflict handling solution; The one or more rule adjustment instructions are determined by applying the current conflict handling solution and combining the first set of candidate rule adjustment parameters and the second set of candidate rule adjustment parameters.

3. The oxygen respirator monitoring and alarm method according to claim 2, characterized in that: When the meta-rule indicates that the at least two conflict handling strategies need to be combined to generate the current conflict handling solution, the step of combining the at least two conflict handling strategies to generate the current conflict handling solution based on the preset meta-rule set and in combination with the obtained quantitative characteristics of the conflict situation, the obtained time series change trend, and the obtained feedback data includes: Identifying a combination parameter of the at least two conflict handling strategies, the number of which is greater than one and which can be adjusted, the combination parameter being used to characterize a combination mode of the at least two conflict handling strategies, the combination mode including at least one of weight distribution, execution order, and fusion logic; Based on the indication of the meta-rule, or with reference to the valid combination parameters recorded in the feedback data that are similar to the quantitative characteristics of the conflict situation at that moment and the time series change trend, setting a set of combination parameter values ​​as a starting point for the obtained combination parameters that are greater than one and can be adjusted; Generating a set of combination parameter values ​​greater than one and different for selection by adjusting the value of at least one parameter among the combination parameters greater than one and adjustable; Using a preset combination effect evaluation logic, for each generated set of selectable combination parameter values, and in combination with the quantitative characteristics of the conflict situation at that moment, the time series change trend, and the feedback data, a numerical evaluation index of the corresponding expected effect is calculated. The combination effect evaluation logic is used to output a numerical evaluation index of the expected effect of combining the selectable combination parameter values ​​based on the input quantitative characteristics of the conflict situation at that moment, the time series change trend, the feedback data, and the set of selectable combination parameter values; From the calculated numerical evaluation indicators of each set of selectable combination parameter values, select a set of combination parameter values ​​that makes the numerical evaluation indicators meet a pre-set optimization goal as the final determined combination parameter; The at least two conflict handling strategies are combined by applying the selected finally determined combination parameter to generate the current conflict handling solution.

4. The oxygen respirator monitoring and alarm method according to claim 1, characterized in that: The monitoring parameters from multiple different sources include at least two of: environmental gas concentration parameters, firefighter breathing parameters, firefighter posture parameters, firefighter heart rate parameters, and firefighter body surface temperature parameters.

5. The oxygen respirator monitoring and alarm method according to claim 1, characterized in that: The real-time environmental parameter includes at least one of ambient temperature and gas concentration.

6. The oxygen respirator monitoring and alarm method according to claim 2, characterized in that: The meta-rule set is a set of rules that defines the logic for selecting one of the at least two conflict handling strategies included in the preset conflict handling mechanism, or determining the combination of the at least two conflict handling strategies based on the quantitative characteristics of the conflict situation, the time series change trend, and the feedback data.

7. The oxygen respirator monitoring and alarm method according to claim 2, characterized in that: The preset combination effect evaluation logic includes at least: Mapping the input quantitative characteristics of the conflict situation, the time series change trend, the key elements in the feedback data, and the currently selected combined parameter values ​​to be evaluated into a set of predefined scoring items; Set corresponding weight coefficients and scoring rules for each scoring item; The numerical evaluation index is obtained by calculating the score of each scoring item and performing weighted summation or logical reasoning.

8. An oxygen respirator monitoring and alarm system, used to execute the method according to any one of claims 1 to 7, characterized in that: The system includes: The first acquisition module is used to obtain the response result of the firefighter to the preset status confirmation request; a second acquisition module, configured to acquire monitoring parameters from a plurality of different sources related to the physiological discomfort state of the firefighter, wherein each of the monitoring parameters from the plurality of different sources does not reach a preset independent alarm threshold, and the monitoring parameters from the plurality of different sources exhibit abnormal fluctuations indicative of the physiological discomfort; a judgment module, configured to judge, based on preset correlation analysis rules, whether there is a preset correlation between the response result obtained by the response result obtaining module and abnormal fluctuations of the monitoring parameters from multiple different sources obtained by the monitoring parameter obtaining module, indicating that the firefighter failed to effectively respond to the status confirmation request due to the physiological discomfort; The early warning module is used to determine that the firefighter is at risk of early disability due to reasons other than hypoxia, where the oxygen supply is sufficient but the mobility may be impaired, when the correlation determined by the correlation judgment module meets the preset correlation analysis rules, and trigger a warning or alarm for the early disability risk based on the determined early disability risk.

Images