Closed space operation safety early warning method, system and assembly thereof

Abstract

The invention relates to the field of safety production management, and discloses a closed space operation safety early warning method, system and component thereof, and the method comprises the steps: S1, continuously collecting the air quality and temperature and humidity information in a closed space; s2, when the potential danger is detected, a ventilation system is automatically started to adjust the indoor environment and send a warning signal; s3, continuously detecting the state of the access control system at the entrance of the closed space; and S4, continuously collecting and analyzing the body movement data of the operator, and sending a warning signal to the related personnel when the operator is in a static state for a long time. By collecting and analyzing the body motion data of the operator, combining the input of the three-axis acceleration sensor and the gyroscope, and utilizing the deep learning model to classify the motion mode, the dangerous states such as the normal working state and the long-time static state are effectively distinguished. By accurately identifying motion abnormity, early warning is given out in time, rescue is triggered, and the safety risk in operation in the closed space is remarkably reduced.

Description

Technical Field

[0001] The present invention relates to the field of safety production management, and specifically to a safety warning method, system and its components for confined space operations. Background Art

[0002] In safety production management, especially in the safety control of operations in confined spaces, existing technologies have achieved a preliminary automated warning mechanism by installing various sensors to monitor environmental indicators and using video surveillance to assist in observing personnel dynamics. However, this mode relies highly on manual intervention and has certain limitations in terms of real-time response speed and accuracy.

[0003] Solutions of the prior art corresponding to this technical solution: Currently, most confined space safety management solutions mainly rely on a combination of fixedly deployed environmental monitoring sensors and remote video surveillance. When an anomaly is detected, an alarm is triggered through a preset threshold to notify on-site management personnel to intervene and handle it.

[0004] Defects of the prior art: Existing confined space operation safety warning systems lack effective means to monitor human activities, and cannot timely identify the situation where operators remain stationary for a long time, resulting in delays in rescue opportunities in case of emergencies. In addition, the automation level of existing systems is low, and they cannot intelligently adjust ventilation and air exchange strategies, affecting work efficiency and safety. Summary of the Invention

[0005] In view of the deficiencies of the prior art, the present invention provides a safety warning method, system and its components for confined space operations, which solves the problems of delayed rescue risks caused by the inability to effectively monitor the dynamic changes of individuals in the prior art, as well as the waste of human resources and potential safety hazards caused by insufficient automation.

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: A safety warning method for confined space operations, including the following steps: S1. Continuously collect air quality, temperature and humidity information in the confined space, and evaluate the suitability of the space environment according to preset standards; S2. According to the evaluation result of the space environment suitability, when a potential danger is detected, automatically start the ventilation system to adjust the indoor environment, and at the same time send a warning signal to relevant personnel; S3. Continuously detect the status of the access control system at the entrance of the confined space to ensure that the access control is in a continuously closed state, and ensure the isolation of the confined space; S4. Continuously collect and analyze the body movement data of the operator, and send a warning signal to relevant personnel when the operator remains stationary for a long time.

[0007] A safety warning system for confined space operations, including: Environmental monitoring module, which is used to collect air quality, temperature and humidity information in the confined space, and evaluate the suitability of the space environment according to preset standards; The ventilation system automatically starts the ventilation system to adjust the indoor environment and send a warning signal to relevant personnel when potential danger is detected based on the spatial environment suitability assessment results; The confined space entrance access control system is used to detect the status of the access control in real time and ensure that the access control is continuously closed to ensure the isolation of the confined space; The human motion monitoring module is used to continuously collect and analyze the body motion data of the operators, identify whether the operators have been stationary for a long time, and send warning signals to relevant personnel.

[0008] Preferably, the environmental monitoring module includes a gas sensor and a temperature and humidity sensor for monitoring oxygen concentration, carbon dioxide concentration, toxic gas content, and temperature and humidity parameters.

[0009] Preferably, the human motion monitoring module comprises: Three-axis accelerometer and gyroscope to collect operator motion data; The deep learning model training unit analyzes and iteratively optimizes the operator's movement patterns based on historical data sets of three-axis accelerometers and gyroscopes. It is used to analyze crawling, bending, standing, normal working conditions and long-term static conditions to avoid judging necessary working postures as potential dangers.

[0010] Preferably, the deep learning model is trained based on a multi-layer neural network, and a classification model is established by inputting three-axis acceleration sensor and gyroscope data to output the motion state category of the operator, including normal movement, small movement and static state.

[0011] Preferably, the human motion monitoring module further includes an alarm triggering module, which is used to automatically send an alarm signal to the monitoring center based on the analysis results of the deep learning model training unit and the motion status of the operator, and to cooperate with the environmental monitoring module and the ventilation system to adjust the operating status of the ventilation system.

[0012] Preferably, when the human motion monitoring module recognizes that the operator has been stationary for a long time and the accumulated time of the stationary state exceeds a set threshold, an alarm signal is triggered and a rescue procedure is started.

[0013] Preferably, the ventilation system includes an exhaust device and an air intake device, and adjusts the wind speed in real time to optimize the air quality according to the output results of the environment monitoring module and then the human movement monitoring module.

[0014] Preferably, the system further includes an edge computing module for locally analyzing environmental monitoring data and human motion monitoring data.

[0015] A safety warning component for confined space operations, comprising: An environmental monitoring component for collecting air quality, temperature, and humidity information inside a confined space and evaluating the suitability of the space environment according to preset standards; A ventilation component including an exhaust device and an intake device for automatically starting a ventilation system to adjust the indoor environment and sending a warning signal to relevant personnel when a potential hazard is detected based on the evaluation result of the space environment suitability; An access control component for the entrance to the confined space for continuously detecting the status of the access control to ensure that the access control remains closed at all times to guarantee the isolation of the confined space; A human motion monitoring component for continuously collecting and analyzing the body motion data of workers, identifying whether there is a situation where a worker remains stationary for a long time, and sending a warning signal to relevant personnel The present invention provides a safety warning method, system, and its components for confined space operations. It has the following beneficial effects: 1. By continuously collecting and analyzing the body motion data of workers, the present invention realizes real-time monitoring and accurate analysis of the dynamic state of workers. The human motion monitoring module combines the data input from a three-axis acceleration sensor and a gyroscope, and uses a deep learning model to classify the motion patterns of workers, which can effectively distinguish normal working states such as crawling, bending, standing, etc. from potential dangerous states such as remaining stationary for a long time. Through the accurate identification of the motion state, the present invention can issue an early warning and trigger a rescue procedure in a timely manner when a worker has an abnormality, significantly reducing the risk of life safety accidents in the confined space operation environment.

[0016] 2. By using edge computing technology to perform real-time analysis and processing of environmental data and motion state data, and combining with ventilation equipment to form an intelligent decision-making mechanism, the present invention can quickly start an emergency plan in case of an emergency. And by reducing the influence of dependence on the cloud and communication delay, the system can complete data processing and response within milliseconds, greatly improving the safety and controllability of confined space operations, and providing a more robust safety guarantee for high-risk operation scenarios.

[0017] 3. Through the linkage between the environmental monitoring module, the human motion monitoring module, the ventilation system, and the alarm trigger module, the present invention detects the environment and personnel actions. Therefore, when the environmental parameters exceed the standard and it is found that a person is stationary at the same time, the system can comprehensively judge the current risk state and automatically adjust the ventilation strategy, while triggering a multi-level alarm mechanism. This ensures all-round coverage and real-time optimization of the safety status of the confined space, and greatly improves the practicability and reliability of the system in complex operation environments. Brief Description of the Drawings

[0018] Figure 1 It is a schematic flowchart of the method of the present invention; Figure 2 It is a schematic diagram of the system architecture of the present invention. Detailed Embodiments

[0019] Next, in combination with the drawings in the specification of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present invention.

[0020] In order to better understand the present invention, the above content will be described in detail below in combination with specific embodiments.

[0021] Embodiment 1: Please refer to the attached Figure 1 The present invention provides a safety warning method for working in a confined space, including the following steps: S1. Continuously collect the air quality, temperature and humidity information in the confined space, and evaluate the suitability of the space environment according to preset standards; in this embodiment, the air quality in the confined space, such as oxygen concentration, carbon dioxide concentration, toxic gas content and temperature and humidity information, is collected in real time through sensors to form dynamic monitoring of environmental data. Combining preset standards, such as gas concentration ranges and temperature and humidity thresholds, the collected data is analyzed and evaluated to determine whether the current environment meets the standards and requirements for normal operation. When it does not meet the standards, an alarm is issued, so as to ensure timely discovery of potential environmental problems and ensure the safety of the working environment.

[0022] S2. According to the evaluation result of the suitability of the space environment, when a potential danger is detected, automatically start the ventilation system to adjust the indoor environment, and at the same time send a warning signal to relevant personnel; In this embodiment, in the analysis and evaluation result of the air in the confined space, when any one of the oxygen concentration is low and the toxic gas exceeds the standard, the ventilation system is triggered for automatic adjustment. The ventilation system optimizes the air quality and reduces risk factors by dynamically adjusting the operation of the exhaust and intake devices. At the same time, it is also responsible for sending warning signals to relevant personnel to ensure that relevant personnel are informed of the current dangerous state in a timely manner, so as to make necessary manual intervention or take further measures.

[0023] S3. Continuously detect the status of the access control system at the entrance of the confined space to ensure that the access control is in a continuously closed state and ensure the isolation of the confined space; In this embodiment, by monitoring the opening and closing state of the access control in real time, the isolation of the enclosed space is ensured. Its main function is to prevent external pollutants (such as dust and toxic gases) from entering the enclosed space, thereby interfering with the stability of the working environment. In addition, the status monitoring of the access control system also ensures the accuracy of the environmental monitoring data and avoids accidental impacts caused by the accidental opening of the door. By ensuring the continuous closed state of the access control, the safety and controllability of the working environment are further enhanced.

[0024] S4. Continuously collect and analyze the body movement data of the operator, and send a warning signal to relevant personnel when the operator is in a stationary state for a long time.

[0025] In this embodiment, the movement data is continuously collected by a three-axis acceleration sensor and a gyroscope, and the movement state of the operator (such as stationary, normal movement or abnormal movement) can be analyzed. When the system detects that the operator is in a stationary state for a long time (possibly due to an accident or sudden situation), a warning signal will be sent to relevant personnel to attract attention, so as to start the rescue procedure in time. To avoid false alarms, at the same time, this step combines a deep learning algorithm to accurately analyze the movement pattern, which can distinguish between short-term stillness (such as rest) during normal work and long-term stillness in dangerous situations, improving the accuracy and efficiency of early warning.

[0026] Embodiment 2: Please refer to the attached Figure 2 , the present invention provides a system for implementing the above method, including: An environmental monitoring module, which is used to collect the air quality, temperature and humidity information in the enclosed space, and evaluate the suitability of the space environment according to the preset standards; A ventilation system, according to the evaluation result of the space environment suitability, when a potential danger is detected, automatically starts the ventilation system to adjust the indoor environment, and sends a warning signal to relevant personnel; An access control system for the entrance of the enclosed space, which is used to detect the status of the access control in real time, and ensure that the access control is continuously in a closed state to ensure the isolation of the enclosed space; A human movement monitoring module, which is used to continuously collect and analyze the body movement data of the operator, identify whether the operator has a situation of staying still for a long time, and send a warning signal to relevant personnel.

[0027] The environmental monitoring module includes a gas sensor and a temperature and humidity sensor, which are used to monitor the oxygen concentration, carbon dioxide concentration, toxic gas content and temperature and humidity parameters.

[0028] In this embodiment, the environmental monitoring module provides an accurate basis for the assessment of environmental suitability by collecting environmental data in the enclosed space in real time, and at the same time provides basic support for the linkage response of the ventilation system and other modules. Generally, this module includes gas sensors and temperature and humidity sensors, which are used to detect key environmental parameters such as oxygen concentration, carbon dioxide concentration, toxic gas content, as well as temperature and humidity. The dynamic collection and processing of these data are the prerequisite for ensuring the safety of the enclosed space environment.

[0029] In a possible implementation, the present invention arranges gas sensors and temperature and humidity sensors, and combines an edge computing unit to perform local processing on the collected data to ensure the real-time and accuracy of the monitoring data. This module can not only operate independently, but also be linked with the human motion monitoring module and the ventilation system to further improve the intelligent level of early warning.

[0030] First of all, the gas sensors are configured to continuously monitor the oxygen concentration, carbon dioxide concentration, and the content of specific toxic gases. Through a preset standard range, this module can judge whether the current environment is suitable for personnel operation. For example, in some embodiments, the safe range of oxygen concentration is set to 19.5% to 23.5%. When the monitored value is lower or higher than this range, the module will trigger a danger signal. In addition, the concentration thresholds of toxic gases (such as carbon monoxide, hydrogen sulfide, etc.) can be set according to different operation scenarios, and exceeding the threshold will directly trigger the alarm mechanism.

[0031] Secondly, the temperature and humidity sensors are used to monitor the temperature and humidity parameters in the space. In a possible implementation, this module can accurately measure the temperature range from -20°C to 50°C and the humidity range from 10% to 90% RH. By adjusting the air intake or exhaust rate of the ventilation system in real time, the environmental monitoring module can ensure that the temperature and humidity parameters are always within the set range. Therefore, by combining the air and temperature and humidity data, the data of the space suitability can be obtained, and the calculation of the environmental suitability is comprehensively evaluated in combination with a weighted model. Specifically, the environmental suitability A is calculated by the following formula: where Q 氧气 is the current oxygen concentration, Q 氧气,标准 is the standard value of oxygen concentration, generally taking 21%, Q 二氧化碳 is the current carbon dioxide concentration, Q 二氧化碳,标准 is the standard value of carbon dioxide concentration, generally taking 0.04%, C 毒气 is the concentration of toxic gas (such as carbon monoxide concentration), C 毒气,标准 is the maximum allowable value of the concentration of toxic gas, set according to the specific scenario, T is the current temperature, T 标准 is the ideal temperature, H is the current humidity, H 标准 is the ideal humidity, w1 , w 2 , w 3 , w 4 , w 5 are weight parameters respectively, reflecting the importance of different environmental factors; In some embodiments, this module combines with an edge computing unit for real-time data processing, which specifically includes the following steps: Data preprocessing: Before the collected environmental data enters the suitability evaluation formula, it needs to be denoised and normalized. Wavelet denoising technology is used for data denoising, and its formula is as follows: Among them, is the denoised signal; W k (t) is the wavelet coefficient; φ k (t) is the wavelet basis function; n is the decomposition level; The normalization process is carried out according to the following formula: Among them, X is the original data; max(X) and min(X) are the maximum and minimum values of the data; X ′ is the normalized data, which is used to input the suitability evaluation formula.

[0032] Real-time analysis: Generally, when the calculation result of A is lower than the preset suitability threshold, the system will trigger linkage control, including adjusting the wind speed of the ventilation system and starting the alarm module.

[0033] And the environmental monitoring module not only realizes real-time monitoring of environmental parameters and suitability evaluation during confined space operations, but also further sets up a multi-level early warning mechanism to adopt differential linkage control strategies for different risk levels. Through the multi-level early warning mechanism, false alarms of the system can be effectively avoided, and at the same time, the allocation and use of resources can be optimized.

[0034] Specifically: The environmental monitoring module calculates the suitability A of the current environment by collecting gas concentration, temperature and humidity data in real time. Based on the numerical range of A, the environment is divided into three levels: "safe", "warning" and "emergency", and the corresponding system linkage operations are triggered.

[0035] Safe level: When A ≥ 0.8, the environment is in a safe state, and the system does not need to trigger any alarms or linkage controls, and only needs to maintain the normal monitoring state. At this time: The ventilation system maintains the lowest operating rate, which is only used for conventional air circulation; The data collection frequency can be reduced to once every 2 minutes to save system resources; The alarm module is in a silent state and only records the changing trends of environmental parameters.

[0036] Early warning level: When 0.5 ≤ A < 0.8, the environment enters the early warning level, and the system will trigger a primary linkage response to prevent the environment from deteriorating further. Specifically: The ventilation system starts a medium-intensity exhaust and intake mode to adjust air quality, temperature, and humidity; The alarm module sends early warning signals to operators and the monitoring center, such as a yellow alarm; The data collection frequency is increased to once every 0.5 minutes, and the edge computing module analyzes the changing trends of data in real time; In some embodiments, historical data trends can be combined to predict future environmental changes. For example, if the oxygen concentration shows a downward trend, the ventilation system will pre-elevate the exhaust rate.

[0037] The purpose of the early warning level is to prevent environmental parameters from further dropping to a dangerous state through early intervention.

[0038] Emergency level: When A < 0.5, the environment enters an emergency state, and immediate comprehensive linkage response measures are required to ensure the safety of operators. Specifically: The ventilation system starts the highest-intensity mode to comprehensively optimize air quality; The alarm module sends a red emergency alarm signal to operators, the monitoring center, and the rescue team, and attaches environmental parameter data and location information; In some embodiments, the emergency alarm also triggers the flashing of lights or the ringing of sirens in the operation area to ensure that all personnel receive warnings in a timely manner; The data collection frequency is increased to once per second, and all sensors transmit data to the monitoring center in real time; The human motion monitoring module is linked to the environmental monitoring module to determine whether the personnel status is safe. For example, if it is detected that an operator remains stationary for a long time while the environmental parameters are at the emergency level, the system will prioritize starting the rescue procedure; The ventilation system further combines the output results of the environmental monitoring module to dynamically adjust the wind speed. The calculation formula is as follows: F = k·(Q 目标 -Q 当前 ) where F is the wind speed; Q 目标 is the air quality target value; Q 当前 is the current air quality value; k is a proportionality coefficient. At this time, the system dynamically adjusts the wind speed at the second level to restore the air quality to the safe range at the fastest speed.

[0039] In some embodiments, the multi-level warning mechanism of the environmental monitoring module further includes the following extended functions: Multi-level linkage: In the event of a warning and an emergency, the environmental monitoring module can link other modules according to the specific risk level. For example: At the warning level, the operating personnel are notified first to remind them to pay attention to environmental changes; At the emergency level, the access control system is directly linked to prevent other personnel from entering the confined space, and at the same time, the area warning mode is activated.

[0040] Risk trend prediction: By analyzing the historical environmental data trends in real time through the edge computing module, the changes in environmental suitability are predicted. For example, the system can estimate in advance the value that the oxygen concentration may reach within the next 10 minutes and trigger an early response according to the prediction results.

[0041] Sub-region monitoring and warning: In some large confined spaces, the environmental monitoring module can achieve multi-level warning for different regions. For example, when the oxygen concentration in a certain region is abnormal while that in other regions is normal, only the warning linkage of that region is triggered to avoid wasting resources due to a global alarm.

[0042] Therefore, through the multi-level warning and linkage mechanism of the environmental monitoring module, not only the real-time and accuracy of environmental monitoring are achieved, but also efficient responses can be made according to different risk levels, fully ensuring the safety of operating personnel and optimizing the allocation and use of system resources at the same time. The collaborative work of this module with the ventilation system, the human motion monitoring module, and the edge computing module further improves the intelligent level of the safety warning system for confined space operations.

[0043] The human motion monitoring module includes: A three-axis acceleration sensor and a gyroscope for collecting the motion data of operating personnel; A deep learning model training unit that analyzes and iteratively optimizes the motion patterns of operating personnel based on the historical data sets of the three-axis acceleration sensor and the gyroscope, for analyzing crawling, bending, standing, and normal working states and long-term stationary states, to avoid misjudging necessary working postures as potential risks.

[0044] The deep learning model is trained based on a multi-layer neural network. By inputting the data of the three-axis acceleration sensor and the gyroscope, a classification model is established, and the motion state categories of operating personnel are output, including normal motion, micro-motion, and stationary state.

[0045] In this embodiment, the human motion monitoring module is used to collect and analyze the motion data of the operator, identify their motion state, and trigger an alarm signal when necessary. It includes a three-axis acceleration sensor and a gyroscope for collecting the motion data of the operator. At the same time, combined with the deep learning model training unit, the motion patterns of the operator are analyzed and iteratively optimized to avoid misjudging necessary working postures (such as crawling or bending) as dangerous states. Therefore, it can not only accurately identify the stationary, normal motion, and abnormal states of the operator, but also achieve efficient alarm linkage and response control.

[0046] And the data is realized through the cooperation of the edge computing unit, providing the classification result of the motion state in real time. Specifically: The three-axis acceleration sensor is used to collect the three-dimensional acceleration data of the operator in the three-dimensional space. At the same time, the gyroscope is used to capture the three-dimensional angular velocity data to more comprehensively reflect the motion state of the operator. In some embodiments, the sampling frequency of the three-axis acceleration sensor is set to ensure that subtle motion changes can be accurately captured. For example, when the operator is crawling or bending, the sensor can record low-frequency and small-amplitude motion data, while when standing still, the collected acceleration is close to zero. The gyroscope is used to monitor the change of angular velocity to determine the body posture of the operator, such as turning around, bending, or tilting.

[0047] And after these data are preprocessed, they will be used as the input of the deep learning model. To improve the applicability of the sensing data, the data preprocessing includes two parts: denoising and normalization. In the denoising process, wavelet transform technology is used, and then through normalization processing, the data preprocessing is realized; The implementation method of the deep learning model training unit is as follows: The deep learning model uses a multi-layer neural network, combines the three-axis acceleration and gyroscope data, and establishes a classification model of the motion state. The model structure includes an input layer, a hidden layer, and an output layer, which is described by the following formula: y = f(W·X + b) Where, y is the output result of the model, used to represent the motion state category; W is the neural network weight matrix; X is the input vector of the preprocessed sensor data, b is the bias vector; f is the activation function, In the classification stage, the output layer converts the model output into the probability distribution of the motion state through the Softmax function: Where, P(S t = i|X t ) represents the probability that the operator is in state i at time t; z i is the linear transformation result of the hidden layer; n is the number of classification state categories (such as stationary, normal motion, micro motion, etc.), S tis the motion state of the operator at time t, is the sum of the activation value indices of all state categories.

[0048] Specifically, the output categories of the model include the following states: Normal motion: When the operator is walking or performing other continuous movements, the motion data characteristics show regular fluctuations, and the model outputs this category.

[0049] Minor movements: When the operator is adjusting his posture or is temporarily still, such as lowering his head or raising his hand, the model can accurately classify this state to avoid misjudgment as stillness.

[0050] Static state: When the acceleration and angular velocity are close to zero and the static time reaches a certain threshold, the model outputs a static state.

[0051] In order to distinguish long-term stillness from short-term stillness during normal work, the model further introduces the calculation formula for the cumulative stillness time: Among them, T s Indicates the cumulative static time; P(S t = stationary) is the probability of being in a stationary state; P threshold is the judgment threshold of the static state; H(x) is a step function, which takes 1 when x>0 and 0 otherwise. 1 is the starting time of monitoring; t 2 is the end time of monitoring, dt is the integral variable, representing a small increment in time Therefore, when T s Exceeding the set threshold P threshold , the system will trigger an alarm signal and link the environmental monitoring module and ventilation system to ensure personnel safety.

[0052] This formula is obtained by 1 ,t 2 ] in the stationary state probability P(S t = static), analyze, combine the threshold P threshold The step function H accumulates the time when the operator is in a stationary state. When the stationary time T s If it exceeds a certain set value, the system will consider that the operator may be in a dangerous state and trigger the corresponding alarm.

[0053] For example: Assume t 1 =0,t 2 =10, the probability of static state P(S t =Stationary) are shown in the following table: (t) 0 0.3 0.5 -0.2 0 1 0.6 0.5 0.1 1 2 0.7 0.5 0.2 1 3 0.4 0.5 -0.1 0 4 0.8 0.5 0.3 1 5-10 0.3 0.5 -0.2 0 According to the above table, the cumulative stationary time is the time periods of time intervals 1s, 2s, and 4s. Therefore: 1 + 1 + 1 = 3s If P is set threshold to 3s, then the alarm trigger condition is exactly reached at this time.

[0054] In some embodiments, the deep learning model training unit further has the following extended functions: The training data set of the model not only includes the historically collected motion data, but can also be updated online through the edge computing unit to adapt to the specific requirements of different operation scenarios.

[0055] In a possible implementation manner, in order to cope with different working environments (such as high temperature or airtight environment), the model can adjust the feature weights to enhance the classification accuracy.

[0056] As an option, the system can also compare the real-time motion data with the historical data to detect whether there are abnormal behaviors of the operating personnel, such as long-term stillness or abnormal acceleration.

[0057] Through the combination of hardware and deep learning algorithms, the human motion monitoring module can not only accurately judge the motion state of the operating personnel, but also form a linkage mechanism with the environmental monitoring module and the ventilation system to achieve intelligent real-time monitoring and early warning functions. This design improves the response efficiency and reliability while ensuring the monitoring accuracy.

[0058] The human motion monitoring module further includes an alarm trigger module, which is used to automatically send an alarm signal to the monitoring center according to the analysis result of the deep learning model training unit and in combination with the determination of the motion state of the operating personnel, and is linked with the environmental monitoring module and the ventilation system to adjust the operating state of the ventilation system.

[0059] When the human motion monitoring module identifies that the operating personnel are stationary for a long time and the cumulative time of the stationary state exceeds the set threshold, an alarm signal is triggered and the rescue procedure is started.

[0060] In this embodiment, the alarm trigger module is an important part of the human motion monitoring module, mainly used to automatically trigger an alarm signal and link other modules to respond when identifying the state that the operating personnel are stationary for a long time. The module can timely discover potential dangers and quickly take corresponding measures through continuous analysis of the motion state of the operating personnel and in combination with the judgment of the cumulative stationary time; Specifically, when the human body movement monitoring module recognizes that the operator is in a stationary state and the duration of the stationary state exceeds a preset safety threshold, the alarm triggering module will immediately send an alarm signal to the monitoring center to alert the operator of possible abnormal situations. Meanwhile, this module can be linked with the environmental monitoring module and the ventilation system to dynamically adjust the parameters of the working environment, such as enhancing air circulation or activating the emergency ventilation mode, to reduce the impact of the dangerous environment on the operator.

[0061] In some embodiments, the alarm triggering module supports a multi-level early warning function, dividing different alarm levels according to the risk degree. For example, when the duration of the stationary state approaches the set safety threshold, the system can trigger a first-level alarm to alert the operator and the monitoring center of possible dangerous situations. If the stationary time exceeds the safety threshold, an emergency alarm is triggered to notify the relevant parties to take immediate action. At the same time, this module can be linked with other modules, such as sending a signal to the ventilation system to increase the air circulation intensity, or collaborating with the access control system to prevent other people from entering the dangerous area.

[0062] Specifically: The alarm triggering module combines the real-time data of the environmental monitoring module to further optimize the accuracy of the alarm. If the air quality assessment of the environmental monitoring module shows that the current environmental parameters exceed the safety range, such as too low oxygen concentration or excessive toxic gas concentration, the alarm triggering module will raise the alarm level and accelerate the response speed. For example, in the case of linkage with the ventilation system, the system can dynamically adjust the wind speed and direction according to the air quality parameters to quickly restore the environment to a safe level.

[0063] As a possible implementation, the alarm triggering module is combined with the edge computing unit to complete the analysis of the stationary state and the generation of alarms locally, thereby reducing the delay of data transmission and improving the response efficiency. In the case of network communication interruption, this module can still operate independently to ensure the safety of the operator. For example, the edge computing unit can predict the future state changes of the operator through historical data and initiate early warning when necessary.

[0064] In some embodiments, the alarm triggering module can also generate a more accurate alarm signal by combining the location information of the operator. For example, when the monitoring center receives an alarm, the module can synchronously transmit the location information of the operator, the current environmental data, and the historical movement state to help the rescue team quickly locate the person and take appropriate rescue measures. In this way, the module effectively improves the rescue efficiency and reduces possible misjudgments or delays.

[0065] In addition, this module supports adjusting the alarm strategy according to the requirements of different operation scenarios. For example, in a low-risk environment, the system can reduce unnecessary alarm times by extending the safety threshold time; while in a high-risk environment, the system can shorten the threshold time and generate higher-frequency alarm signals by combining the analysis results of other modules. This flexible design enables the module to adapt to a variety of complex operation environments and improves the reliability of the system at the same time.

[0066] Therefore, through the alarm trigger module, dangerous states can be identified and alarms can be issued in a timely and accurate manner. At the same time, through cooperation with other modules, intelligent risk response and environmental regulation can be achieved. This module design not only improves the system safety but also enhances its adaptability and application value in complex operation environments.

[0067] The ventilation system includes an exhaust device and an intake device, and adjusts the wind speed in real time according to the output results of the environmental monitoring module and the human motion monitoring module to optimize the air quality.

[0068] In this embodiment, the ventilation system is used to adjust the air quality in the enclosed space in real time to ensure that the operation environment is always in a safe state. The specific process is as follows: The ventilation system receives the output data of the environmental monitoring module, dynamically adjusts the air quality, removes harmful gases and introduces fresh air. When the air quality in the operation environment exceeds the preset safety range, the ventilation system will be immediately activated to adjust the intensity and direction of exhaust and intake, so as to optimize the operation environment. And the system combines the output results of the human motion monitoring module to achieve intelligent linkage response.

[0069] Among them, when the environmental monitoring module detects that the oxygen concentration in the air is lower than the set value, or the concentration of toxic gases is higher than the preset threshold, the ventilation system will automatically increase the operating intensity of the exhaust and intake devices to quickly improve the environmental conditions. At the same time, when the human motion monitoring module identifies that the operator is in a long-term stationary state and may have abnormal physical conditions due to the deterioration of air quality, the ventilation system can further optimize the operation according to the comprehensive data. For example, in an emergency, the system can give priority to increasing the exhaust speed to quickly remove dangerous gases, and at the same time control the intake device to introduce an appropriate amount of fresh air to restore the oxygen concentration.

[0070] At the same time, the ventilation system dynamically adjusts the operating state by calculating the wind speed and the air quality target value. For example, when it detects insufficient oxygen concentration, the system will adjust the ratio of exhaust and intake in real time according to the difference between the current air quality and the target value. The ventilation system can also combine the changing trend of environmental parameters, such as the gradual increase in carbon dioxide concentration, and increase the exhaust intensity in advance to avoid the environment entering a dangerous state.

[0071] As an option, the ventilation system can also operate in conjunction with the results of the human motion monitoring module. For example, when the operator's motion state is monitored to be "stationary" and the ambient air quality is close to the danger threshold, the system will give priority to increasing the supply rate of fresh air and trigger the environmental monitoring module to perform high-frequency sampling of air quality to ensure the stability of environmental parameters and that the exhaust device and the air intake device are divided into two parts. The exhaust device is mainly responsible for discharging waste gas or toxic gas from the confined space, while the air intake device is responsible for introducing fresh air to maintain the oxygen concentration. In some embodiments, the ventilation system can realize independent regulation of different regions according to the environmental parameters of different regions. Assume that when the carbon dioxide concentration is detected to exceed the standard in a certain area, the system can only perform exhaust operations on the area without affecting the normal operation of other areas. Therefore, the energy saving and operation efficiency of the system are effectively improved.

[0072] Finally, the operation logic of the ventilation system can be further optimized. For example, when the environmental monitoring module and the human motion monitoring module output emergency signals at the same time, the ventilation system can enter the highest intensity operation mode to ensure that the air quality returns to a safe range in the shortest time. In addition, the system can also combine the alarm trigger module to send the current ventilation status and the location information of the operator to the monitoring center so that the rescue personnel can grasp the environmental conditions in real time.

[0073] The system further includes an edge computing module, which is used to perform localized analysis of environmental monitoring data and human motion monitoring data.

[0074] Embodiment 3: The components of the above-mentioned confined space operation safety warning system include: Environmental monitoring components, which are used to collect information on air quality, temperature and humidity in a confined space, and evaluate the suitability of the space environment according to preset standards; Ventilation components, including exhaust devices and air intake devices, are used to automatically start the ventilation system to adjust the indoor environment and send warning signals to relevant personnel when potential dangers are detected based on the spatial environment suitability assessment results; The confined space entrance access control component is used to detect the status of the access control in real time to ensure that the access control is continuously closed to ensure the isolation of the confined space; The human motion monitoring component is used to continuously collect and analyze the body motion data of the operators, identify whether the operators have been stationary for a long time, and send warning signals to relevant personnel.

[0075] The components of this embodiment can be used to execute the above-mentioned method and system embodiments. The principles and technical effects are similar and will not be repeated here.

[0076] Although embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A confined space operation safety early warning method, characterized in that: The following steps are involved: S1. Continuously collect air quality, temperature and humidity information in the confined space, and evaluate the suitability of the space environment according to preset standards; S2. Based on the evaluation results of the space environment suitability, when potential danger is detected, the ventilation system is automatically activated to adjust the indoor environment and a warning signal is sent to relevant personnel; S3. Continuously monitor the status of the access control system at the entrance of the confined space to ensure that the access control is in a continuously closed state and to ensure the isolation of the confined space; S4. Continuously collect and analyze the body movement data of operators, and send warning signals to relevant personnel when operators are in a static state for a long time.

2. A confined space operation safety warning system, based on the confined space operation safety warning method according to claim 1, characterized in that: include: Environmental monitoring module, which is used to collect air quality, temperature and humidity information in the confined space, and evaluate the suitability of the space environment according to preset standards; The ventilation system automatically starts the ventilation system to adjust the indoor environment and send a warning signal to relevant personnel when potential danger is detected based on the spatial environment suitability assessment results; The confined space entrance access control system is used to detect the status of the access control in real time and ensure that the access control is continuously closed to ensure the isolation of the confined space; The human motion monitoring module is used to continuously collect and analyze the body motion data of the operators, identify whether the operators have been stationary for a long time, and send warning signals to relevant personnel.

3. The confined space operation safety warning system according to claim 2 is characterized in that: The environmental monitoring module includes a gas sensor and a temperature and humidity sensor for monitoring oxygen concentration, carbon dioxide concentration, toxic gas content, and temperature and humidity parameters.

4. The confined space operation safety warning system according to claim 2, characterized in that: The human motion monitoring module comprises: Three-axis accelerometer and gyroscope, used to collect the operator's motion data; The deep learning model training unit analyzes and iteratively optimizes the operator's movement patterns based on historical data sets of three-axis accelerometers and gyroscopes. It is used to analyze crawling, bending, standing, normal working conditions and long-term static conditions to avoid judging necessary working postures as potential dangers.

5. The confined space operation safety warning system according to claim 4, characterized in that: The deep learning model is trained based on a multi-layer neural network. A classification model is established by inputting data from a three-axis acceleration sensor and a gyroscope to output the motion state categories of the operator, including normal motion, small motion, and static state.

6. The confined space operation safety warning system according to claim 5, characterized in that: The human motion monitoring module further includes an alarm triggering module, which is used to automatically send an alarm signal to the monitoring center based on the analysis results of the deep learning model training unit and the motion status of the operator, and to cooperate with the environmental monitoring module and the ventilation system to adjust the operating status of the ventilation system.

7. The confined space operation safety warning system according to claim 6, characterized in that: When the human motion monitoring module identifies that the operator has been stationary for a long time and the cumulative time of the stationary state exceeds the set threshold, an alarm signal is triggered and the rescue procedure is started.

8. The confined space operation safety warning system according to claim 7, characterized in that: The ventilation system includes an exhaust device and an air intake device, and adjusts the wind speed in real time to optimize the air quality according to the output results of the environment monitoring module and then the human movement monitoring module.

9. The confined space operation safety warning system according to claim 2, characterized in that: The system further includes an edge computing module, which is used to perform localized analysis on environmental monitoring data and human motion monitoring data.

10. A confined space operation safety warning component, based on the confined space operation safety warning system according to any one of claims 2 to 9, characterized in that: include: Environmental monitoring components, which are used to collect information on air quality, temperature and humidity in a confined space, and evaluate the suitability of the space environment according to preset standards; Ventilation components, including exhaust devices and air intake devices, are used to automatically start the ventilation system to adjust the indoor environment and send warning signals to relevant personnel when potential dangers are detected based on the spatial environment suitability assessment results; The confined space entrance access control component is used to detect the status of the access control in real time to ensure that the access control is continuously closed to ensure the isolation of the confined space; The human motion monitoring component is used to continuously collect and analyze the body motion data of the operators, identify whether the operators have been stationary for a long time, and send warning signals to relevant personnel.

Images