Respirator self-adaptive control method, system and equipment and storage medium

Abstract

The invention provides a respirator self-adaptive control method, system and equipment and a storage medium, and relates to the field of respirator control. The method comprises the steps that physiological parameters and environmental parameters of a respirator wearer and initial running state data of the respirator are obtained, man-machine state evaluation is conducted based on the physiological parameters and the environmental parameters, and a man-machine state evaluation result is determined; determining an initial control strategy of the respirator based on the man-machine state evaluation result and the initial operation state data of the respirator; acquiring historical data in the use process of the respirator, and adaptively adjusting the initial control strategy based on the historical data to obtain an optimized dynamic control strategy; multi-source sensing data of the respirator are obtained in real time, the dynamic control strategy is adjusted in real time based on the multi-source sensing data, and a self-adaptive control strategy is obtained; and the respirator is controlled based on the self-adaptive control strategy. According to the scheme, the wearing comfort and the protection effect of the respirator can be improved.

Description

Technical Field

[0001] The present application relates to the field of ventilator control, and in particular to a ventilator adaptive control method, system, device and storage medium. Background Art

[0002] Respirators are a commonly used personal protective equipment, widely used in the fields of industry, medical treatment, epidemic prevention, etc. Traditional respirators usually adopt a fixed control strategy, that is, air supply, filtration and protection control are performed according to the set parameters. However, in actual use, the wearer's physiological state, the surrounding environment and the operating status of the equipment are all dynamically changing. Fixed control strategies are difficult to adapt to these changes, resulting in the respirator's wearing comfort and protective effect cannot be guaranteed. Summary of the invention

[0003] The present application provides a respirator adaptive control method, system, device and storage medium, which can improve the wearing comfort and protective effect of the respirator.

[0004] In a first aspect, the present application provides a method for adaptive control of a ventilator, the method comprising: Acquire physiological parameters and environmental parameters of the respirator wearer and initial operating status data of the respirator, and perform a human-machine status assessment based on the physiological parameters and environmental parameters to determine a human-machine status assessment result; Determine the initial control strategy of the respirator based on the human-machine status assessment results and the initial operating status data of the respirator; Acquire historical data during the use of the respirator, the historical data including historical physiological feedback data and respirator operation status data, and adaptively adjust the initial control strategy based on the historical data to obtain an optimized dynamic control strategy; Acquire multi-source sensor data of the ventilator in real time, and adjust the dynamic control strategy in real time based on the multi-source sensor data to obtain an adaptive control strategy; The ventilator is controlled based on an adaptive control strategy.

[0005] By adopting the above technical solution, by obtaining the physiological parameters, environmental parameters and initial operating status data of the respirator wearer, and performing human-machine status assessment based on the physiological parameters and environmental parameters, and determining the human-machine status assessment results, it is possible to fully grasp the physiological needs of the wearer and the risk level of the environment they are in. On this basis, combined with the initial operating status data of the respirator, the initial control strategy of the respirator is determined, including the initial air supply volume, air supply frequency, filtration level and safety control strategy, etc., to achieve the initial parameter setting of the respirator, ensuring that the respirator can provide reasonable initial protection measures according to the actual needs of the wearer and environmental risks.

[0006] In order to further improve the intelligence and adaptability of the control strategy, the present invention also obtains historical data during the use of the respirator, including historical physiological feedback data and respirator operating status data. By analyzing these historical data, the association rules between the wearer's historical physiological feedback data and the respirator control parameters are mined, and combined with the reinforcement learning algorithm, the initial control strategy is adaptively adjusted with the improvement of wearing comfort and protection safety as the optimization goal to obtain an optimized dynamic control strategy. This strategy optimization method based on historical data can fully consider the wearer's actual use experience and the operating performance of the respirator, make targeted improvements to the initial strategy, and generate a more intelligent and personalized control strategy.

[0007] During the actual use of the respirator, the wearer's physiological state, environmental risks, and equipment operating status may all change in real time. In order to cope with these changes, the present invention also obtains multi-source sensor data of the respirator in real time, including real-time physiological parameters, real-time environmental parameters, and real-time equipment status parameters. By fusion analysis of these multi-source data, a state perception matrix integrating man-machine-environment is constructed, and it is input into a preset strategy optimization model. The dynamic control strategy is corrected in real time using adaptive weight adjustment and multi-objective optimization algorithms to generate the optimal adaptive control strategy. This control method based on real-time multi-source perception and strategy adaptive optimization can dynamically track the state changes of man, machine, and environment, and adjust the control strategy in time to ensure that the wearer is always provided with the best protection effect in complex and changeable actual use scenarios.

[0008] Finally, according to the generated adaptive control strategy, the present invention utilizes technical means such as PID control, filter wind resistance adjustment, and safety warning to accurately control the air supply system, filter system, and safety protection system of the respirator, ensuring that control instructions can be accurately and efficiently executed, truly realizing real-time adaptive control of the respirator, and creating a safe, comfortable, and intelligent protection environment for the wearer.

[0009] In a second aspect of the present application, a ventilator adaptive control method system is provided, comprising: A data acquisition module, used to acquire physiological parameters and environmental parameters of the respirator wearer and initial operating state data of the respirator, and to perform a human-machine state assessment based on the physiological parameters and environmental parameters, and determine a human-machine state assessment result; An initial strategy generation module, used for determining an initial control strategy of the respirator based on the human-machine status evaluation result and the initial operation status data of the respirator; An optimization module is used to obtain historical data during the use of the respirator, the historical data including historical physiological feedback data and respirator operation status data, and adaptively adjust the initial control strategy based on the historical data to obtain an optimized dynamic control strategy; An adjustment module is used to obtain multi-source sensor data of the respirator in real time, and adjust the dynamic control strategy in real time based on the multi-source sensor data to obtain an adaptive control strategy; The control module is used for controlling the respirator based on an adaptive control strategy.

[0010] In a third aspect of the present application, a computer storage medium is provided, wherein the computer storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the above method steps.

[0011] In the fourth aspect of the present application, an electronic device is provided, including a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes the above method.

[0012] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. This application obtains the physiological parameters, environmental parameters and initial operating status data of the respirator wearer, and performs a human-machine status assessment based on the physiological parameters and environmental parameters to determine the human-machine status assessment results, so as to fully grasp the physiological needs of the wearer and the risk level of the environment. On this basis, combined with the initial operating status data of the respirator, the initial control strategy of the respirator is determined, including the initial air supply volume, air supply frequency, filtration level and safety control strategy, etc., to achieve the initial parameter setting of the respirator, ensuring that the respirator can provide reasonable initial protection measures according to the actual needs of the wearer and environmental risks.

[0013] 2. In order to further improve the intelligence and adaptability of the control strategy, the present invention also obtains historical data during the use of the respirator, including historical physiological feedback data and respirator operating status data. By analyzing these historical data, the association rules between the wearer's subjective feelings and the respirator control parameters are excavated, and combined with the reinforcement learning algorithm, the initial control strategy is adaptively adjusted with the improvement of wearing comfort and protection safety as the optimization goals to obtain an optimized dynamic control strategy. This strategy optimization method based on historical data can fully consider the wearer's actual use experience and the operating performance of the respirator, make targeted improvements to the initial strategy, and generate a more intelligent and personalized control strategy.

[0014] 3. In the actual use of the respirator of this application, the wearer's physiological state, environmental risks, and equipment operating status may all change in real time. In order to cope with these changes, the present invention also obtains multi-source sensor data of the respirator in real time, including real-time physiological parameters, real-time environmental parameters, and real-time equipment status parameters. By fusion analysis of these multi-source data, a state perception matrix of man-machine-environment integration is constructed, and it is input into a preset strategy optimization model. The dynamic control strategy is corrected in real time using adaptive weight adjustment and multi-objective optimization algorithms to generate the optimal adaptive control strategy. This control method based on real-time multi-source perception and strategy adaptive optimization can dynamically track the state changes of man, machine, and environment, and adjust the control strategy in time to ensure that the wearer is always provided with the best protection effect in complex and changeable actual use scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A schematic flow chart of a ventilator adaptive control method provided in an embodiment of the present application; Figure 2 An architecture diagram of a ventilator adaptive control system provided in an embodiment of the present application; Figure 3 A schematic diagram of the structure of an electronic device provided in this application. DETAILED DESCRIPTION

[0016] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0017] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.

[0018] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0019] In order to facilitate understanding of the method and system provided by the embodiments of the present application, before introducing the embodiments of the present application, the background of the embodiments of the present application is first introduced.

[0020] Respirators are a commonly used personal protective equipment, widely used in the fields of industry, medical treatment, and epidemic prevention. The main function of a respirator is to provide the wearer with clean breathing air and block harmful substances in the environment to ensure the health and safety of the wearer. However, traditional respirators usually adopt a fixed control strategy, that is, air supply, filtration, and protection control are performed according to preset parameters. This "one-size-fits-all" control method has obvious limitations and is difficult to adapt to the dynamic changes in personnel status, environmental conditions, and equipment performance during actual use, resulting in a significant reduction in the wearing comfort and protective effect of the respirator.

[0021] Specifically, the fixed control strategy of traditional respirators has the following main shortcomings: First, it fails to fully consider the changes in the wearer's physiological state. In actual use, the wearer's breathing needs will change dynamically due to factors such as their activity intensity and physical condition. However, traditional respirators cannot automatically adjust the air supply parameters according to these changes, resulting in the air supply volume being difficult to match the actual needs. When the air supply is insufficient, the wearer will feel stuffy and uncomfortable; when the air supply is excessive, it will cause energy waste and noise pollution. Second, it fails to respond to changes in environmental risks in a timely manner. The risk factors in industrial sites and epidemic areas are often dynamic, and the concentration of harmful substances may fluctuate with time and location. However, traditional respirators use fixed filtration levels, which are difficult to adapt to such risk changes, may lead to insufficient or excessive protection, which not only affects the wearing comfort, but also reduces the service life of the respirator. Third, there is a lack of effective monitoring and early warning of the equipment's operating status. Traditional respirators are rarely equipped with real-time monitoring and fault diagnosis functions, and are unable to detect and solve potential equipment problems in a timely manner, posing safety hazards. At the same time, due to the lack of intelligent control means, even if problems are found, it is difficult to give targeted control instructions according to actual needs.

[0022] In general, the fixed control strategy of traditional respirators can no longer meet the increasingly complex protection needs. How to achieve an adaptive respirator control method that can automatically adjust the control strategy according to the wearer's physiological state, environmental risks, and dynamic changes in the equipment's operating status to improve wearing comfort and protection safety has become a key technical problem that needs to be broken through in the field of respiratory protection. This is not only related to the physical and mental health of the majority of respirator users, but also affects the safe production and sustainable development of related industries.

[0023] After the background introduction of the above content, those skilled in the art can understand the problems existing in the prior art. The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments.

[0024] Based on the above background technology, please refer to Figure 1 , Figure 1 A flow chart of a ventilator adaptive control method provided in an embodiment of the present application. The system can be implemented by a computer program or can be run as an independent tool application. Specifically, in an embodiment of the present application, the method can be applied on a server, but can also be applied on electronic devices such as a server. A ventilator adaptive control method includes the following steps: S101, obtaining physiological parameters and environmental parameters of the respirator wearer and initial operating state data of the respirator, and performing a human-machine state assessment based on the physiological parameters and environmental parameters to determine a human-machine state assessment result; Specifically, in order to comprehensively assess the actual needs and risks faced by respirator wearers, the physiological parameters and environmental parameters of the respirator wearer and the initial operating status data of the respirator are first obtained. Among them, physiological parameters are collected through physiological sensors installed inside the respirator mask, including indicators such as respiratory rate, tidal volume, and blood oxygen saturation, which reflect the wearer's physiological state and breathing needs. Environmental parameters are collected through environmental sensors installed outside the respirator mask, including indicators such as temperature, humidity, and harmful gas concentration, which reflect the risk level of the wearer's environment. The initial operating status data of the respirator is obtained through the built-in operating status monitoring module of the respirator, including parameters such as battery power, filter usage time, and mask tightness, which reflect the performance status and remaining protection capabilities of the respirator.

[0025] After obtaining the above parameters, the respirator control system inputs the physiological parameters and environmental parameters into the preset human-machine state assessment model for comprehensive analysis and assessment. Specifically, the assessment model constructs a physiological state matrix based on physiological parameters to quantitatively assess the wearer's physiological load level and respiratory demand level; at the same time, it constructs an environmental risk matrix based on environmental parameters to quantitatively assess the degree of danger of the environment and the level of protection demand. Then, the physiological state assessment results are weighted and fused with the environmental risk assessment results to obtain a comprehensive human-machine state assessment result that fully reflects the wearer's actual needs and risks. For example, when the physiological load is high and the environmental risk is high, the assessment results will indicate that a higher intensity of respiratory protection is required; conversely, when the physiological load is low and the environmental risk is low, the assessment results indicate that the protection intensity can be appropriately reduced to take into account comfort.

[0026] Specifically, the physiological parameters and environmental parameters of the respirator wearer and the initial operating status data of the respirator are obtained, including: Acquiring wearer's physiological parameters through physiological sensors in the respirator mask; Collect the wearer's environmental parameters through environmental sensors outside the respirator mask; The initial operating status data of the respirator is collected through the sensors inside the respirator.

[0027] Specifically, in order to fully perceive the multi-source information in the respirator usage scenario and accurately evaluate user needs and equipment status, the respirator control system adopts a distributed sensor network to collect the wearer's physiological parameters, environmental parameters and the initial operating status data of the respirator through sensors inside and outside the respirator mask and sensors of the respirator itself.

[0028] First, the respirator mask is equipped with several physiological sensors, including respiratory flow sensors, blood oxygen concentration sensors, carbon dioxide concentration sensors, etc. Among them, the respiratory flow sensor is located at the inhalation valve and exhalation valve of the mask, and measures the wearer's respiratory parameters such as respiratory frequency and tidal volume in real time through the principle of thermistor or differential pressure meter; the blood oxygen concentration sensor is located on the inside of the mask, using photoelectric volumetric pulse wave recording technology to non-invasively and continuously measure the wearer's blood oxygen saturation; the carbon dioxide concentration sensor is also located on the inside of the mask, using non-dispersive infrared absorption method to monitor the carbon dioxide partial pressure of the wearer's exhaled gas in real time. These built-in physiological sensors in the mask can comprehensively collect the wearer's respiratory physiological information and accurately reflect their physiological load status and oxygenation demand level.

[0029] Secondly, the respirator mask is equipped with several environmental sensors, including temperature and humidity sensors, PM2.5 sensors, and harmful gas sensors. Among them, the temperature and humidity sensor uses a capacitive or resistive measurement principle to monitor the temperature and relative humidity of the respirator's use environment in real time; the PM2.5 sensor uses the laser scattering principle to measure the concentration of particles with a diameter of less than or equal to 2.5μm in the ambient air; the harmful gas sensor uses electrochemical or metal oxide semiconductor principles for quantitative detection of common toxic and harmful gases such as carbon monoxide and hydrogen sulfide. These external environmental sensors can collect air quality information in the respirator's environment in multiple dimensions, and comprehensively assess the degree of danger and protection needs of the environment.

[0030] Finally, the respirator itself also has several sensors built in to collect the initial operating status data of the respirator. For example, the battery management system uses voltage, current, temperature and other sensors to monitor the battery's capacity, charge and discharge status, health and other key parameters in real time; the filter management system uses differential pressure sensors, particulate matter concentration sensors, etc. to monitor the filter's usage time, filtration efficiency, degree of blockage and other status information; the air supply system uses flow meters, pressure gauges, etc. to collect operating data such as the speed, air supply volume, and working pressure of the motor and air valve. These state sensors that come with the respirator can comprehensively diagnose the performance of each component of the respirator and evaluate the protection capability and remaining life of the entire machine.

[0031] Based on the above embodiment, as an optional embodiment, the human-machine state assessment is performed based on the physiological parameters and the environmental parameters, and the human-machine state assessment result is determined, including: S201, constructing a physiological state matrix based on physiological parameters; Specifically, after the ventilator control system obtains the wearer's respiratory rate, tidal volume, blood oxygen saturation, carbon dioxide partial pressure and other physiological parameters from the physiological sensors built into the mask, it first pre-processes these raw data. On the one hand, through digital filtering, anomaly detection and other algorithms, noise, interference and invalid data in physiological signals are removed to improve data quality; on the other hand, through data synchronization, timestamp alignment and other methods, physiological data from different sources and frequencies are unified to a time base to ensure the time consistency of the data.

[0032] After obtaining high-quality physiological parameter data, the control system fills each parameter into different element positions of the matrix in a certain order and arrangement according to the pre-set matrix template, and constructs a multi-dimensional, multi-time scale physiological state matrix. The row vectors of the matrix represent different physiological parameter indicators, the column vectors represent the measured values ​​of the same parameter at different times, and the sub-matrix represents the parameter change trend within a certain time window. Through this matrix representation, multi-source physiological signals are organically unified into a tensor space, reflecting the multi-variable coupling characteristics of the human physiological system.

[0033] In the process of constructing the physiological state matrix, the control system fully considers the medical significance, time dynamic characteristics, correlation and constraint relationship of different physiological parameters, and optimizes the algebraic properties such as the sparsity, rank and condition number of the matrix, so that it can more accurately, stably and efficiently represent the complex physiological state changes of the human body. At the same time, according to the individual differences of the wearer such as age, gender, physique, etc., the control system can also flexibly adjust the parameter selection and weight distribution of the matrix template to realize the personalized customization of the matrix construction method, and further improve the pertinence and accuracy of the physiological state assessment.

[0034] S202, constructing an environmental risk matrix based on environmental parameters; Specifically, after the respirator control system obtains multiple parameters such as ambient temperature and humidity, PM2.5 concentration, and harmful gas concentration from the environmental sensors outside the mask, it first standardizes these environmental data. On the one hand, it uses statistical normalization and data compression to map environmental parameters of different dimensions and numerical ranges to a unified scale space to eliminate the impact of numerical size and distribution differences on subsequent analysis; on the other hand, it refers to relevant industry standards and ambient air quality evaluation indicators to determine the safety threshold, warning threshold, and danger threshold for each environmental parameter, and uses threshold normalization to characterize the risk level of each parameter.

[0035] After obtaining the standardized environmental parameter data, the control system fills the environmental parameters representing different risk factors into different rows of the matrix according to the matrix template set in the expert knowledge base, and fills the threshold normalized sequence reflecting the dynamic change trend of the same factor into different columns of the matrix, thereby constructing a multi-risk source and multi-time series environmental risk matrix. The larger the value of the element in the matrix, the higher the hazard level of the risk factor at that moment. Through the comprehensive analysis of row vectors, column vectors, and sub-matrices, the spatiotemporal distribution and evolution law of the composite environmental risk can be determined.

[0036] In the process of constructing the environmental risk matrix, the control system fully considers the characteristics of the harmful mechanism, mode of action, cumulative effect, etc. of different risk factors, optimizes the risk weight matrix of the matrix based on toxicology and environmental medicine knowledge, and highlights the risk contribution of key factors such as PM2.5 and toxic gases. At the same time, according to factors such as the wearer's physical condition and work intensity, the control system can also dynamically adjust the threshold setting and evaluation criteria of the matrix template to achieve real-time adaptation of the risk assessment model, and further improve the sensitivity and reliability of environmental risk warning.

[0037] S203, evaluating the wearer's physiological state based on the physiological state matrix to obtain a real-time physiological load index, and mapping the real-time physiological load index to a respirator air supply demand level; Specifically, the ventilator control system uses a multi-parameter fusion algorithm based on matrix analysis to perform feature extraction and dimensionality reduction on the physiological state matrix. First, the physiological state matrix is ​​decomposed into several principal component sub-matrices through matrix decomposition methods such as singular value decomposition, and the key eigenvectors that have the greatest impact on the change of physiological state are screened out; then, multivariate statistical methods such as weighted summation and principal component regression are used to linearly combine these key eigenvectors into a one-dimensional physiological load index, which changes continuously in value and reflects the overall stress level of the human physiological system.

[0038] After generating the real-time physiological load index, the control system further uses piecewise function mapping, fuzzy logic reasoning and other methods to discretize the continuously changing load index into several air supply demand levels. On the one hand, referring to human physiological knowledge and the experience of respiratory protection experts, several load index thresholds are pre-set, and the value range of the load index is divided into multiple intervals, each of which corresponds to an air supply demand level; on the other hand, a mapping relationship between dynamic characteristics such as the load index change rate and change trend and the air supply demand level is established to achieve dynamic adaptive adjustment of air supply demand assessment. By combining the above static threshold mapping with dynamic feature correction, a sequence of air supply demand levels that is synchronized with physiological load changes in real time is finally obtained.

[0039] In the process of extracting the physiological load index and mapping the air supply demand level, the control system makes full use of knowledge from disciplines such as ergonomics, respiratory physiology, and sports medicine, establishes a quantitative relationship between the physiological load level and the body's oxygen demand, and optimizes the scientific nature of the index synthesis method and the level classification standard. At the same time, for different groups of people and work scenarios, the control system can also flexibly adjust the calculation weight and level threshold setting of the physiological load index to achieve personalized adaptation of the evaluation model, further improving the accuracy and real-time performance of air supply demand analysis.

[0040] S204, evaluating the environmental status based on the environmental risk matrix, obtaining an environmental risk index, and mapping the environmental risk index to a respirator protection requirement level; Specifically, the respirator control system adopts a multi-criteria evaluation method based on matrix analysis to evaluate and integrate the environmental risk matrix. First, the risk weights of various risk factors are calculated through methods such as hierarchical analysis and entropy weight method to obtain a risk weight matrix that matches the size of the environmental risk matrix; then, the environmental risk matrix is ​​multiplied with the risk weight matrix using weighted matrix multiplication to obtain a risk assessment matrix in which each element has been weighted; finally, all elements of the risk assessment matrix are summed up to obtain a scalar environmental risk index, which simultaneously summarizes the risk contribution and dynamic change trend of each risk factor.

[0041] After generating the real-time environmental risk index, the control system further uses piecewise function mapping, artificial neural network and other methods to discretize the continuously changing risk index into several protection demand levels. On the one hand, referring to the quantitative relationship of disciplines such as occupational health, several risk index thresholds are pre-set, and the range of risk index changes is divided into multiple intervals, each of which corresponds to a protection demand level; on the other hand, a nonlinear mapping relationship between dynamic characteristics such as risk index change rate and cumulative effect and protection demand level is established to achieve intelligent adaptive adjustment of protection demand assessment. By combining the above static threshold division with dynamic feature correction, a dynamic protection demand level sequence closely related to the environmental risk state is finally obtained.

[0042] In the process of extracting environmental risk index and mapping protection demand level, the control system fully considers the characteristics of environmental risk factors such as harmful mechanism, synergy, dose-effect relationship, etc., optimizes the weight distribution and threshold selection in the matrix operation process, and highlights the assessment weights of key risk sources such as PM2.5 and toxic gases. At the same time, for different protection scenarios and wearers, the control system can also flexibly adjust the parameter settings and mapping function form of the risk assessment matrix to achieve scenario adaptation of the assessment model, and further improve the pertinence and sensitivity of protection demand analysis.

[0043] S205, taking the respirator protection requirement level and the respirator air supply requirement level as the human-machine status assessment result.

[0044] Specifically, after obtaining the real-time changing physiological load index and environmental risk index, the respirator control system uses the pre-established piecewise function mapping relationship to convert the two indexes into discrete air supply demand levels and protection demand levels. Since the level classification standards and threshold selection of the two dimensions are independent of each other, the air supply demand level and the protection demand level may not be completely consistent in value, but they reflect the wearer's current human-machine status and demand intensity from different aspects.

[0045] In order to facilitate the representation and transmission of the results, the control system uses a structured data format to encapsulate the two demand levels in parallel in the same state assessment result data packet. On the one hand, the data packet contains the real-time values ​​of the two demand levels, which are dynamically updated at certain time intervals; on the other hand, it also includes metadata such as timestamps, duration, and change trends, which comprehensively records the evolution trajectory of the human-machine state over time. The entire state assessment result data packet uses compact binary encoding, with small data volume and high transmission efficiency, which is very suitable for real-time flow in embedded systems.

[0046] By simultaneously outputting the respirator air supply demand level and the protection demand level as the results of the human-machine state assessment, this embodiment constructs a multi-dimensional state representation mechanism that takes into account the two core influencing factors of physiology and environment. Compared with a single physiological state assessment or a single environmental state assessment, it can more comprehensively and objectively reflect the real needs of complex human-machine systems, and achieve a higher level in terms of completeness and accuracy of state description. At the same time, since the two demand levels are obtained through a consistent matrix analysis method and encapsulated in a unified data format, the result representation of this parallel output is also highly consistent and comparable, which can effectively simplify the computational complexity of subsequent state fusion and strategy optimization, and improve the efficiency of human-machine collaboration.

[0047] S102, determining an initial control strategy of the ventilator based on the human-machine status evaluation result and the initial operating status data of the ventilator; Specifically, the respirator control system first extracts two key parameters, namely the respirator air supply demand level and the protection demand level, from the human-machine status evaluation results. The air supply demand level reflects the wearer's demand for the respirator's airflow supply, while the protection demand level reflects the wearer's demand for the respirator's filtering and protection capabilities. At the same time, the control system also obtains the current battery power, filter status, mask fit and other equipment performance parameters from the respirator's initial operating status data.

[0048] Based on the above parameters, the respirator control system determines a set of optimized initial control strategies through a preset strategy generation algorithm, taking into account human-machine requirements and equipment constraints. First, the initial air supply volume and frequency are determined based on the air supply demand level and the battery power of the respirator, while meeting the wearer's breathing needs and taking into account the energy consumption and endurance of the equipment. Secondly, the initial filtration level and filter replacement prompt threshold are determined based on the protection demand level and the filter status, while providing sufficient protection strength, extending the filter life and improving the economy of the equipment. Thirdly, the initial protection level and mask wearing prompt of the respirator are determined based on the protection demand level and the mask fit, while ensuring the reliability of protection and improving the wearing comfort. Finally, a set of initial control strategies that take into account multiple goals such as protection performance, wearing comfort, and equipment energy consumption are generated by comprehensively balancing various control parameters, and the strategy is sent to each execution module of the respirator to guide its operation.

[0049] Based on the above embodiment, as an optional embodiment, based on the human-machine state evaluation result and the initial operation state data of the respirator, the initial control strategy of the respirator is determined, including: S301, determining an initial air supply volume and an initial air supply frequency based on the air supply demand level of the ventilator and the initial operation status data; Specifically, after the state assessment module inputs the air supply demand level of the respirator, the respirator control system first reads the initial operating state data of the respirator from the storage module, including the current fan speed, air path resistance, pressure level, battery power, remaining usage time and other parameters. Then, the control system matches the air supply demand level with a pre-designed air supply strategy lookup table, which maps different air supply demand levels to corresponding air flow and frequency parameter ranges based on human physiological needs and respiratory protection experience. By looking up the table, the control system obtains the target range of air flow and frequency corresponding to the current air supply demand level.

[0050] After clarifying the target range of air supply volume and frequency, the control system further considers the constraints of the initial operating state parameters and uses a multi-objective optimization algorithm to determine the specific values ​​of the initial air supply volume and initial air supply frequency. On the one hand, the control system substitutes parameters such as fan speed and air path resistance into the aerodynamic model of the respirator system to calculate the maximum air volume and maximum frequency that the respirator can stably provide under the current state; on the other hand, the control system uses battery power and remaining usage time as constraints to calculate the theoretical values ​​of air volume and frequency under the maximum power limit while ensuring the shortest operating time. After comprehensively considering aerodynamic constraints and power constraints, the control system uses numerical methods such as weighted summation and convex optimization to solve the optimal initial air supply volume and air supply frequency values ​​that meet the current operating state constraints within the target range as the core parameters of the initial control strategy.

[0051] In the process of determining the initial air supply volume and frequency based on the air supply demand level and operating status data, the control system makes full use of the prior knowledge of the respirator system's structural parameters, aerodynamic model, energy consumption model, etc., and fine-tunes the target weights and constraints in the calculation process, while meeting the needs of the human body, minimizing the energy consumption and loss of the respirator itself, and achieving a high level of precision in collaborative optimization. At the same time, since the determination of the initial air supply volume and frequency uses a standard multi-objective optimization algorithm, the control strategy generation process has good robustness and adaptability, and can quickly converge to the optimal solution even when the respirator state or external environment fluctuates, thereby effectively ensuring the stability of human-machine collaborative control.

[0052] S302, determining an initial filtration level based on the respirator protection requirement level, and determining a filter replacement requirement level based on the initial filtration level and the initial operating status data; Specifically, after receiving the protection requirement level from the status assessment module, the respirator control system queries the recommended filter type and filtration efficiency under the protection level from the internal filter performance database. These recommended configurations are pre-set according to occupational health protection standards for different environmental hazard levels and pollution characteristics. By looking up the table, the control system obtains the optimal filter selection parameters under the current environmental conditions and defines them as the initial filtration level.

[0053] After determining the initial filtration level, the control system further reads the filter usage time and filtration resistance parameters in the initial operating status data to evaluate the remaining service life and performance degradation of the current filter. On the one hand, the control system compares the filter usage time with the rated replacement cycle to calculate the remaining service life percentage of the filter; on the other hand, it compares the measured filtration resistance with the initial nominal resistance to calculate the relative increase percentage of the filter resistance. After obtaining the quantitative index of the filter performance, the control system matches it with a filter replacement requirement level threshold table, which sets the corresponding relationship between the remaining life and the relative increase degree of resistance and the replacement requirement level. If the measured index reaches a certain replacement level threshold, it is determined that the filter replacement requirement is at that level.

[0054] In the process of determining the initial filtration level and the filter replacement requirement level, the control system adopts a strategy that combines table lookup reasoning based on prior expert knowledge and data-driven threshold judgment. It takes into account the accuracy of replacement demand judgment while quickly matching, and strikes a good balance between the real-time and reliability of decision-making. At the same time, the control system also adaptively corrects the initial filtration level and replacement threshold according to the actual filter configuration of the respirator, improving the personalized level of filtration performance analysis.

[0055] S303, determining an initial protection level of the respirator based on the initial operating state data, and determining an initial safety control strategy according to the initial protection level of the respirator and a preset protection level threshold; Specifically, after reading the initial operating status data, the respirator control system focuses on analyzing several key parameters related to the safety performance of the respirator: air tightness test results, key component working status, battery pack output voltage, alarm device self-test results, etc. The control system compares these parameters with the corresponding safety standard thresholds one by one. The air tightness test results reflect the integrity of the respirator in terms of air circuit sealing, the key component working status reflects the functional health level of the core components, the battery pack output voltage reflects whether the energy reserve of the respirator is sufficient, and the alarm device self-test results reflect whether the safety warning channel is normal. The control system adopts the fuzzy comprehensive evaluation method to weighted sum the compliance status of various indicators to obtain an initial protection level score that reflects the overall safety level of the respirator.

[0056] After obtaining the initial protection level score, the control system further compares it with a preset safety level threshold. This threshold is set according to the general safety standards in the protection field and combined with the actual working environment characteristics of the respirator, and represents the basic requirements for the safety protection level of the respirator. If the initial protection level score is higher than the safety level threshold, it means that the overall safety status of the respirator is good, and the control system generates a conventional initial safety control strategy, including normal frequency of air tightness detection, regular key component status monitoring, standard undervoltage protection measures, etc. When the initial protection level score is lower than the safety level threshold, it means that the respirator has certain safety hazards, and the control system generates an enhanced initial safety control strategy, including increasing the frequency of air tightness detection, real-time monitoring of key component working parameters, early activation of undervoltage warnings, and preparation of emergency protection responses. These strategies can play an early warning and active defense role when the overall function of the respirator has not been damaged, avoiding the deterioration of accidents.

[0057] In the process of realizing the initial protection level assessment and generating safety strategies, the control system adopts a combination of multi-index fusion evaluation and threshold comparison, which can comprehensively consider all key factors of respirator safety protection, quantitatively analyze the overall safety level, and match strategies accordingly, achieving a high level in terms of comprehensive risk assessment and flexibility of response. At the same time, the control system dynamically adjusts the protection level threshold according to the risk level of the respirator's environment and the protection task, making the safety strategy closer to actual needs. S304: The filter replacement requirement level, the initial air supply volume, the initial safety control strategy and the initial air supply frequency are used as the initial control strategy.

[0058] Specifically, the respirator control system uses lightweight data exchange formats such as JSON to encapsulate the four parameters of filter replacement requirement level, initial air supply volume, initial safety control strategy, and initial air supply frequency into a unified initial control strategy object. Among them, the filter replacement requirement level is represented by an enumeration type field, such as "0-normal" "1-replacement recommended", etc.; the initial air supply volume and air supply frequency are directly represented by floating-point values, and the dimension is indicated; the initial safety control strategy is represented by a nested JSON object, which contains multiple specific strategy parameters such as air tightness detection frequency, key component monitoring cycle, undervoltage alarm threshold, etc. The entire initial control strategy object is organized in a tree-like hierarchical structure with clear logic and easy parsing and processing.

[0059] It is worth mentioning that in the process of generating the initial control strategy object, the control system does not simply stack parameters, but follows a strict strategy consistency check mechanism. On the one hand, the control system checks whether there is a contradiction between the filter replacement demand level and the initial air supply volume and frequency. If the replacement demand is high and the air supply parameters are also high, it means that the strategies of the two dimensions are not coordinated and need to be adjusted; on the other hand, the control system must also ensure the logical correspondence between the initial air supply parameters and the safety control strategy. For example, when the air supply volume is large, it is necessary to encrypt the air tightness detection and component monitoring accordingly to prevent leakage or damage. Only the initial control strategy that has passed the consistency check can be officially output to ensure the feasibility and efficiency of the strategy.

[0060] By using the filter replacement demand level, initial air supply volume, initial safety control strategy and initial air supply frequency as the initial control strategy output, this embodiment constructs a startup strategy generation scheme that takes into account protection effect, human-machine coordination and system safety, and comprehensively considers multiple control objectives of the intelligent respirator in the initial stage, reflecting the design concept of "technology as the device". This multi-parameter fusion structured strategy representation is conducive to improving the startup efficiency of the control system and achieving rapid entry into the optimal working state; on the other hand, it is also convenient for the control strategy to be parsed and executed between different functional modules of the respirator, improving the system's collaborative working ability.

[0061] S103, acquiring historical data during the use of the ventilator, the historical data including historical physiological feedback data and ventilator operation status data, and adaptively adjusting the initial control strategy based on the historical data to obtain an optimized dynamic control strategy; Specifically, the respirator control system will continuously record and accumulate two types of key historical data during the use of the device: one is historical physiological feedback data, including wearing comfort scores, breathing smoothness scores, protection safety scores, etc., which reflect the wearer's subjective feelings and evaluations of the respirator control effect; the other is respirator operation status data, including air supply volume, filtration resistance, mask pressure, energy consumption level, etc., which reflects the performance and resource consumption of the respirator under different control strategies. These two types of data are collected through the human-computer interaction interface on the respirator mask and the built-in operation monitoring module of the device, and uploaded to the data analysis engine of the control system for processing.

[0062] The data analysis engine first cleans, integrates and converts historical data to establish a structured data analysis model. Then, using data mining algorithms such as correlation analysis, cluster analysis, and anomaly detection, it extracts valuable features and patterns from massive historical data. For example, it identifies the correlation pattern between the wearer's subjective feedback and the respirator control parameters, finds the common characteristics of the optimal control strategy under different working scenarios, and determines the abnormal deviation of the respirator's operating status. Based on these analysis results, the data analysis engine will output a series of strategy optimization suggestions, such as increasing or decreasing the air supply under specific working conditions, turning on or off a certain protection function, and correcting the mask pressure threshold.

[0063] Finally, the respirator control system will comprehensively consider the strategy optimization suggestions given by the data analysis engine, make adaptive adjustments to the initial control strategy, and generate an optimized dynamic control strategy. The adjustment process follows the principles of gradual optimization and controllable risk. By setting a reasonable change step size and safety threshold, optimization measures are gradually introduced on the basis of the initial strategy. At the same time, the optimization effect is evaluated in real time, and risky optimization actions are rolled back in time. After multiple rounds of iterative optimization, the dynamic control strategy will achieve targeted enhancement and dynamic adaptation based on the original static strategy, and more accurately match the actual needs and usage habits of the wearer.

[0064] S104, acquiring multi-source sensor data of the ventilator in real time, and adjusting the dynamic control strategy in real time based on the multi-source sensor data to obtain an adaptive control strategy; Specifically, the respirator is integrated with various types of sensors, including physiological sensors, environmental sensors, and equipment status sensors, which respectively collect multi-source heterogeneous data such as the wearer's physiological parameters, environmental parameters, and respirator operating parameters in real time. Among them, physiological sensors monitor the wearer's real-time respiratory rate, tidal volume, blood oxygen concentration and other indicators, reflecting the dynamic changes of their physiological load and respiratory needs; environmental sensors monitor the surrounding temperature and humidity, harmful gas concentration, dust concentration and other parameters, reflecting the dynamic changes of the wearer's dangerous environment and protection needs; equipment status sensors monitor the respirator's battery voltage, motor speed, valve opening, filter resistance and other parameters, reflecting the real-time fluctuations of the respirator's performance status. These multi-source sensor data are uploaded to the data processing module of the respirator control system in real time through a high-speed data bus.

[0065] The data processing module first preprocesses the multi-source sensor data, including data cleaning, data synchronization, data normalization and other operations, eliminates noise and anomalies in the data, unifies the data format and timestamp, and maps data from different sources and scales to the same feature space. Then, the processed sensor data is input into the state assessment engine, and through the multi-sensor data fusion algorithm, a multi-dimensional feature vector reflecting the current human-machine state is dynamically constructed to evaluate the wearer's physiological risk level, environmental hazard level and respirator performance level in real time, and generate real-time scores for various key indicators. Next, the state assessment engine compares the real-time state assessment results with the currently executed dynamic control strategy, identifies the deviation between the control strategy and actual needs, and uses the strategy adaptive optimization algorithm to automatically generate strategy adjustment instructions, including air supply parameter adjustment, filter mode switching, alarm prompt triggering, etc.

[0066] Finally, the respirator control system sends the strategy adjustment instructions to each execution module, and corrects the relevant parameters in the dynamic control strategy in real time, so that it can quickly adapt to the changes in the current human-machine state and environmental conditions, and generate the latest adaptive control strategy. During the entire adjustment process, the control system adopts closed-loop control and feedback correction mechanism to evaluate the effect of strategy adjustment in real time, and dynamically balance the timeliness and stability of strategy adjustment. On the one hand, by setting the rapid response threshold of strategy adjustment, it is ensured that the corresponding strategy adjustment can be made in time when the human-machine state fluctuates violently; on the other hand, by setting the strategy rollback mechanism and safety constraints, it is avoided that the strategy adjustment is too large or the direction is wrong, resulting in instability of the control strategy or endangering the safety of the user. After a series of links such as real-time sensing, fusion analysis, strategy adjustment, and effect evaluation, the adaptive control strategy can quickly, smoothly, and accurately converge to the optimal control mode in the current scenario, providing the wearer with consistent high-quality protection services.

[0067] Based on the above embodiment, as an optional embodiment, the multi-source sensor data includes real-time physiological parameters, real-time environmental parameters and real-time device status parameters; the dynamic control strategy is adjusted in real time based on the multi-source sensor data to obtain an adaptive control strategy, including: S401, constructing a state perception matrix integrating man-machine-environment based on real-time physiological parameters, real-time environmental parameters and real-time device state parameters; Specifically, in order to achieve comprehensive perception of the wearer's physiological state, external environmental changes, and the operating status of the respirator, the present invention integrates multiple types of sensors on the respirator. Among them, the physiological sensor is responsible for collecting the wearer's real-time physiological parameters, such as respiratory rate, tidal volume, blood oxygen saturation, etc.; the environmental sensor is used to obtain the real-time parameters of the environment in which the respirator is located, such as temperature, humidity, air pressure, pollutant concentration, etc.; the equipment status sensor monitors the real-time operating parameters of the respirator itself, such as filter resistance, fan speed, battery power, etc.

[0068] The multi-source heterogeneous data collected by these sensors can fully reflect the real-time status of people, machines and the environment. However, the formats, frequencies and units of these raw data are different, making it difficult to directly use them for state analysis and strategy optimization. Therefore, the present invention proposes to construct a state perception matrix that integrates man-machine-environment, normalize and extract features from multi-source sensor data, and form a unified matrix representation.

[0069] Specifically, the state perception matrix consists of three sub-matrices: physiological state matrix, environmental state matrix and device state matrix. The physiological state matrix contains key features such as the wearer's breathing parameters and physical load; the environmental state matrix includes environmental quality indicators such as temperature, humidity, and pollution index; and the device state matrix covers performance parameters such as filtration efficiency, wind resistance coefficient, and energy consumption ratio.

[0070] The three sub-matrices are aligned according to a unified timestamp and the final state perception matrix is ​​generated through weighted fusion. Each element in the matrix represents the comprehensive state characteristics of the three dimensions of human, machine, and environment at a certain moment, and has the same dimension and value range, which is convenient for subsequent state analysis and strategy tuning.

[0071] The advantages of building an integrated human-machine-environment state perception matrix are: first, it regards humans, machines, and the environment as a highly coupled complex system, and describes their internal connections in a unified matrix form, which helps to understand and control the dynamic changes of the entire system; second, the state perception matrix reduces the dimension of multi-source heterogeneous data, simplifies the complexity of data processing and analysis, and lays an efficient data foundation for real-time strategy optimization; finally, the perception matrix can quantify the degree of state matching between humans, machines, and the environment. When the states of the three change significantly, the corresponding strategy adjustments are triggered in time, thereby realizing the adaptability and real-time nature of ventilator control.

[0072] S402, inputting the state perception matrix into a preset strategy optimization model, and performing real-time correction on the dynamic control strategy through preset adaptive weight adjustment and multi-objective optimization to generate an optimal adaptive control strategy.

[0073] Specifically, the strategy optimization model uses a multi-objective optimization algorithm based on reinforcement learning. The algorithm uses the state perception matrix as the environmental state and the air supply parameters of the respirator as the execution action. Through continuous interaction with the environment and trial and error learning, it finds a control strategy that can achieve the best balance between various performance indicators. During the optimization process, the algorithm dynamically adjusts the weight coefficients of each goal to adapt to the real-time changes in the status of people, machines, and the environment. For example, when environmental pollution intensifies, the model will increase the weight of protective performance; when the wearer is in a state of strenuous exercise, the model will appropriately sacrifice energy saving to ensure comfort.

[0074] The core of multi-objective optimization is to design a comprehensive evaluation function, which weights and sums multiple performance indicators such as respiratory resistance, filtration efficiency, and energy consumption to form a scalar optimization target. Then, under the constraints determined by the state perception matrix, the global optimal solution of the evaluation function is found through an intelligent search algorithm to obtain the corresponding optimal control strategy. This strategy not only weighs various performances as a whole, but also makes adaptive adjustments to the real-time status of people, machines, and the environment, and can be regarded as the best strategy at the current moment.

[0075] As the state perception matrix is ​​updated in real time, the optimal control strategy of the respirator will also change dynamically. The strategy optimization model will continuously monitor the changing trends of the human, machine, and environment states. When significant changes are detected, adaptive weight adjustment and multi-objective optimization will be immediately initiated to correct the current control strategy so that it always keeps pace with the actual state. This real-time closed-loop optimization mechanism can quickly respond to external interference and improve the timeliness and robustness of the control strategy.

[0076] S105, controlling the ventilator based on the adaptive control strategy.

[0077] Specifically, the adaptive control strategy of the respirator will involve multiple aspects such as air supply control, filtration control, mask pressure control, power supply control, and human-computer interaction control. Each control link has corresponding target parameters and actuators. The respirator control system first parses and converts the received adaptive control strategy, extracts the target parameters of each control link, such as air supply volume, filtration level, mask pressure threshold, power consumption budget, interaction mode, etc., and converts these parameters into control instructions for each actuator.

[0078] For air supply control, the control system will achieve precise airflow output control by controlling the motor speed and valve opening of the respirator according to the air supply parameters determined by the adaptive strategy. By real-time monitoring of air supply pressure and flow, and combining the wearer's breathing rhythm and tidal volume changes, the control system can dynamically adjust the air supply strategy to minimize energy consumption and extend battery life while continuing to meet the user's breathing needs. At the same time, in an emergency, the air supply control can also quickly switch to forced air supply mode according to the strategy instructions to ensure the wearer's breathing safety.

[0079] For filtration control, the control system will select the appropriate type and number of filter elements according to the protection level determined by the adaptive strategy, and achieve the best filtration effect by controlling the speed and airflow direction of the filter fan. The system will also monitor the use status and blockage degree of the filter in real time, and dynamically adjust the filtration strategy according to the changes in the filtration resistance, so as to maximize the life of the filter while meeting the protection needs. When the filter reaches the replacement period or fails, the system can remind the user to replace it in time according to the strategy instructions to ensure the protection effect.

[0080] For mask pressure control, the control system will accurately control the pressure inside the mask by controlling the solenoid valve and pressure relief device according to the pressure range determined by the adaptive strategy, ensuring the tightness of the mask while maximizing the wearing comfort. The system will also monitor the pressure and airtightness inside the mask in real time, dynamically adjust the control strategy through a closed-loop control algorithm, and compensate for the pressure fluctuations caused by mask leakage. When the pressure exceeds the safe range, the control system will promptly trigger an alarm prompt to guide the user to check the mask wearing condition.

[0081] For power supply control, the control system will coordinate various power consumption modules according to the power consumption budget determined by the adaptive strategy, and dynamically optimize the energy allocation plan. By real-time monitoring of the battery status and discharge curve, the system can accurately estimate the remaining working time, reasonably arrange various tasks, and avoid excessive discharge and sudden power outages. At the same time, the control system will also start the backup power supply or trigger the low power mode when necessary according to the strategy instructions to maximize the battery life of the respirator.

[0082] For human-machine interactive control, the control system will transmit various status information, alarm reminders and operation instructions to users through multiple interactive channels such as the display screen on the respirator mask, voice prompts, vibration feedback, etc. according to the interactive mode determined by the adaptive strategy, and guide users to make necessary parameter settings and mode selections. The system will also collect subjective feedback from users through the interactive interface to evaluate the effect of the current control strategy and provide reference for the next step of strategy optimization.

[0083] Based on the above embodiment, as an optional embodiment, the ventilator is controlled based on an adaptive control strategy, including: S501, controlling the speed of the fan and the opening of the air supply valve using a PID control algorithm according to the target air supply volume and target air supply frequency of the adaptive control strategy; Specifically, the air supply control system of the present invention is composed of a fan, an air supply valve, a drive circuit and a PID controller. Among them, the fan is the power source of air supply, and its rotation speed directly determines the size of the air supply volume; the air supply valve is an actuator for adjusting the air supply resistance, and its opening size affects the air supply frequency. According to the target value of the adaptive control strategy, the PID controller measures the actual value of the fan speed and the opening of the air supply valve in real time, calculates the deviation between the two, and outputs the corresponding control signal through the drive circuit to feedback adjust the fan and the air supply valve, and finally stabilizes the actual air supply volume and frequency near the target value.

[0084] The PID control algorithm consists of three control functions: proportional, integral, and differential. It can take into account the rapidity, accuracy, and stability of control at the same time. Its basic principle is: the proportional term responds quickly according to the deviation between the target value and the actual value, providing the main control function; the integral term eliminates the static error of the system to ensure the accuracy of long-term control; the differential term compensates in advance according to the change trend of the deviation to suppress the oscillation and overshoot of the system. Through the optimal combination of the three functions, the PID controller can stabilize the fan speed and air supply valve opening at the target value in the shortest time, and has a strong ability to suppress external interference.

[0085] In a specific embodiment, the control formula of the PID controller is: u(t)=Kp*e(t)+Ki*∫e(t)dt+Kd*de(t) / dt Among them, u(t) represents the output of the controller, that is, the control amount applied to the controlled object; e(t) represents the deviation between the target value and the actual value, that is, e(t)=r(t)-y(t), r(t) is the target value, and y(t) is the actual value; Kp, Ki, and Kd are the proportional coefficient, integral coefficient, and differential coefficient, respectively, which determine the weights of the three control actions.

[0086] The proportional term Kp*e(t) reflects the sensitivity of the controller to the error size. The larger the Kp, the stronger the controller's response to the error and the faster the control speed. However, if Kp is too large, it is easy to cause system oscillation.

[0087] The integral term Ki*∫e(t)dt reflects the controller's response to the accumulated error. ∫e(t)dt is the integral of the error e(t) over time t. The larger the Ki, the stronger the integral effect and the smaller the steady-state error of the system. However, if Ki is too large, the system response will be slower.

[0088] The differential term Kd*de(t) / dt reflects the sensitivity of the controller to the error change rate. de(t) / dt is the differential of the error e(t) with respect to time t, which represents the instantaneous change speed of the error. The larger the Kd, the stronger the differential effect and the smaller the overshoot of the system response. However, if Kd is too large, it will amplify the noise interference and cause system jitter.

[0089] In practical applications, it is necessary to repeatedly debug and optimize the three parameters Kp, Ki, and Kd according to the characteristics of the controlled object and the control requirements to find the best parameter combination so that the control system can achieve the best balance in terms of speed, accuracy, stability, etc. Taking the fan speed control as an example, the speed deviation e(t) is used as the input of the PID controller, and the controller output u(t) is used as the control voltage of the drive circuit to change the working current of the fan, thereby adjusting the fan speed so that it can quickly and stably track the target speed. Similarly, for the air supply valve control, the valve opening deviation is used as the input of the PID controller, and the controller output is used as the control signal of the valve driver to adjust the actual opening of the valve so that it can quickly and stably track the target opening.

[0090] It should be pointed out that the above formula gives the continuous time form of the PID controller. In the digital control system, the formula needs to be discretized and converted into a discrete recursive formula for computer programming. The discretized PID control formula can be expressed as: u(k)=Kp*e(k)+Ki*∑e(k)+Kd*[e(k)-e(k-1)] Where k is the serial number of the discrete time point, u(k) is the controller output at the kth sampling moment, e(k) is the error at the kth sampling moment, ∑e(k) is the cumulative sum of errors from the beginning to the kth sampling moment, and [e(k)-e(k-1)] is the difference between the errors at two adjacent sampling moments. The discrete PID controller samples the error signal at a fixed sampling period T, calculates the control quantity according to the above recursive formula, and then converts it into a continuous control signal through the holding link to drive the actuator to operate.

[0091] In the actual control process, the parameters of the PID controller need to be adjusted according to the dynamic characteristics of the respirator system. In order to adapt to the real-time changes in the state of people, machines, and the environment, the present invention adopts an adaptive parameter setting method to dynamically adjust the proportional, integral, and differential coefficients of the PID controller according to the feedback information of the state perception matrix, so that it is always in the optimal working state. For example, when the wearer's respiratory load suddenly increases, the controller will increase the proportional coefficient and integral coefficient accordingly to speed up the system's response speed and tracking accuracy; when the ambient temperature drops sharply, the controller will appropriately reduce the differential coefficient to reduce the overshoot of the system and ensure the stability of the air supply temperature.

[0092] In a specific embodiment, in order to further enhance the stability and reliability of the ventilator control system, the present invention introduces a redundant design and a self-diagnosis mechanism in key functional modules. By configuring hot backup on core components such as fans, filters, and power supplies, once the main components fail, the backup components can take over seamlessly and immediately to ensure that the core functions of the ventilator are not interrupted. At the same time, the system's built-in self-diagnosis program will regularly and comprehensively scan the working status of each module, and detect and warn of potential problems and hidden dangers as early as possible, thereby effectively avoiding performance degradation caused by cumulative failures and ensuring that the ventilator is always in the best working condition during long-term continuous use.

[0093] Taking the fan of the air supply module as an example, the present invention is additionally equipped with a backup fan with equivalent performance in addition to the main fan. The two fans are connected in parallel on the same air supply duct through an intelligent switching circuit. Normally, the main fan works independently, and the backup fan is in hot standby state. The control system will monitor the operating parameters of the main fan in real time, including speed, current, noise, temperature, vibration, etc. Once abnormal data is found, such as long-term exceeding the standard or sudden change, the system will immediately determine that the main fan may have a fault, automatically cut off its power supply and put it into use. The backup fan. Since the response time of the switching circuit can be controlled at the millisecond level, the backup fan can achieve disturbance-free switching before the main fan completely stops, and the user will not feel any interruption in air supply.

[0094] In addition to hardware redundancy, the system also deploys comprehensive fault diagnosis and maintenance reminder services. When the main fan switches to the backup fan, the fault diagnosis program will further analyze the various parameters before the fan stops, and combine the previous work logs to infer the most likely cause of the fault and generate a fault diagnosis report. At the same time, the system immediately sends a warning message to the user's mobile terminal, informing the user that "the respirator fan has switched to the backup state, please repair or replace the main fan in time." If the user authorizes in advance, the system can also upload the diagnostic report to the cloud management platform or send it directly to the designated maintenance center to realize automatic fault reporting and remote after-sales. This proactive warning and prompt can urge users to carry out maintenance in a timely manner and reduce the risk of secondary failures caused by component loss.

[0095] S502, adjusting the filter wind resistance of the filter in real time according to the filter level instruction in the adaptive control strategy; Specifically, the filter of the present invention adopts a multi-level adjustable structure, which is composed of a primary filter layer, a medium-efficiency filter layer and a high-efficiency filter layer. The primary filter layer mainly filters large particles and visible droplets, the medium-efficiency filter layer intercepts fine particles and some harmful gases, and the high-efficiency filter layer deeply purifies submicron particles and biochemical agents. The filtration accuracy and wind resistance coefficient of the three filter layers are in a progressive relationship, and can be combined and switched according to actual needs to achieve dynamic adjustment of the filtration level.

[0096] Electric control valves are set at the air inlet and outlet of the filter. By changing the opening and closing state of the valves, the airflow can be controlled to pass through different filter layers in sequence, thereby changing the total wind resistance of the filter. For example, when only the primary valve is opened, the airflow only passes through the primary filter layer, with the smallest wind resistance and the lowest filtration accuracy; when all three valves are opened, the airflow passes through the three-level filter in sequence, with the largest wind resistance and the highest filtration accuracy; when some valves are closed, a balance can be made between the filtration effect and the breathing resistance, achieving intermediate adjustment of the filtration level.

[0097] The adaptive control strategy will generate filtering level instructions in real time according to the environmental pollution conditions reflected by the state perception matrix. When the environmental pollution is light, the strategy will reduce the filtering level and reduce the filtering wind resistance to reduce the breathing burden; when the pollution is aggravated, the strategy will increase the filtering level and increase the filtering wind resistance to meet higher protection requirements. After the instruction is generated, the control system will drive the corresponding electronically controlled valve to open and close, switch the working state of the filter, and make its filtering wind resistance dynamically adapt to environmental changes.

[0098] It is worth noting that the adjustment of the filter resistance needs to be coordinated with the air supply control. Since the change of the filter resistance will affect the total resistance of the respirator, and then affect the air supply volume and frequency, when adjusting the filter, it is also necessary to adjust the fan speed and the air supply valve opening accordingly through the PID control algorithm to maintain the stability of the air supply performance. The state perception matrix plays an important role in information fusion in this process. It comprehensively considers the changes in the environmental state and the system state, and provides a basis for the formulation of adaptive control strategies, so that the filter control and air supply control can work together efficiently to cope with complex and changeable operating conditions.

[0099] S503, according to the target safety control strategy in the adaptive control strategy, when the protection performance drops to a preset threshold, a safety warning is issued.

[0100] Specifically, the present invention sets a special safety status assessment module in the state perception matrix to evaluate the comprehensive protection performance of the respirator in real time. The module comprehensively analyzes multiple key indicators such as environmental pollution level, filter saturation, mask sealing, etc., and obtains a quantitative protection performance score through weighted fusion. Under normal circumstances, the score should be stable at a relatively high safety level, indicating that the respirator can effectively block external pollution and provide clean breathing air for the wearer.

[0101] However, when external pollution suddenly increases or equipment fails, the protection performance score may drop sharply. To deal with this situation, the present invention presets multiple safety thresholds in the adaptive control strategy, and formulates corresponding target safety control measures for different thresholds. For example, when the protection performance score drops to the warning threshold, the control strategy will instruct the respirator to issue a voice or visual warning to remind the wearer that the danger of the current environment has increased and that they need to be vigilant; when the score further drops to the alarm threshold, the control strategy will trigger the alarm device of the respirator, generate a strong sound and light alarm signal, and urge the wearer to evacuate the dangerous area as soon as possible; when the score is lower than the emergency evacuation threshold, the control strategy will even forcibly cut off the air supply of the respirator, forcing the wearer to stop working and immediately move to a safe area.

[0102] It is worth mentioning that the issuance of safety warnings needs to consider the balance between timeliness and reliability. To avoid false positives and false negatives, the safety status assessment module will smooth and filter the instantaneous fluctuations of the protection performance, and the warning will only be triggered if the score is lower than the threshold for multiple consecutive times. At the same time, to ensure that the warning information is conveyed in a timely manner, the present invention also integrates a variety of redundant warning methods on the respirator, including sound alarms, flashing lights, face screen displays, vibration reminders, etc., and selects the most suitable warning combination for the current occasion through an intelligent scheduling algorithm, striving to attract the wearer's attention at the first time.

[0103] See also Figure 2 , Figure 2 An architecture diagram of a ventilator adaptive control system provided in an embodiment of the present application, the ventilator adaptive control system may include: A data acquisition module 1 is used to acquire physiological parameters and environmental parameters of the respirator wearer and initial operating state data of the respirator, and to perform a human-machine state assessment based on the physiological parameters and environmental parameters to determine a human-machine state assessment result; An initial strategy generating module 2, for determining an initial control strategy of the ventilator based on the human-machine status evaluation result and the initial operation status data of the ventilator; The optimization module 3 is used to obtain historical data during the use of the ventilator, the historical data including historical physiological feedback data and ventilator operation status data, and adaptively adjust the initial control strategy based on the historical data to obtain an optimized dynamic control strategy; An adjustment module 4 is used to obtain multi-source sensor data of the respirator in real time, and adjust the dynamic control strategy in real time based on the multi-source sensor data to obtain an adaptive control strategy; The control module 5 is used to control the ventilator based on an adaptive control strategy.

[0104] It should be noted that: when the system provided in the above embodiment realizes its functions, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the system and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0105] Please refer to Figure 3 The application also discloses an electronic device. Figure 3 The electronic device 300 may include: at least one processor 301, at least one network interface 304, a user interface 303, a memory 305, and at least one communication bus 302 or end-to-end wireless communication.

[0106] The communication bus 302 is used to realize the connection and communication between these components.

[0107] The user interface 303 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 303 may also include a standard wired interface and a wireless interface.

[0108] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0109] Among them, the processor 301 may include one or more processing cores. The processor 301 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 305, and calling data stored in the memory 305. Optionally, the processor 301 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field~Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 301 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processor (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 301, and it can be implemented separately through a chip.

[0110] Among them, the memory 305 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read~Only Memory). Optionally, the memory 305 includes a non-transitory computer-readable medium. The memory 305 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 305 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 305 may optionally be at least one storage system located away from the aforementioned processor 301. Refer to Figure 3 , the memory 305 as a computer storage medium may include an operating system, a network communication module, a user interface module and an application program of a ventilator adaptive control method.

[0111] exist Figure 3In the electronic device 300 shown, the user interface 303 is mainly used to provide an input interface for the user and obtain the data input by the user; and the processor 301 can be used to call the application program storing the ventilator adaptive control method in the memory 305. When executed by one or more processors 301, the electronic device 300 executes one or more methods in the above-mentioned embodiments. It should be noted that for the aforementioned method embodiments, for the sake of simple description, they are all expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited to the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application. In the above embodiments, the description of each embodiment has its own emphasis. For the parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0112] In the several implementations provided in this application, it should be understood that the disclosed system can be implemented in other ways. For example, the system embodiments described above are only schematic, such as the division of modules, which is only a logical function division. There may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, indirect coupling or communication connection of systems or modules, which can be electrical or other forms.

[0113] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0114] The present application also provides a computer storage medium that can store multiple instructions, which are suitable for being loaded and executed by a processor as described above. Figure 1 The ventilator adaptive control method of the embodiment shown in the figure can be found in the specific execution process. Figure 1 The specific description of the illustrated embodiment will not be repeated here.

[0115] In addition, each functional module in each embodiment of the present application can be integrated into a processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The above integrated modules can be implemented in the form of hardware or software functional modules.

[0116] If the integrated module is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: various media that can store program codes, such as USB flash drives, mobile hard drives, magnetic disks or optical disks.

[0117] The above are only exemplary embodiments of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification and the truth of practice, those skilled in the art will easily think of other embodiments of the present disclosure.

[0118] This application is intended to cover any variation, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the art not described in the present disclosure. The description and examples are to be regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A ventilator adaptive control method, characterized in that: The method comprises: Acquire physiological parameters and environmental parameters of the respirator wearer and initial operating state data of the respirator, and perform a human-machine state assessment based on the physiological parameters and the environmental parameters to determine a human-machine state assessment result; Determining an initial control strategy for the ventilator based on the human-machine status assessment result and the initial operating status data of the ventilator; Acquire historical data during the use of the ventilator, the historical data including historical physiological feedback data and ventilator operation status data, and adaptively adjust the initial control strategy based on the historical data to obtain an optimized dynamic control strategy; Acquire multi-source sensor data of the ventilator in real time, and adjust the dynamic control strategy in real time based on the multi-source sensor data to obtain an adaptive control strategy; The ventilator is controlled based on the adaptive control strategy.

2. The method according to claim 1, characterized in that The step of obtaining physiological parameters and environmental parameters of the respirator wearer and initial operating state data of the respirator includes: collecting the physiological parameters of the wearer through a physiological sensor in the respirator mask; collecting the wearer's environmental parameters through an environmental sensor outside the respirator mask; The initial operating state data of the respirator is collected by a sensor in the respirator.

3. The method according to claim 1, characterized in that The performing of the human-machine state assessment based on the physiological parameters and the environmental parameters and determining the human-machine state assessment result includes: constructing a physiological state matrix based on the physiological parameters; constructing an environmental risk matrix based on the environmental parameters; Evaluate the wearer's physiological state based on the physiological state matrix to obtain a real-time physiological load index, and map the real-time physiological load index to a respirator air supply demand level; Evaluate the environmental status based on the environmental risk matrix to obtain an environmental risk index, and map the environmental risk index to a respirator protection requirement level; The respirator protection requirement level and the respirator air supply requirement level are used as the human-machine status evaluation result.

4. The method according to claim 1, characterized in that: Determining an initial control strategy of the ventilator based on the human-machine state evaluation result and the initial operating state data of the ventilator includes: Determining an initial air supply volume and an initial air supply frequency based on the ventilator air supply demand level and the initial operating state data; determining an initial filtration level based on the respirator protection requirement level, and determining a filter replacement requirement level based on the initial filtration level and the initial operating status data; Determining an initial protection level of the respirator based on the initial operating state data, and determining an initial safety control strategy according to the initial protection level of the respirator and a preset protection level threshold; The filter replacement requirement level, the initial air supply volume, the initial safety control strategy and the initial air supply frequency are used as the initial control strategy.

5. The method according to claim 1, characterized in that The adaptively adjusting the initial control strategy based on the historical data to obtain an optimized dynamic control strategy includes: Extracting features from the historical physiological feedback data and the respirator operating state data to obtain feedback feature data and operating state feature data, and performing association analysis based on the feedback feature data and the operating state feature data to obtain association rules between the wearer's subjective feelings and the respirator control parameters; By using a reinforcement learning algorithm and taking improving wearing comfort and protective safety as optimization goals, the initial respirator control strategy is adaptively optimized in combination with association rules to obtain the optimized dynamic control strategy.

6. The method according to claim 1, characterized in that The multi-source sensor data includes real-time physiological parameters, real-time environmental parameters and real-time device status parameters; the dynamic control strategy is adjusted in real time based on the multi-source sensor data to obtain an adaptive control strategy, including: Based on the real-time physiological parameters, the real-time environmental parameters and the real-time device status parameters, a human-machine-environment integrated state perception matrix is ​​constructed; The state perception matrix is ​​input into a preset strategy optimization model, and the dynamic control strategy is corrected in real time through preset adaptive weight adjustment and multi-objective optimization to generate the optimal adaptive control strategy.

7. The method according to claim 1, characterized in that The controlling of the ventilator based on the adaptive control strategy comprises: According to the target air supply volume and target air supply frequency of the adaptive control strategy, the speed of the fan and the opening of the air supply valve are controlled by using a PID control algorithm; According to the filtering level instruction in the adaptive control strategy, the filtering wind resistance of the filter is adjusted in real time; According to the target safety control strategy in the adaptive control strategy, when the protection performance drops to a preset threshold, a safety warning is issued.

8. A ventilator adaptive control system, characterized in that: The system comprises: A data acquisition module, used to acquire physiological parameters and environmental parameters of the respirator wearer and initial operating state data of the respirator, and to perform a human-machine state assessment based on the physiological parameters and the environmental parameters to determine a human-machine state assessment result; An initial strategy generation module, used to determine an initial control strategy of the ventilator based on the human-machine state evaluation result and the initial operation state data of the ventilator; An optimization module, used for acquiring historical data during the use of the ventilator, the historical data including historical physiological feedback data and ventilator operation status data, and adaptively adjusting the initial control strategy based on the historical data to obtain an optimized dynamic control strategy; An adjustment module, used for acquiring multi-source sensor data of the respirator in real time, and adjusting the dynamic control strategy in real time based on the multi-source sensor data to obtain an adaptive control strategy; A control module is used to control the ventilator based on the adaptive control strategy.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: It includes a processor, a memory and a transceiver, the memory is used to store instructions, the transceiver is used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes the method according to any one of claims 1 to 7.

Images