Patent Application
20250222283
The invention belongs to the technical field of respirators, in particular to a method and system for controlling the pressure in a mask, and a respirator.
Respirators, as protective equipment, adopt a filter cartridge to filter air inhaled by users to prevent harmful substances from entering the respiratory tract. The filter cartridges, as the key component of the respirators, are made from different materials, adopt different techniques and are used for filtering out particulate matter, gas or vapor in air. Different filter cartridges can selectively filter out different pollutants according to different designs and service environments of equipment. Respirators are widely used for, for example, mining, stone and wood processing, agricultural production, loading, unloading and transportation in freight yards at the quay, and other scenarios, and are also used by various processing enterprises, chemical manufacturers, laboratories, etc.
Different types of filter cartridges generally adopt different filter materials which are different in breathability. Some efficient filter materials may produce a greater resistance against the air flow, and users, when breathing, need to overcome such a resistance. In addition, the design of the filter cartridges also has an influence on the respiratory resistance, and some more compact and denser filter cartridges may lead to a larger respiratory resistance.
In use of the respirators, particulate matter or gas will gradually accumulate on the filter cartridge with time, leading to a gradual increase in the respiratory resistance; and when the filter cartridge is saturated to some extent with the accumulation of the particulate matter or gas, it needs to be replaced with a new one, otherwise users will suffer from a greater respiratory resistance.
How to solve the problems caused by the respiratory resistance of existing respirator masks becomes a technical problem to be solved.
The objective of the invention is to provide a method and system for controlling the pressure in a mask, and a respirator to effectively solve the problems mentioned above.
To fulfill the above objective, the technical solutions adopted by the invention are as follows:
A method for controlling the pressure in a mask comprises:
Further, the step of determining a correlation between a fan speed, a pressure in the mask and respiratory behaviors of a user comprises:
Further, the reference air pressure is input to a PID controller, and with the reference air pressure as a set point, the PID controller controls the fan speed according to a deviation of a currently acquired air pressure from the set point to reduce the deviation.
Further, a control model of the PID controller is:
Further, a method for determining the fan speed correction function ƒ(t) comprises:
Further, the step of performing feature extraction based on the motion data, the speed data and the air pressure data, and establishing the fan speed correction function ƒ(t) based on extracted features by means of a machine learning algorithm comprises:
Further, the step of determining a correlation between a fan speed, a pressure in the mask and respiratory behaviors of a user comprises:
Further, the neural network model comprises:
A system for controlling the pressure in a mask uses the method for controlling the pressure in a mask and comprises:
A respirator comprises a mask and the system for controlling the pressure in a mask, wherein a pressure in the mask is controlled by the system for controlling the pressure in a mask.
The technical solutions of the invention fulfill the following beneficial effects:
In the invention, the fan, after being started, continuously supplies air into the mask, the fan speed is dynamically controlled according to the acquired pressure and the determined correlation, and the system can control the air supply according to real-time requirements of users to minimum the respiratory resistance. The method for controlling the pressure in a mask provides an individualized and real-time ventilation solution, which allows for an adjustment according to specific respiratory requirements of users to effectively deal with the respiratory resistance.
To better clarify the technical solutions in the embodiments of the invention or the prior art, drawings used for describing the embodiments of the invention or the prior art are briefly introduced below. Obviously, the following drawings only illustrate some embodiments of the invention, and those ordinarily skilled in the art can obtain other drawings according to the following ones without creative labor.
The technical solutions in the embodiments of the invention are clearly and completely described below in conjunction with accompanying drawings of these embodiments. Obviously, the embodiments in the following description are merely illustrative ones, and are not all possible ones of the invention.
Unless otherwise defined, all technical and scientific terms used here have the same meanings as commonly understood by those skilled in the art. The terms used here are merely for the purpose of describing specific embodiments and are not intended to limit the invention. The term “and/or” used here indicates the inclusion of any or all combinations of one or more related items listed.
A method for controlling the pressure in a mask, as shown in
In the invention, the fan is arranged on the mask to draw air into the mask from the outside to reduce the respiratory resistance produced by a filter cartridge; the fan, after being started, continuously supplies air into the mask to ensure that air flows in the mask in the respiration process of users, so as to maintain ventilation. In the actual respiration process, the fan speed is dynamically controlled according to the acquired pressure and the determined correlation, and the system can control the air supply according to actual requirements of users to minimize the respiratory resistance. The method provides an individualized and real-time ventilation solution, which allows for an adjustment according to specific respiratory requirements of users to effectively deal with the respiratory resistance.
In the implementation process, in a case where the correlation needs to be determined depending on the actual operating process of the fan after installation, for example, depending on acquired actual operating data, the sequence of A1 and A2 cannot be changed. In a case where the correlation can be determined independent of the actual operating process, for example, the correlation can be directly output by means a set algorithm after related parameters of the fan and the mask are input, the sequence of A1 and A2 can be changed, or A1 and A2 can be performed separately.
Preferably, as shown in
By adopting this optimized scheme, a system can control the fan speed more accurately according to the respiratory pattern of the user to provide more natural and comfortable respiration experience, and can more effectively reduce the respiratory resistance to allow users to use a respirator more comfortably.
In this preferred scheme, by setting the reference air pressure, the system can more flexibly satisfy the requirements of different users. Because different users may have different normal respiratory patterns, the reference air pressure is set to make the system more individualized and adapt to different physiological differences. By setting the reference air pressure, the system can adapt to changes of the respiratory requirements of users in different working environments and on different activity levels, for example, when the respiratory pattern of a user during exercise may be different from the respiratory pattern of the user in a static state, and in this case, the reference air pressure can be set to allow the system to better adapt to these changes.
Preferably, the reference air pressure is input to a PID controller, and with the reference air pressure as a set point, the PID controller controls the fan speed according to a deviation of a currently acquired air pressure from the set point to reduce the deviation.
In the implementation process, as shown in
As a specific implementation, a traditional control model of the PID controller is:
In the implementation process, the initial resistance a filter cartridge adopted by a respirator is certain, and the initial resistance of the filter cartridge is set as P3, as shown in
In the specific implementation process, the user of the respirator is often accompanied by a dynamic motion, which surely has a significant influence on the respiratory behaviors. Fully concerning such an influence, a preferred control model of the PID controller is:
In this preferred scheme, when Kf and f(t) are input to the PID controller, the mechanism of dynamic gain is taken into account, wherein f(t) is a dynamic factor related to the user, and the dynamic factor may affect the change of the air pressure in the respirator.
Kf, as an adjustment factor, indicates the intensity of the added dynamic gain, and by adjusting the value of Kf, the sensitivity of the system to the dynamic factor can be controlled, and the sensitivity and stability of the system need to be balanced when the value of Kf is selected. Specifically, in actual use, the value of Kf should be adjusted according to user feedback and system performance. Specifically,
In actual use, an optimal value of Kf may be determined by system simulation, experiments and user feedback to ensure that the sensitivity and stability of the system can be balanced in the control process in all cases to provide comfortable and reliable respiration support for users.
Preferably, as show in
A traditional PID control model is linear and cannot process some nonlinear system responses. In this preferred scheme, the fan speed correction function is introduced to ensure that the system can well adapt to nonlinear relations, thus improving the capacity to model complex relations of the system. In addition, the traditional PID control model generally depends on preset parameters, while the dynamic gain parameter introduced here can be adjusted in operation according to the actual condition, such that the real-time performance and stability are better balanced, particularly in a real-time system such as a respirator.
In the implementation process, the dynamic gain parameter Kf allows for an adjustment of the sensitivity of the system according to real-time motion data, air pressure and other factors, thus improving the adaptability of the system to different operation conditions and ensuring that stable and effective respiration support can be provided in different usage scenarios. The machine learning model can better process uncertain and nonstructured data to deal with various uncertainties that possibly occur in a respirator system, such as sudden changes of the motion pattern of the user or uncertainties of the external environment.
Preferably, the step of performing feature extraction based on the motion data, the speed data and the air pressure data, and establishing the fan speed correction function ƒ(t) based on extracted features by means of a machine learning algorithm, as shown in
In the training process, the train set is used for model training, the test set is used for evaluating model performance, and hyper-parameters of a decision tree can be adjusted to improve the model performance by cross validation or the like; the test set is used for evaluating the model to guarantee the generalization performance of the model; the decision tree model, after being trained and evaluated, is integrated into the system, the real-time performance of the model is tested to ensure that the model satisfies the real-time requirement in actual use, and the performance of the model is monitored regularly.
In this preferred scheme, the reason why the decision tree model is used is that it has a clear structure and is easy to understand and explain, which, in the respirator system, is very important for professionals and terminal users who have a clear understanding of the operating principle of the model; the decision tree can effectively capture and process nonlinear relations, has a good adaptability to the interaction between features, and can well process variable and complex relations between motion data, speed data and air pressure data in the respirator system, and the decision tree can output the importance of features to help determine critical features for predicting the fan speed, which is beneficial to feature selection optimization and model explanation; and the decision tree can process hybrid data, including numeric data and categorical data, which is beneficial for processing different types of data that may be involved in the fan speed correction function.
As another optimized scheme, as shown in
Wherein, preferably, the neural network model comprises:
In an application scenario of a respirator, seasonal changes may lead to changes of respiratory behaviors, for example, the change in temperature and humidity will affect the comfort and resistance of the respiratory tract, and the LSTM layer can learn and capture such seasonal changes to allow the model to better adapt to dynamic changes of the respiratory behaviors in different times. The respiratory resistance in the mask may be affected by multiple factors which have a long-term correlation in time, and the LSRM layer can capture such a correlation to allow the model to better understand and predict long-term dynamic changes of the respiratory resistance in the mask.
After the neural network model is trained, the currently acquired pressure in the mask can be directly input to the neural network model to predict the fan speed; in the model training process, the correlation model of the fan speed, the pressure in the mask and the respiratory behaviors of the user is established, and the model is trained to learn the relationship between the fan speed, the pressure in the mask and the respiratory behaviors of the user; and when the model is used, to reduce the cost of the respirator, only the pressure in the mask is acquired in the implementation process, and because the model has learnt the correlation between the pressure in the mask and the fan speed in the training process and the LSMT layer allows the model to memorize previous information, the model can better adapt to the current input and predict the fan speed according to the correlation learned in the training process.
A system for controlling the pressure in a mask uses the method for controlling the pressure in a mask in Embodiment 1 and comprises:
A respirator comprises a mask and the system for controlling the pressure in a mask in Embodiment 2, wherein the pressure in the mask is controlled by the system for controlling the pressure in a mask.
The technical effects fulfilled by Embodiments 2 and 3 are the same as those fulfilled by Embodiment 1 and will not be repeated here.
The basic principle, main features and advantages of the invention are illustrated and described above. Those skilled in the art should understand that the invention is not limited to the above embodiments, the above embodiments and descriptions are merely used to explain the principle of the invention, various modifications and improvements can be made without departing from the spirit and scope of the invention, and all these modifications and improvement should also fall within the protection scope of the invention. The protection scope of the invention should be defined by the appended claims and their equivalents.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2024100087441 | Jan 2024 | CN | national |
| Number | Date | Country | |
|---|---|---|---|
| Parent | PCT/CN2024/108764 | Jul 2024 | WO |
| Child | 19028989 | US |
