A flow control method and related apparatus

Abstract

This application relates to the field of medical device control technology, and discloses a flow control method and related apparatus. By acquiring initial airway pressure and flow rates, and filtering them, real-time airway pressure and flow rates are obtained, thereby suppressing noise interference in the initial airway pressure and flow rates. A dynamic compensation flow rate is determined based on the real-time airway pressure and flow rates, and a preset target flow rate is added to the dynamic compensation flow rate to obtain an equivalent target flow rate. This compensates for the flow rate deviation caused by load pressure. A target control speed is generated based on the equivalent target flow rate, and the current airway flow rate is adjusted to approach the preset target flow rate, reducing the impact of load pressure on the flow control process. This decoupling of load pressure and flow control improves the accuracy of turbine fan flow control.

Description

Technical Field

[0001] This application relates to the field of medical device control technology, and in particular to a flow control method and related apparatus. Background Technology

[0002] Respiratory therapy equipment refers to medical devices used to provide respiratory support and improve ventilation and oxygenation in patients. It typically relies on adjusting the speed of a turbine fan to provide ventilation power. In actual ventilation, due to differences in the mechanical properties of patients' lungs, and the possibility of spontaneous breathing, coughing, choking, and severe spontaneous breathing during ventilation, the turbine fan of respiratory therapy equipment must possess extremely high transient response capabilities.

[0003] In related technologies, traditional control strategies often employ proportional-integral-derivative (PID) controllers. These controllers adjust the airflow rate based on the difference between the current airflow rate and a preset target flow rate. However, pressure disturbances in the airflow prevent this adjustment method from quickly and accurately reaching the target flow rate. Even when fuzzy PID or adaptive control is used to optimize the adjustment parameters and dynamically adjust them according to the system's operating state to improve the response speed and stability of traditional PID control under changing operating conditions, it remains essentially limited to feedback control based on existing deviations. This makes it difficult to address the impact of complex pressure changes in the ventilation system on the actual output flow of the turbine blower, hindering the rapid and accurate achievement of the target flow rate. Consequently, it is difficult to effectively reduce the turbine blower's sluggish air supply response caused by pressure disturbances, easily leading to over-ventilation or under-ventilation in the airflow.

[0004] Therefore, how to reduce the lag in the air supply response of turbine fans is a technical problem that urgently needs to be solved. Summary of the Invention

[0005] To address the aforementioned technical problems, this application provides a flow control method and related apparatus.

[0006] The embodiments of this application disclose the following technical solutions: On one hand, embodiments of this application provide a flow control method, the method comprising: Obtain the initial airway pressure and initial airway flow values ​​within a target time period, wherein the target time period is a time period preceding the current time and whose length is less than a time length threshold; The initial airway pressure value is filtered to obtain the real-time airway pressure value, and the initial airway flow rate value is filtered to obtain the real-time airway flow rate value. A dynamic compensation flow rate value is determined based on the real-time airway pressure value and the real-time airway flow rate value; the dynamic compensation flow rate value is used to offset the flow rate value deviation caused by changes in load pressure. The sum of the preset target flow value and the dynamically compensated flow value is determined to be the equivalent target flow value; The target control speed of the turbine fan is generated based on the equivalent target flow rate value; the target control speed is used to adjust the airflow rate value at the current moment.

[0007] On the other hand, embodiments of this application provide a flow control device, the device comprising: The acquisition unit is used to acquire the initial airway pressure value and the initial airway flow value within a target time period, wherein the target time period is a time period located before the current time and whose length is less than a time length threshold. The filtering unit is used to filter the initial airway pressure value to obtain the real-time airway pressure value and to filter the initial airway flow value to obtain the real-time airway flow value. A determining unit is configured to determine a dynamic compensation flow value based on the real-time airway pressure value and the real-time airway flow value; the dynamic compensation flow value is used to offset the flow value deviation caused by changes in load pressure. The determining unit is further configured to determine that the sum of the preset target flow value and the dynamically compensated flow value is an equivalent target flow value; The generation unit is used to generate a target control speed of the turbine fan based on the equivalent target flow rate value; the target control speed is used to adjust the airflow rate value at the current moment.

[0008] Optionally, the target time period includes a first sliding filter window and a second sliding filter window, the initial airway pressure value includes a first initial airway pressure value and a second initial airway pressure value, and the initial airway flow value includes a first initial airway flow value and a second initial airway flow value; the filtering unit is specifically used for: Based on multiple initial airway flow rates within the first sliding filter window, a first average flow rate is calculated, and based on multiple initial airway pressure values ​​within the first sliding filter window, a first average pressure is calculated. A second average flow rate is calculated based on multiple second initial airway flow rates within the second sliding filter window, and a second average pressure is calculated based on multiple second initial airway pressure values ​​within the second sliding filter window; the window length of the first sliding filter window is greater than the window length of the second sliding filter window. The real-time airway flow rate value is obtained by weighting the first average flow rate and the second average flow rate. The real-time airway pressure value is obtained by weighting the first average pressure value and the second average pressure value.

[0009] Optionally, the determining unit is specifically used for: The real-time airway pressure value and the real-time airway flow rate value are input into the dynamic compensation determination model to obtain the dynamic compensation flow rate value.

[0010] Optionally, the determining unit is further configured to: Obtain at least one sample to control the rotation speed; For each of the sample control speeds, at least one load pressure and the first flow rate data corresponding to the load pressure and the second flow rate data corresponding to the state without external pressure disturbance are obtained; The difference between the second flow data and the first flow data is determined as the sample dynamic compensation flow value; Based on the first flow data, the load pressure, and the sample dynamic compensation flow value, the dynamic compensation determination model is determined.

[0011] Optionally, the determining unit is specifically used for: Based on the sample control speed, the reference flow rate speed control model is inversely calculated to determine the second flow rate data corresponding to the sample control speed under the condition of no external pressure disturbance.

[0012] Optionally, the determining unit is further configured to: Obtain at least one reference control speed and the base flow value corresponding to each of the reference control speeds under conditions without external pressure disturbance; Based on the correspondence between the reference control speed and the basic flow rate value, a reference flow rate speed control model is established.

[0013] Optionally, the determining unit is specifically used for: Under each of the sample control speeds, at least one of the load pressures corresponding to multiple initial sample output flow values ​​is collected. The average flow rate and the standard deviation of the flow rate are determined based on multiple initial sample output flow rate values, and initial sample output flow rate values ​​whose absolute value of the difference between them and the average flow rate is greater than three times the standard deviation of the flow rate are eliminated. Kernel density estimation is performed on the remaining initial sample output flow values ​​to obtain the flow probability density; The flow value corresponding to the point where the flow probability density is maximum is determined to be the first flow data.

[0014] Optionally, the generating unit is specifically used for: The equivalent target flow rate value is input into the reference flow rate speed control model, and the target control speed is generated based on the reference flow rate speed control model.

[0015] On the other hand, embodiments of this application provide a computer device, the computer device including a processor and a memory: The memory is used to store computer programs; The processor is configured to perform the methods described above according to the computer program.

[0016] As described in the above technical solution, this application first obtains the initial airway pressure and initial airway flow values ​​to obtain the real-time changes in the airway pressure and flow values ​​at the current moment. Subsequently, the initial airway pressure and flow values ​​are filtered to obtain real-time airway pressure and flow values. This process suppresses sudden changes in airway flow and pressure values ​​caused by sensor noise or signal interference, while retaining the real-time trend of airway pressure and flow values ​​changing with airway operating conditions. This improves the timeliness and accuracy of identifying pressure changes and flow value deviations, avoiding jumps in the dynamic compensation flow value caused by instantaneous fluctuations in the initial airway pressure or flow value, thereby preventing turbine fan target control speed fluctuations due to jumps in the dynamic compensation flow value. Based on this, a dynamic compensation flow value is determined based on the real-time airway pressure and flow values ​​to pre-convert flow value deviations that may be caused by changes in current load pressure into corresponding flow compensation amounts, incorporating flow value deviations caused by load pressure changes into the subsequent flow regulation process. Furthermore, the preset target flow rate value is added to the dynamically compensated flow rate value to obtain an equivalent target flow rate value, thereby offsetting the flow rate deviation caused by load pressure changes. Finally, a target control speed is generated based on the equivalent target flow rate value, and the current airway flow rate value is adjusted using the target control speed. This converts the target flow rate demand under load pressure into the target flow rate demand under conditions without external pressure disturbances, facilitating the rapid calculation of the target control speed. Therefore, this application improves the accuracy of real-time airway pressure and flow rate values ​​through filtering, and determines the dynamically compensated flow rate value based on these two values. It incorporates the flow rate deviation caused by pressure disturbances into the determination process of the equivalent target flow rate value and the target control speed in advance, thus compensating for the impact of pressure disturbances on airway flow rate during the target control speed determination process. This offsets the turbine fan flow control result deviation caused by pressure disturbances and achieves decoupling between pressure disturbances and the turbine fan flow control process. This avoids the lag in turbine air supply response caused by pressure disturbances in the airway, improves the turbine's response speed to changes in airway status, and suppresses the jump in dynamic compensation flow value caused by fluctuations in airway flow detection results, thereby preventing turbine target control speed jitter and ensuring the stability of the respiratory therapy equipment system. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic flowchart of a flow control method provided in an embodiment of this application; Figure 2 A pressure-flow characteristic curve of a turbine fan duct is provided for an embodiment of this application; Figure 3 This application provides a dynamic compensation method for determining a model space surface plot. Figure 4 This application provides a schematic diagram of a flow control process for a respiratory therapy device. Figure 5 A schematic diagram of the structure and control flow of a respiratory therapy device system provided in this application embodiment; Figure 6 This is a schematic diagram of a flow control device provided in an embodiment of this application. Detailed Implementation

[0019] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0020] When a patient's inspiratory demand suddenly increases, a slow response in adjusting the turbine speed can easily lead to insufficient ventilation as airway pressure cannot be maintained at the target level in time. Conversely, when a patient coughs or switches from inspiratory to expiratory states, a failure to reduce the turbine speed promptly can result in excessively high airway pressure, causing expiratory obstruction. Therefore, precise control of the turbine's target speed is necessary to achieve rapid and accurate regulation of airway flow.

[0021] To address the issue of response lag in adjusting turbine blower output flow rate using feedback-based control methods during ventilation in existing respiratory therapy equipment due to airway pressure disturbances, and to reduce the impact of fluctuations in airway flow rate detection results on the accuracy of airway flow rate control, this application provides a flow control method. This method determines flow compensation requirements based on the current airway pressure and flow rate, and adjusts the turbine blower's operating state based on the compensated equivalent target flow rate, enabling the turbine blower to respond proactively to changes in airway pressure. Compared to feedback control that relies on error accumulation, this method proactively adjusts the turbine blower's flow output without waiting for flow deviations to occur. This eliminates lag in air supply response and avoids overshoot oscillations, significantly improving the turbine blower's ability to respond to sudden changes in patient breathing, effectively reducing the risk of airway pressure collapse or overshoot, and improving patient-ventilator synchrony. Therefore, the method provided in this application effectively reduces the impact of pressure disturbances on turbine blower output flow rate, suppresses the response lag problem in turbine blower output flow rate adjustment, and improves the accuracy and stability of turbine blower output flow rate.

[0022] The embodiments of this application will now be described with reference to the accompanying drawings.

[0023] Please see Figure 1 , Figure 1 This is a schematic diagram of a flow control method provided in an embodiment of this application.

[0024] S101: Obtain the initial airway pressure and initial airway flow values ​​within the target time period. The target time period is a time period before the current time and whose length is less than the time length threshold.

[0025] In this embodiment of the application, by acquiring the initial airway pressure value and the initial airway flow value within the target time period in real time, the current airway pressure state and the current airway flow state can be acquired in real time. This allows for the determination of flow control requirements in advance based on changes in the airway state and the adjustment of the turbine fan operating state, thereby enabling the current airway flow value to quickly approach the preset target flow value.

[0026] Because measurement noise and signal interference may be introduced during the acquisition of initial airway pressure and initial airway flow values, they cannot be directly used for the rapid and accurate determination of dynamic compensation flow values. Therefore, by acquiring multiple initial airway pressure and initial airway flow values ​​within a certain time range adjacent to the current time, prior to the current time, airway state information corresponding to multiple sampling times can be integrated. This reduces the error in judging the change in airway state at the current time caused by measurement noise and signal interference when only a single initial airway pressure and initial airway flow value are collected. In this embodiment, the target time period refers to a time range before the current time that is shorter than a time length threshold. In one possible implementation, the target time period can be a time period adjacent to the current time. The time length threshold is a pre-set time parameter used to limit the length of the target time period. This embodiment does not limit its value and can be adaptively adjusted according to the actual usage scenario and the operating conditions of the respiratory therapy equipment.

[0027] In this embodiment, the initial airway pressure value and initial airway flow rate value refer to the actual airway pressure data and airway flow rate data collected within a target time period. The initial airway pressure value represents the load pressure corresponding to the turbine fan output flow rate during the current ventilation process, and the initial airway flow rate value represents the airway flow rate state actually formed by the turbine fan output under that load pressure state. In this embodiment, the initial airway pressure value and initial airway flow rate value can be obtained from a target time period prior to the current moment. The length of the target time period is less than a time length threshold to avoid introducing historical data from a distant time period that could affect the current airway status judgment, thereby accurately determining the pressure and flow rate changes during the current airway status change process.

[0028] S102: Filter the initial airway pressure value to obtain the real-time airway pressure value, and filter the initial airway flow value to obtain the real-time airway flow value.

[0029] In this embodiment, the airflow is output through a turbine fan. The current airflow and pressure values ​​change with the turbine fan's operating status and the airflow pressure load. Since the initial airflow and pressure values ​​may contain random errors—for example, high-frequency measurement noise and signal interference may be introduced during the acquisition of these values ​​using flow and pressure sensors—these random errors could cause abrupt changes in the dynamic compensation flow rate, leading to fluctuations in the turbine fan's target control speed and affecting the stability of the determined flow rate. Therefore, it is necessary to filter the initial airflow and pressure values ​​to obtain the real-time flow rate and pressure values, respectively. This reduces the impact of noise and signal interference on the dynamic compensation flow rate determination process, suppressing abnormal fluctuations while preserving the airflow condition trends reflected by the initial pressure and flow values, thus improving the stability of the determined flow rate. Among them, the real-time airway pressure value is used to represent the current airway pressure status, and the real-time airway flow value is used to represent the current airway flow status. Together, they are used to determine the impact of the current load pressure change on the airway flow.

[0030] S103: Determine the dynamic compensation flow value based on the real-time airway pressure value and the real-time airway flow value; the dynamic compensation flow value is used to offset the flow value deviation caused by changes in load pressure.

[0031] In this embodiment, the dynamic compensation flow rate is the flow rate deviation caused by load pressure disturbances in the control system of the respiratory therapy device under the same control speed, i.e., the flow rate compensation requirement determined based on the current airway pressure state and the current airway flow rate state. Since the theoretical difference between the real-time airway flow rate and the preset target flow rate cannot offset the flow rate deviation caused by load pressure changes when there is a pressure disturbance in the airway, adjusting the flow rate based solely on this theoretical difference cannot quickly and accurately bring the current airway flow rate close to the preset target flow rate. Therefore, the dynamic compensation flow rate is calculated as the difference between the airway flow rate when there is a pressure disturbance and the airway flow rate when there is no pressure disturbance, under the same control speed. This dynamic compensation flow rate offsets the flow rate deviation caused by load pressure changes. Since different airway pressure states correspond to different actual output flow rates when the turbine fan control speed remains constant, the dynamic compensation flow rate can determine the amount of flow rate that needs to be compensated under the current airway state. Therefore, by using real-time airway pressure and real-time airway flow to determine the dynamic compensation flow value, it is possible to determine the flow compensation requirement corresponding to load pressure changes in advance during the flow regulation process, and incorporate the flow value deviation caused by load pressure changes into the flow adjustment process.

[0032] S104: Determine the sum of the preset target flow value and the dynamic compensation flow value as the equivalent target flow value.

[0033] In this embodiment, the preset target flow rate value represents the target ventilation flow rate that the respiratory therapy device expects to achieve; the equivalent target flow rate value represents the flow rate value used to control the airway flow rate at the current moment to reach the preset target flow rate value after considering the flow rate deviation caused by changes in the current load pressure. Please refer to [link to relevant documentation]. Figure 2 , Figure 2 This application provides a turbine fan duct pressure-flow characteristic curve. Since the turbine fan output duct flow rate gradually decreases as the duct pressure gradually increases when the turbine fan control speed remains constant, i.e., the duct pressure and duct flow rate show a negative correlation trend, after superimposing the dynamically compensated flow rate value onto the preset target flow rate value, the target flow rate requirement under load pressure influence can be converted into the target flow rate requirement under the state without external pressure disturbance. This allows the target control speed determined based on the equivalent target flow rate value to take into account the impact of load pressure changes on the output flow rate in advance.

[0034] As one possible implementation, the equivalent target flow value can be determined using the following formula: , in, Represents the equivalent target flow value. This indicates the preset target traffic value. This represents the dynamically compensated flow value.

[0035] Therefore, this application's embodiments determine a dynamic compensation flow rate value based on real-time airway pressure and flow rate values, and then superimpose the dynamic compensation flow rate value with a preset target flow rate value to obtain an equivalent target flow rate value. Compared to related technologies that only adjust based on target flow rate deviation, this application utilizes real-time airway pressure change information to pre-determine the corresponding dynamic compensation flow rate value and applies it to the target flow rate adjustment process. This avoids the impact of feedback adjustment lag on airway flow control, improves the turbine fan's response speed to changes in airway status, and thus improves the human-machine interface.

[0036] S105: Generate the target control speed of the turbine fan based on the equivalent target flow rate; the target control speed is used to adjust the airflow rate at the current moment.

[0037] In this embodiment, the turbine fan is the gas supply component in the respiratory therapy device, and its rotational speed affects the gas output flow rate. The target control speed is a command used to control the turbine fan's rotational speed. The corresponding target control speed is determined by the equivalent target flow rate value, and the turbine fan is controlled to operate at the target control speed to adjust the airway flow rate at the current moment. Generating the target control speed based on the equivalent target flow rate value allows the turbine fan's speed adjustment process to pre-consider dynamically compensating for flow rate deviations, avoiding response lag during the current airway flow rate adjustment process. This ensures that the current airway flow rate quickly and accurately approaches the preset target flow rate value, guaranteeing the accuracy of flow control in the airway.

[0038] This application first acquires initial airway pressure and flow rates to obtain the real-time changes in these values. Then, the initial airway pressure and flow rates are filtered to obtain real-time airway pressure and flow rates. This process suppresses sudden changes in airway pressure and flow rates caused by sensor noise or signal interference while preserving the real-time trends of these values ​​as airway conditions change. This improves the timeliness and accuracy of identifying pressure fluctuations and flow rate deviations, preventing jumps in the dynamic compensation flow rate due to instantaneous fluctuations in the initial airway pressure or flow rate, thus preventing turbine fan target speed fluctuations caused by such jumps. Based on this, a dynamic compensation flow rate is determined using the real-time airway pressure and flow rates to pre-convert potential flow rate deviations caused by changes in current load pressure into corresponding flow compensation values, incorporating these deviations into the subsequent flow regulation process. Furthermore, the preset target flow rate value is added to the dynamically compensated flow rate value to obtain an equivalent target flow rate value, thereby offsetting the flow rate deviation caused by load pressure changes. Finally, a target control speed is generated based on the equivalent target flow rate value, and the current airway flow rate value is adjusted using the target control speed. This converts the target flow rate demand under load pressure into the target flow rate demand under conditions without external pressure disturbances, facilitating the rapid calculation of the target control speed. Therefore, this application improves the accuracy of real-time airway pressure and flow rate values ​​through filtering, and determines the dynamically compensated flow rate value based on these two values. It incorporates the flow rate deviation caused by pressure disturbances into the determination process of the equivalent target flow rate value and the target control speed in advance, thus compensating for the impact of pressure disturbances on airway flow rate during the target control speed determination process. This offsets the turbine fan flow control result deviation caused by pressure disturbances and achieves decoupling between pressure disturbances and the turbine fan flow control process. Therefore, this avoids the lag in turbine air supply response caused by pressure disturbances in the airway, improves the turbine's response speed to changes in airway status, and suppresses the jump in dynamic compensation flow value caused by fluctuations in airway flow value detection results, thereby preventing turbine target control speed jitter and ensuring the stability of the respiratory therapy equipment system.

[0039] To balance filtering stability and airway condition change response capability, dual time-domain filtering can be used to process the initial airway pressure and initial airway flow values. This reduces the impact of fluctuations, preserves the trend of airway condition changes, and improves the accuracy of real-time airway pressure and real-time airway flow values ​​in reflecting the current airway condition. As one possible implementation, the target time period includes a first sliding filter window and a second sliding filter window. The initial airway pressure value includes a first initial airway pressure value and a second initial airway pressure value. The initial airway flow value includes a first initial airway flow value and a second initial airway flow value. The initial airway pressure value is filtered to obtain a real-time airway pressure value. The initial airway flow value is filtered to obtain a real-time airway flow value. This includes: calculating a first average flow rate based on multiple first initial airway flow values ​​within the first sliding filter window, and calculating a first average pressure based on multiple first initial airway pressure values ​​within the first sliding filter window; calculating a second average flow rate based on multiple second initial airway flow values ​​within the second sliding filter window, and calculating a second average pressure based on multiple second initial airway pressure values ​​within the second sliding filter window; the window length of the first sliding filter window is greater than the window length of the second sliding filter window; a weighted average calculation is performed on the first average flow rate and the second average flow rate to obtain a real-time airway flow value; and a weighted average calculation is performed on the first average pressure and the second average pressure to obtain a real-time airway pressure value.

[0040] In this embodiment, the first sliding filter window is a long-term sliding filter window in the dual-time-domain filtering process, used to process the initial airway flow rate and initial airway pressure values ​​collected over a relatively long time range to improve the stability of the filtering results. Based on multiple initial airway flow rate values ​​within the first sliding filter window, a first average flow rate is calculated, and based on multiple initial airway pressure values ​​within the first sliding filter window, a first average pressure is calculated. The first average flow rate represents the average airway flow rate state within the corresponding time range of the long-term sliding filter window, and the first average pressure represents the average airway pressure state within the corresponding time range of the long-term sliding filter window.

[0041] In this embodiment, the methods for calculating the first average flow rate and the first average pressure are not limited. As one possible implementation, the following formula can be used when calculating the first average flow rate: ,in, The first average flow rate is given; the following formula can be used to calculate the first average pressure: ,in, The average value of the first pressure. m The length of the first sliding filter window is determined by... To obtain multiple initial airway flow values ​​within the first sliding filter window through real-time high-frequency sampling, To obtain multiple initial airway pressure values ​​within the first sliding filter window through real-time high-frequency sampling.

[0042] In this embodiment, the second sliding filter window is a short-time sliding filter window in the dual-time-domain filtering process. It is used to process the initial airway flow rate and initial airway pressure values ​​collected within a shorter time range to improve the responsiveness of the filtering results to changes in airway status. Based on multiple initial airway flow rate values ​​within the second sliding filter window, a second average flow rate is calculated, and based on multiple initial airway pressure values ​​within the second sliding filter window, a second average pressure is calculated. The second average flow rate represents the average airway flow rate state within the corresponding time range of the short-time sliding filter window, and the second average pressure represents the average airway pressure state within the corresponding time range of the short-time sliding filter window.

[0043] In this embodiment, the methods for calculating the second average flow rate and the second average pressure are not limited. As one possible implementation, the following formula can be used when calculating the second average flow rate: ,in, The second average flow rate is given; the second average pressure can be calculated using the following formula: ,in, The second average pressure value. n The length of the second sliding filter window. These are multiple initial airway flow values ​​within the second sliding filter window obtained through real-time high-frequency sampling. These are multiple initial airway pressure values ​​within the second sliding filter window obtained through real-time high-frequency sampling.

[0044] In this application embodiment, the sampling frequency of high-frequency sampling is not limited. As one possible implementation, the sampling frequency of high-frequency sampling can be 100Hz~10kHz.

[0045] In this embodiment, the length of the first sliding filter window is greater than the length of the second sliding filter window, and the lengths of both the first and second sliding filter windows are less than or equal to a time length threshold. This embodiment does not limit the magnitude of the length difference between the first and second sliding filter windows; preferably, the window length of the first sliding filter window... m Window length greater than or equal to the second sliding filter window n 2 times.

[0046] In this embodiment, the average first pressure and the average second pressure are weighted to obtain the real-time airway pressure value. This allows the real-time airway flow rate and real-time airway pressure value to integrate airway status information across different time ranges, taking into account both the stability and responsiveness to changes in the real-time airway pressure and flow rate values.

[0047] In this embodiment, the weighted calculation method for obtaining the real-time airway flow rate and the real-time airway pressure is not limited. As one possible implementation, the following formula can be used for weighted fusion calculation: , in, and These are the filter weighting coefficients. Used to adjust the fusion ratio between the first average flow rate and the second average flow rate. Used to adjust the fusion ratio between the first and second average pressure values. By adjusting and It can adjust the deviation between the data corresponding to the long-term sliding filter window and the data corresponding to the short-term sliding filter window, reduce the impact of sudden data on real-time airway flow and real-time airway pressure values, and reduce abnormal speed adjustment caused by sudden changes in flow and pressure.

[0048] In this embodiment, since the short-term data corresponding to the second sliding filter window is used to reflect the current airway status change trend, while the long-term data corresponding to the first sliding filter window can reduce the impact of short-term fluctuations, there may be differences between the first average flow rate and the second average flow rate, and between the first average pressure and the second average pressure. To allow real-time airway flow rate and real-time airway pressure values ​​to be corrected according to different change trends, when the first average flow rate is greater than the second average flow rate, it indicates that the current airway flow rate is lower than the airway flow rate over a longer period. In this case, the real-time airway flow rate is obtained by subtracting a portion of the difference between the first and second average flow rates from the first average flow rate, i.e., using the formula... Calculate the real-time airway flow rate. When the first average flow rate is less than or equal to the second average flow rate, it indicates that the current airway flow rate is higher than the airway flow rate over a longer period. In this case, the real-time airway flow rate is obtained by adding a portion of the difference between the first and second average flow rates to the first average flow rate. Calculate the real-time airway flow rate. For the real-time airway pressure, for similar reasons, the first average pressure is corrected according to the relationship between the first and second average pressures, and then the real-time airway pressure is calculated using the different formulas mentioned above.

[0049] Therefore, this embodiment performs dual-time-domain filtering on the initial airway pressure and initial airway flow values ​​based on a first and a second sliding filter window, and weights and fuses the mean values ​​corresponding to different time ranges. This allows for the suppression of abnormal fluctuations in the initial airway pressure and initial airway flow values ​​using data from a longer time range, while retaining the trend of rapid changes in airway status by combining data from a shorter time range, thus improving the response speed to changes in airway status. While reducing noise interference in the initial airway pressure and initial airway flow values, the resulting real-time airway pressure and real-time airway flow values ​​more accurately reflect the current airway status, adapting to changes in lung mechanics parameters and airway conditions under different patient pathological states, and improving the stability and accuracy of subsequent dynamic compensation flow value determination.

[0050] In actual ventilation, the actual output flow rate of the turbine fan varies under different control speeds and load pressures. Therefore, in order to accurately determine the dynamic compensation flow rate value, as a possible implementation method, the dynamic compensation flow rate value is determined based on the real-time airway pressure value and the real-time airway flow rate value. This includes: inputting the real-time airway pressure value and the real-time airway flow rate value into the dynamic compensation determination model to obtain the dynamic compensation flow rate value.

[0051] In this embodiment, the dynamic compensation determination model refers to a model used to determine the dynamic compensation flow rate value based on the real-time airway pressure and real-time airway flow rate values. After determining the real-time airway pressure and real-time airway flow rate values, these values ​​are input into the dynamic compensation determination model to obtain the dynamic compensation flow rate value. Please refer to [link to relevant documentation]. Figure 3 , Figure 3 This is a dynamic compensation determination model space surface diagram provided in an embodiment of this application. For example... Figure 3 As shown, the dynamic compensation determination model takes real-time airflow and real-time air pressure values ​​as inputs and the dynamically compensated flow rate value as the output. Under the same control speed, changes in real-time air pressure will cause changes in the actual output flow rate of the turbine fan, resulting in airflow deviation due to load pressure changes. For example, when the real-time air pressure increases, the corresponding dynamically compensated flow rate value generally shows an increasing trend to compensate for the decrease in airflow caused by the increase in real-time air pressure.

[0052] Therefore, by using a dynamic compensation determination model to determine the corresponding dynamic compensation flow rate value based on the real-time airway status during ventilation, the influence of real-time airway pressure and real-time airway flow rate on the compensation result can be comprehensively considered, thereby improving the accuracy and timeliness of the dynamic compensation flow rate value determination and effectively improving human-machine synchronization.

[0053] To establish a dynamic compensation determination model that reflects the relationship between control speed, load pressure, and flow rate changes, as a possible implementation method, the construction method of the dynamic compensation determination model includes: acquiring at least one sample control speed; acquiring at least one load pressure and corresponding first flow rate data and second flow rate data under each sample control speed, and the corresponding second flow rate data under the state without external pressure disturbance; determining the difference between the second flow rate data and the first flow rate data as the sample dynamic compensation flow rate value; and determining the dynamic compensation determination model based on the first flow rate data, load pressure, and sample dynamic compensation flow rate value.

[0054] In this embodiment, the sample control speed refers to the speed control command of the turbine fan when acquiring the first flow rate data and the second flow rate data; the load pressure refers to the pressure in the air duct when acquiring the first flow rate data; the first flow rate data refers to the flow rate data acquired under the sample control speed and load pressure conditions, i.e., the flow rate data acquired when there is an external pressure disturbance; the second flow rate data refers to the basic air duct flow rate value under the same sample control speed conditions but without external pressure disturbance. Since the actual output flow rate of the turbine fan differs under different load pressure states at the same control speed, it is necessary to acquire at least one first flow rate data corresponding to a load pressure to obtain the correspondence between different load pressures and the first flow rate data.

[0055] In this embodiment, a controlled experimental environment can be constructed to obtain the flow rate changes under different load pressure conditions. One possible implementation is to disconnect the expiratory circuit to open the airway, set different turbine fan speeds, measure the steady-state output flow rate at the corresponding turbine fan speed, and then obtain the baseline airway flow rate values ​​at different control speeds under conditions without external pressure disturbances. Combined with the flow rate data obtained under conditions with external pressure disturbances, the impact of different load pressures on airway flow rate can be determined.

[0056] In this embodiment of the application, since the load pressure change will cause the actual output flow rate value to change compared with the basic airway flow rate value under the same control speed, the difference between the second flow rate data and the first flow rate data is determined as the sample dynamic compensation flow rate value based on the difference between the first flow rate data obtained under the condition of external pressure disturbance and the second flow rate data obtained under the condition of no external pressure disturbance.

[0057] As one possible implementation, the sample dynamic compensation flow value can be determined using the following formula: ( , ) = - , in, ( , ) This represents the sample's dynamic compensation flow value. Indicates load pressure. Indicates control of rotational speed. This indicates that under conditions without external pressure disturbance, the speed is controlled according to the speed. ω Determined second flow data, This represents the first flow data acquired under conditions of external pressure disturbance.

[0058] In this embodiment of the application, in order to establish a dynamic compensation determination model, the first flow data and load pressure are used as input parameters of the dynamic compensation determination model, and the sample dynamic compensation flow value is used as the output parameter of the dynamic compensation determination model. The correspondence between the first flow data, load pressure and sample dynamic compensation flow value is established, thereby obtaining the dynamic compensation determination model.

[0059] As one possible implementation, the dynamic compensation deterministic model can be expressed as: = F_ff( , ) , in, This represents the sample's dynamic compensation flow value. This represents the first flow data. This indicates the load pressure.

[0060] In this embodiment, the specific implementation of the dynamic compensation determination model is not limited, as long as it can determine the corresponding dynamic compensation flow rate value based on the real-time airway pressure and real-time airway flow rate values. As one possible implementation, the dynamic compensation determination model can be implemented using an analytical function obtained through polynomial fitting, a nonlinear mapping relationship achieved through table lookup and interpolation algorithms, or a neural network trained on offline data.

[0061] Therefore, this embodiment of the application determines the dynamic compensation determination model based on the first flow rate data, load pressure, and sample dynamic compensation flow rate value. This enables the determination of the corresponding sample dynamic compensation flow rate value based on the first flow rate data and load pressure, improving the matching degree of the dynamic compensation determination model to the current airway state when determining the dynamic compensation flow rate value, thereby improving the accuracy of the dynamic compensation flow rate value determination. Furthermore, since the dynamic compensation determination model is established before the actual turbine fan output flow control process in the airway, there is no need to rely on complex online parameter identification during the actual turbine fan output flow control process. This reduces the computational complexity of the dynamic compensation determination model, lowers the microprocessor computing power requirements, and is suitable for resource-constrained embedded system deployments.

[0062] As one possible implementation method, the acquisition of the second flow rate data corresponding to the state without external pressure disturbance includes: performing an inverse operation on the reference flow rate speed control model based on the sample control speed to determine the second flow rate data corresponding to the sample control speed under the state without external pressure disturbance.

[0063] In this embodiment, the reference flow rate and speed control model is used to represent the relationship between the control speed and output flow rate of the turbine fan under pressure disturbance-free conditions. As one possible implementation, the reference flow rate and speed control model can be expressed as: = f_base( ) , in, Indicates control of rotational speed. This represents the baseline airway flow rate under conditions of no external pressure disturbance.

[0064] Based on this, the inverse operation of the reference flow rate and speed control model can be expressed as: = f_base -1 ( ) , Therefore, by taking the sample control speed as input and performing an inverse operation on the reference flow-speed control model, the corresponding second flow data can be obtained. Thus, this embodiment of the application, based on the control speed, performs an inverse operation on the reference flow-speed control model to determine the second flow data corresponding to the control speed under conditions without external pressure disturbance. This allows for the acquisition of a basic airway flow value corresponding to the control speed, providing a reference for subsequently determining the flow change caused by load pressure variations and improving the accuracy of determining the dynamic compensation flow value.

[0065] In this embodiment, the reference flow rate and speed control model can be established in various ways. One possible implementation is that, under conditions without external pressure disturbance, the known correspondence between the control speed and output flow rate can be determined based on the turbine equipment information. Therefore, the reference flow rate and speed control model can be directly established based on this correspondence. To ensure that the established reference flow rate and speed control model more closely reflects the actual operating conditions of the turbine, one possible implementation is a method for constructing the reference flow rate and speed control model, which includes: obtaining at least one reference control speed and the corresponding basic flow rate value for each reference control speed under conditions without external pressure disturbance; and establishing the reference flow rate and speed control model based on the correspondence between the reference control speed and the basic flow rate value.

[0066] In this embodiment, the reference control speed refers to the turbine control speed set when establishing the reference flow rate speed control model. Under a controlled experimental environment, different reference control speeds can be sequentially applied to the turbine system, and the airflow rate can be maintained at a state free from external pressure disturbances under each second sample control command. The turbine system is then subjected to full-condition excitation tests using multiple reference control speeds to obtain the basic flow rate values ​​under different reference control speeds. Thus, the corresponding basic flow rate values ​​under different reference control speeds are obtained. Furthermore, the basic flow rate values ​​can be used as input parameters, and the basic control speed as output parameters to fit multiple basic control speeds and their corresponding basic flow rate values, thereby obtaining the reference flow rate speed control model.

[0067] In this embodiment, the specific implementation of the reference flow rate speed control model is not limited, as long as the corresponding control speed can be determined based on the input flow rate value. As one possible implementation, the reference flow rate speed control model can employ a linear function, piecewise linear function, polynomial function, or a mapping relationship implemented using lookup tables and interpolation algorithms obtained from experimental data fitting.

[0068] Therefore, by obtaining the measured basic flow rate value based on the reference control speed and determining the reference flow rate speed control model based on the correspondence between the two, the control speed corresponding to different flow control requirements can be accurately determined. The reference flow rate speed control model obtained by fitting can better fit the actual working conditions of the turbine fan, thereby providing a stable and unified flow control reference and improving the stability of the current turbine fan flow adjustment process.

[0069] In the actual process of acquiring the first flow rate data, the output flow rate of the turbine fan may fluctuate under the same sample control speed and load pressure conditions. Therefore, the multiple initial sample output flow rate values ​​may contain abnormal data caused by system noise and random fluctuations. Thus, to reduce the impact of abnormal data on the determination of the first flow rate data, statistical processing is performed on the multiple initial sample output flow rate values.

[0070] As one possible implementation, acquiring at least one load pressure and the corresponding first flow rate data for each sample control speed includes: acquiring multiple initial sample output flow rate values ​​corresponding to at least one load pressure for each sample control speed; determining the average flow rate and standard deviation of the flow rate based on the multiple initial sample output flow rate values, and removing initial sample output flow rate values ​​whose absolute difference from the average flow rate is greater than three times the standard deviation of the flow rate; performing kernel density estimation on the remaining initial sample output flow rate values ​​to obtain the flow probability density; and determining the flow rate value corresponding to the maximum flow probability density as the first flow rate data.

[0071] In this embodiment, the initial sample output flow rate value refers to multiple output flow rate data corresponding to at least one load pressure at each sample control speed collected during testing of the turbine fan system under a controlled experimental environment. By setting different load pressures corresponding to the sample control speed, multiple initial sample output flow rate values ​​are determined, forming a raw dataset. This raw dataset can be used to establish a baseline flow rate speed control model and a dynamic compensation determination model.

[0072] As one possible implementation, the sample-controlled rotational speed can be expressed as: The corresponding different load pressures can be expressed as The output flow values ​​of multiple initial samples can be expressed as The original dataset can be represented as { , , } .

[0073] In this embodiment, the method for determining anomaly data for initial sample output flow rates where the absolute value of the difference between the flow rate and the average flow rate is greater than three times the standard deviation of the flow rate is the 3σ criterion anomaly data determination method. Since multiple initial sample output flow rates under the same load pressure and sample control speed conditions typically exhibit a normal distribution around the actual output flow rate state, and data deviating significantly from the average flow rate are more likely to be affected by system noise or random fluctuations, the 3σ criterion is used to determine anomaly data. This allows for accurate determination of anomaly data based on the dispersion of multiple initial sample output flow rates corresponding to the current load pressure and sample control speed.

[0074] In this embodiment, the average flow rate and the standard deviation of the flow rate are determined based on multiple initial sample output flow rate values, and initial sample output flow rate values ​​whose absolute value of the difference between them and the average flow rate is greater than three times the standard deviation of the flow rate are eliminated.

[0075] As one possible implementation, the average flow rate can be used This means that the standard deviation of flow can be expressed as... Therefore, the outlier removal formula can be determined as follows: Among them, the average flow rate The standard deviation of flow rate is used to represent the average state of multiple initial sample output flow rate values ​​under the same load pressure and sample control speed conditions. This is used to represent the dispersion of multiple initial sample output flow values ​​relative to the average flow value. When the initial sample output flow value satisfies the outlier removal formula, the corresponding initial sample output flow value is removed, i.e., outlier removal is performed using the 3σ criterion.

[0076] In this embodiment, after removing the corresponding initial sample output flow values, kernel density estimation is used to statistically analyze the remaining initial sample output flow values ​​to further determine their distribution. Kernel density estimation is a method for determining probability density based on sample data distribution, capable of fitting the continuous distribution corresponding to multiple initial sample output flow values.

[0077] For the initial sample output flow rates after outlier removal using the 3σ criterion, kernel density estimation is used to fit the sample probability density function, obtaining a continuous distribution of the probability of each output flow rate value. The flow rate value corresponding to the position with the highest probability density is selected as the first flow rate data under this operating condition. The fitted sample probability density function can be expressed as: ρ ( ) The first traffic data can be represented as .

[0078] System noise causes random fluctuations in the output flow rate values ​​collected under the same operating conditions around the steady-state value. Since the location with the highest probability density reflects the concentrated distribution area of ​​multiple sample output flow rate values, determining the steady-state value corresponding to the location with the highest probability density reduces the impact of single sampling point deviations on the results, resulting in a preprocessed dataset after eliminating system noise interference. This preprocessed dataset can be represented as follows: { , , } .

[0079] Therefore, the embodiments of this application can reduce the impact of abnormal data caused by system noise and random fluctuations on the determination result of the first flow rate by obtaining multiple initial sample output flow rate values ​​corresponding to at least one load pressure under each sample control speed and using the 3σ criterion and sample probability density fitting method to determine the corresponding first flow rate data, thereby improving the accuracy of representing the output flow rate state under the corresponding sample control speed and load pressure based on the first flow rate data.

[0080] As one possible implementation, generating the target control speed based on the equivalent target flow rate value includes: inputting the equivalent target flow rate value into a reference flow rate speed control model, and generating the target control speed based on the reference flow rate speed control model.

[0081] In this embodiment of the application, since the reference flow rate speed control model stores the correspondence between the flow rate value output by the turbine fan and its control speed under the condition of no pressure disturbance, the reference flow rate speed control model used to represent the correspondence can be directly determined based on this existing correspondence. The equivalent target flow rate value is used as the input parameter, and the target control speed matching the equivalent target flow rate value is determined according to the correspondence. Then, the airway flow rate value at the current moment is adjusted using the target control speed.

[0082] As one possible implementation, the formula for calculating the target control speed by substituting the equivalent target flow rate into the reference flow rate speed control model can be expressed as: = f_base ( ） , in, Indicates the target control speed. This represents the equivalent target flow value.

[0083] Furthermore, after determining the target control speed, the target control speed determined based on the reference flow rate speed control model is sent to the turbine fan controller. The turbine fan adjusts the fan output speed according to the received target control speed, so that the current airflow value is accurately adjusted to the preset target flow value.

[0084] Please see Figure 4 , Figure 4 This is a schematic diagram of a flow control process for a respiratory therapy device provided in an embodiment of this application. During the flow control process, initial airway pressure and initial airway flow values ​​are first acquired, and then subjected to dual time-domain filtering to obtain real-time airway pressure and flow values. Subsequently, the real-time airway pressure and flow values ​​are input into a dynamic compensation determination model to determine the dynamic compensation flow value. A preset target flow value is then calculated using the dynamic compensation flow value to obtain an equivalent target flow value. Based on the equivalent target flow value, a target control speed is generated, and the turbine fan is controlled to operate at the target control speed to adjust the airway flow value at the current moment.

[0085] In this embodiment, the flow control process of the respiratory therapy device can be a periodic execution process. After determining the target control speed and controlling the turbine fan speed for the current control cycle, the airway pressure and airway flow values ​​at the current moment can be retrieved again, and the next control cycle can begin. In this embodiment, the control cycle is not limited and can be adaptively adjusted according to real-time flow regulation requirements. As one possible implementation, the control cycle can be set to 100µs to 10ms.

[0086] Therefore, the embodiments of this application input the equivalent target flow rate value into the reference flow rate speed control model, generate the target control speed based on the reference flow rate speed control model, and match the target control speed with the target flow rate after considering the current airway state change. This enables the turbine fan to adjust its operating state according to the flow rate adjustment requirements corresponding to the current airway state, thereby improving the accuracy of the current airway flow rate adjustment process towards the preset target flow rate value.

[0087] The following section provides an overview of the flow control process. Please refer to [link / reference needed]. Figure 5 , Figure 5This application provides a system structure and control flowchart for a respiratory therapy device. A dynamic compensation determination model is determined based on first flow data, load pressure, and sample dynamic compensation flow value. A reference flow-speed control model is determined based on the correspondence between flow value and control speed under conditions without external pressure disturbance. During flow control of the respiratory therapy device, real-time airway pressure and flow values ​​are acquired, and the dynamic compensation flow value is determined using the dynamic compensation determination model. An equivalent target flow value is calculated by combining the preset target flow value and the dynamic compensation flow value. Finally, the reference flow-speed control model is used to generate the target control speed, thereby controlling the respiratory therapy device to adjust the airway flow value at the current moment.

[0088] As one possible implementation, embodiments of this application can be applied to a respiratory therapy device, which includes a sensor module, a controller module, a turbine fan module, and an expiratory valve module. The sensor module includes a flow sensor and a pressure sensor for acquiring initial airway pressure and initial airway flow values. The controller module includes a processor and a memory. The memory stores a pre-determined baseline flow-speed control model and a dynamic compensation determination model. The processor includes an acquisition unit, a filtering unit, a determination unit, and a generation unit for performing filtering, dynamic compensation flow value determination, equivalent target flow value determination, and target control speed generation processes. The expiratory valve module adjusts the airway pressure to achieve pressure control during inhalation and exhalation. The turbine fan module receives the target control speed generated by the controller module and adjusts the turbine fan's operating state according to the target control speed to adjust the current airway flow value.

[0089] Please see Figure 6 , Figure 6 This is a schematic diagram of a flow control device provided in an embodiment of this application. As shown in the figure, this application embodiment provides a flow control device, the device comprising: The acquisition unit 601 is used to acquire the initial airway pressure value and the initial airway flow value within a target time period, wherein the target time period is a time period located before the current time and whose length is less than a time length threshold. The filtering processing unit 602 is used to filter the initial airway pressure value to obtain the real-time airway pressure value and to filter the initial airway flow value to obtain the real-time airway flow value. The determining unit 603 is used to determine a dynamic compensation flow value based on the real-time airway pressure value and the real-time airway flow value; the dynamic compensation flow value is used to offset the flow value deviation caused by changes in load pressure. The determining unit 603 is further configured to determine that the sum of the preset target flow value and the dynamic compensation flow value is an equivalent target flow value; The generation unit 604 is used to generate a target control speed of the turbine fan based on the equivalent target flow rate value; the target control speed is used to adjust the airflow rate value at the current moment.

[0090] As one possible implementation, the target time period includes a first sliding filter window and a second sliding filter window; the initial airway pressure value includes a first initial airway pressure value and a second initial airway pressure value; the initial airway flow value includes a first initial airway flow value and a second initial airway flow value; and the filtering unit is specifically used for: Based on multiple initial airway flow rates within the first sliding filter window, a first average flow rate is calculated, and based on multiple initial airway pressure values ​​within the first sliding filter window, a first average pressure is calculated. A second average flow rate is calculated based on multiple second initial airway flow rates within the second sliding filter window, and a second average pressure is calculated based on multiple second initial airway pressure values ​​within the second sliding filter window; the window length of the first sliding filter window is greater than the window length of the second sliding filter window. The real-time airway flow rate value is obtained by weighting the first average flow rate and the second average flow rate. The real-time airway pressure value is obtained by weighting the first average pressure value and the second average pressure value.

[0091] As one possible implementation, the determining unit is specifically used for: The real-time airway pressure value and the real-time airway flow rate value are input into the dynamic compensation determination model to obtain the dynamic compensation flow rate value.

[0092] As one possible implementation, the determining unit is further configured to: Obtain at least one sample to control the rotation speed; For each of the sample control speeds, at least one load pressure and the first flow rate data corresponding to the load pressure and the second flow rate data corresponding to the state without external pressure disturbance are obtained; The difference between the second flow data and the first flow data is determined as the sample dynamic compensation flow value; Based on the first flow data, the load pressure, and the sample dynamic compensation flow value, the dynamic compensation determination model is determined.

[0093] As one possible implementation, the determining unit is specifically used for: Based on the sample control speed, the reference flow rate speed control model is inversely calculated to determine the second flow rate data corresponding to the sample control speed under the condition of no external pressure disturbance.

[0094] Optionally, the determining unit is further configured to: Obtain at least one reference control speed and the base flow value corresponding to each of the reference control speeds under conditions without external pressure disturbance; Based on the correspondence between the reference control speed and the basic flow rate value, a reference flow rate speed control model is established.

[0095] As one possible implementation, the determining unit is specifically used for: Under each of the sample control speeds, at least one of the load pressures corresponding to multiple initial sample output flow values ​​is collected. The average flow rate and the standard deviation of the flow rate are determined based on multiple initial sample output flow rate values, and initial sample output flow rate values ​​whose absolute value of the difference between them and the average flow rate is greater than three times the standard deviation of the flow rate are eliminated. Kernel density estimation is performed on the remaining initial sample output flow values ​​to obtain the flow probability density; The flow value corresponding to the point where the flow probability density is maximum is determined to be the first flow data.

[0096] As one possible implementation, the generation unit is specifically used for: The equivalent target flow rate value is input into the reference flow rate speed control model, and the target control speed is generated based on the reference flow rate speed control model.

[0097] On the other hand, embodiments of this application provide a computer device, the computer device including a processor and a memory: The memory is used to store computer programs; The processor is configured to perform the methods described above according to the computer program.

[0098] This application first acquires initial airway pressure and flow rates to obtain the real-time changes in these values. Then, the initial airway pressure and flow rates are filtered to obtain real-time airway pressure and flow rates. This process suppresses sudden changes in airway pressure and flow rates caused by sensor noise or signal interference while preserving the real-time trends of these values ​​as airway conditions change. This improves the timeliness and accuracy of identifying pressure fluctuations and flow rate deviations, preventing jumps in the dynamic compensation flow rate due to instantaneous fluctuations in the initial airway pressure or flow rate, thus preventing turbine fan target speed fluctuations caused by such jumps. Based on this, a dynamic compensation flow rate is determined using the real-time airway pressure and flow rates to pre-convert potential flow rate deviations caused by changes in current load pressure into corresponding flow compensation values, incorporating these deviations into the subsequent flow regulation process. Furthermore, the preset target flow rate value is added to the dynamically compensated flow rate value to obtain an equivalent target flow rate value, thereby offsetting the flow rate deviation caused by load pressure changes. Finally, a target control speed is generated based on the equivalent target flow rate value, and the current airway flow rate value is adjusted using the target control speed. This converts the target flow rate demand under load pressure into the target flow rate demand under conditions without external pressure disturbances, facilitating the rapid calculation of the target control speed. Therefore, this application improves the accuracy of real-time airway pressure and flow rate values ​​through filtering, and determines the dynamically compensated flow rate value based on these two values. It incorporates the flow rate deviation caused by pressure disturbances into the determination process of the equivalent target flow rate value and the target control speed in advance, thus compensating for the impact of pressure disturbances on airway flow rate during the target control speed determination process. This offsets the turbine fan flow control result deviation caused by pressure disturbances and achieves decoupling between pressure disturbances and the turbine fan flow control process. This avoids the lag in turbine air supply response caused by pressure disturbances in the airway, improves the turbine's response speed to changes in airway status, and suppresses the jump in dynamic compensation flow value caused by fluctuations in airway flow detection results, thereby preventing turbine target control speed jitter and ensuring the stability of the respiratory therapy equipment system.

[0099] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems or apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple, and relevant parts can be referred to the method section.

[0100] The term "comprising" and its variations as used herein are open-ended inclusions, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the description below.

[0101] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0102] It should also be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0103] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0104] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A flow control method, characterized by, The method includes: Obtain the initial airway pressure and initial airway flow values ​​within a target time period, wherein the target time period is a time period preceding the current time and whose length is less than a time length threshold; The initial airway pressure value is filtered to obtain the real-time airway pressure value, and the initial airway flow rate value is filtered to obtain the real-time airway flow rate value. A dynamic compensation flow rate value is determined based on the real-time airway pressure value and the real-time airway flow rate value; the dynamic compensation flow rate value is used to offset the flow rate value deviation caused by changes in load pressure. The sum of the preset target flow value and the dynamically compensated flow value is determined to be the equivalent target flow value; The target control speed of the turbine fan is generated based on the equivalent target flow rate value; the target control speed is used to adjust the airflow rate value at the current moment.

2. The method according to claim 1, characterized in that, The target time period includes a first sliding filter window and a second sliding filter window. The initial airway pressure value includes a first initial airway pressure value and a second initial airway pressure value. The initial airway flow value includes a first initial airway flow value and a second initial airway flow value. The process of filtering the initial airway pressure value to obtain a real-time airway pressure value and filtering the initial airway flow value to obtain a real-time airway flow value includes: Based on multiple initial airway flow rates within the first sliding filter window, a first average flow rate is calculated, and based on multiple initial airway pressure values ​​within the first sliding filter window, a first average pressure is calculated. A second average flow rate is calculated based on multiple second initial airway flow rates within the second sliding filter window, and a second average pressure is calculated based on multiple second initial airway pressure values ​​within the second sliding filter window; the window length of the first sliding filter window is greater than the window length of the second sliding filter window. The real-time airway flow rate value is obtained by weighting the first average flow rate and the second average flow rate. The real-time airway pressure value is obtained by weighting the first average pressure value and the second average pressure value.

3. The method according to claim 1, characterized in that, The determination of the dynamic compensation flow value based on the real-time airway pressure value and the real-time airway flow value includes: The real-time airway pressure value and the real-time airway flow rate value are input into the dynamic compensation determination model to obtain the dynamic compensation flow rate value.

4. The method according to claim 3, characterized in that, The method for constructing the dynamic compensation determination model includes: Obtain at least one sample to control the rotation speed; For each of the sample control speeds, at least one load pressure and the first flow rate data corresponding to the load pressure and the second flow rate data corresponding to the state without external pressure disturbance are obtained; The difference between the second flow data and the first flow data is determined as the sample dynamic compensation flow value; Based on the first flow data, the load pressure, and the sample dynamic compensation flow value, the dynamic compensation determination model is determined.

5. The method according to claim 4, characterized in that, The method for obtaining the second flow data corresponding to the state without external pressure disturbance includes: Based on the sample control speed, the reference flow rate speed control model is inversely calculated to determine the second flow rate data corresponding to the sample control speed under the condition of no external pressure disturbance.

6. The method according to claim 5, characterized in that, The method for constructing the reference flow rate and speed control model includes: Obtain at least one reference control speed and the base flow value corresponding to each of the reference control speeds under conditions without external pressure disturbance; Based on the correspondence between the reference control speed and the basic flow rate value, a reference flow rate speed control model is established.

7. The method according to claim 4, characterized in that, The step of acquiring at least one load pressure and the first flow rate data corresponding to the load pressure under each sample control speed includes: Under each of the sample control speeds, at least one of the load pressures corresponding to multiple initial sample output flow values ​​is collected. The average flow rate and the standard deviation of the flow rate are determined based on multiple initial sample output flow rate values, and initial sample output flow rate values ​​whose absolute value of the difference between them and the average flow rate is greater than three times the standard deviation of the flow rate are eliminated. Kernel density estimation is performed on the remaining initial sample output flow values ​​to obtain the flow probability density; The flow value corresponding to the point where the flow probability density is maximum is determined to be the first flow data.

8. The method according to any one of claims 1-7, characterized in that, The generation of the target control speed of the turbine fan based on the equivalent target flow rate value includes: The equivalent target flow rate value is input into the reference flow rate speed control model, and the target control speed is generated based on the reference flow rate speed control model.

9. A flow control device, characterized in that, The device includes: The acquisition unit is used to acquire the initial airway pressure value and the initial airway flow value within a target time period, wherein the target time period is a time period located before the current time and whose length is less than a time length threshold. The filtering unit is used to filter the initial airway pressure value to obtain the real-time airway pressure value and to filter the initial airway flow value to obtain the real-time airway flow value. A determining unit is configured to determine a dynamic compensation flow value based on the real-time airway pressure value and the real-time airway flow value; the dynamic compensation flow value is used to offset the flow value deviation caused by changes in load pressure. The determining unit is further configured to determine that the sum of the preset target flow value and the dynamically compensated flow value is an equivalent target flow value; The generation unit is used to generate a target control speed of the turbine fan based on the equivalent target flow rate value; the target control speed is used to adjust the airflow rate value at the current moment.

10. A computer device, characterized in that, The computer device includes a processor and memory: The memory is used to store computer programs; The processor is configured to perform the method according to any one of claims 1-8 according to the computer program.

Images