CN116183136A - An air leakage detection system for an exhalation valve - Google Patents

Abstract

The invention discloses an air leakage detection system for an exhalation valve, which comprises: an air leakage detection device including a first air leakage detection module and a second air leakage detection module, and an air leakage detection device including a horizontal cavity and a longitudinal cavity to be tested. Valve; the first gas leakage detection module, used to input the first pressure and the first gas flow in the longitudinal cavity to the preset gas leakage detection model, so that the gas leakage detection model can predict the first Gas leakage: determine whether the first gas leakage is greater than the first threshold, if so, determine that the exhalation valve to be tested is leaking; if not, input the second pressure and the second gas flow in the horizontal cavity to the preset gas Leakage detection model, so that the gas leakage detection model predicts the second gas leakage corresponding to the horizontal cavity; judge whether the second gas leakage is greater than the second threshold; if so, determine that the exhalation valve to be tested is a leak If not, it is determined that the exhalation valve to be tested is not leaking.

Description

An air leakage detection system for an exhalation valve

technical field

The invention relates to the technical field of gas detection, in particular to an air leakage detection system of an exhalation valve.

Background technique

In the medical field, the ventilator is a widely used machine, and the exhalation valve is an important part of the ventilator. Before each use of the ventilator, the ventilator needs to be tested for air leakage, which causes the ventilator gas Excessive leakage is usually caused by the exhalation valve in the ventilator. In the prior art, manual detection is used to detect whether the exhalation valve is leaking, but manual detection has a strong subjective judgment, resulting in the Air leak test results for air leaks are not accurate.

Contents of the invention

An embodiment of the present invention provides an air leakage detection system for an exhalation valve, which can effectively solve the problem of inaccurate air leakage detection results of the exhalation valve leakage in the prior art.

An embodiment of the present invention provides an air leakage detection system for an exhalation valve, including: an air leakage detection device, a gas input device, and an exhalation valve to be tested;

The exhalation valve to be tested includes a transverse cavity and a longitudinal cavity; wherein, the transverse cavity and the longitudinal cavity are vertically intersected;

The gas input device is respectively connected with the transverse cavity and the longitudinal cavity;

The air leakage detection device includes: a first air leakage detection module and a second air leakage detection module;

The first air leakage detection module is arranged on the gas passage formed by the gas input device and the longitudinal cavity; the second air leakage detection module is arranged on the gas passage formed by the gas input device and the transverse cavity;

The gas input device is used to deliver gas to the longitudinal cavity and the horizontal cavity;

The first gas leakage detection module is used to input the first pressure in the longitudinal cavity and the first gas flow rate in the longitudinal cavity to the preset gas leakage detection model, so that the gas leakage detection model can predict the longitudinal cavity The first gas leakage amount corresponding to the body; judge whether the first gas leakage amount is greater than the first threshold value, if so, then determine whether the exhalation valve to be tested is leaking; if not, then detect whether the horizontal cavity is leaking;

The second gas leakage detection module is configured to input the second pressure in the lateral chamber and the second gas flow in the lateral chamber to the preset gas leakage amount when the first gas leakage amount is less than or equal to the first threshold Detection model, so that the gas leakage detection model predicts the second gas leakage corresponding to the horizontal cavity; judging whether the second gas leakage is greater than the second threshold; if so, judging that the exhalation valve to be tested is leaking ; If not, it is determined that the exhalation valve to be tested is not leaking.

Further, said generating a preset gas leakage detection model includes:

Build an initial gas leakage detection model;

Obtain the sample pressure, sample gas flow rate and the corresponding sample gas leakage at each sampling moment at several sampling moments;

For each sampling moment, the sample pressure and sample gas flow rate at each sampling moment are used as input, and the sample gas leakage at the corresponding sampling moment is used as output to train the initial gas leakage detection model to generate the predicted The established gas leakage detection model.

Further, the preset gas leakage detection model includes: an input layer, a hidden layer and an output layer;

The input layer is used to divide the pressure to be measured and the flow to be measured to obtain a number of pressure segments to be measured and flow segments to be measured, and transmit the pressure segments to be measured and the flow segments to be measured to the hidden layer;

The hidden layer is used to extract the correlation information between the pressure segments to be measured and the flow segments to be measured according to the pressure segments to be measured and the flow segments to be measured, and to combine the pressure segments to be measured and the flow segments to be measured The correlation information of is transmitted to the output layer;

The output layer is configured to generate a predicted gas leakage corresponding to the pressure to be measured and the flow to be measured in the input layer according to the correlation information between the pressure segments to be measured and the flow segments to be measured.

Further, according to the correlation information between the pressure segments to be measured and the flow segments to be measured, generating the predicted gas leakage corresponding to the pressure to be measured and the flow rate to be measured in the input layer includes:

According to the correlation information between the pressure segments to be measured and the flow segments to be measured, combined with the normalized exponential function and the logistic function, the predicted gas leakage corresponding to the pressure to be measured and the flow rate to be measured in the input layer is generated quantity.

Further, the first air leakage detection module includes: a first solenoid valve, a first flow sensor and a first pressure sensor

The first flow sensor is used to measure the first gas flow in the longitudinal cavity;

The first pressure sensor is used to measure the first pressure in the longitudinal cavity;

The first solenoid valve is used to acquire the first gas flow rate transmitted by the first flow sensor and the first pressure transmitted by the first pressure sensor, and input the first gas flow rate and the first pressure to the preset gas leakage detection Model.

Further, the second air leakage detection module includes: a second solenoid valve, a second flow sensor and a second pressure sensor;

The second flow sensor is used to measure the second gas flow in the transverse cavity;

The second pressure sensor is used to measure the second pressure in the transverse cavity;

The second solenoid valve is used to obtain the second gas flow rate transmitted by the first flow sensor and the second pressure transmitted by the first pressure sensor, and input the second gas flow rate and the second pressure to the preset gas leakage detection Model.

Further, the first air leakage detection module further includes: a first valve, a first air resistance, and a first air containment chamber;

The first valve is connected to one end of the first air resistance, the other end of the first air resistance is connected to the first electromagnetic valve, and the first pressure sensor is connected to the first air volume chamber;

The first valve is used to control whether the gas provided by the gas input device can enter the first gas leakage detection module;

The first air resistance is used to reduce the pressure input into the first air leakage detection module;

The first gas chamber is used to buffer the gas input into the first gas leakage detection module;

The first solenoid valve is also used to adjust the opening degree of the first solenoid valve according to the received first gas flow rate and first pressure.

Further, the second air leakage detection module further includes: a second valve, a second air resistance, and a second air containment chamber;

The second valve is connected to one end of the second air resistance, the other end of the second air resistance is connected to the second solenoid valve, and the second pressure sensor is connected to the second air volume chamber;

The second valve is used to control whether the gas provided by the gas input device can enter the second gas leakage detection module;

The second air resistance is used to reduce the air pressure input into the second air leakage detection module;

The second gas chamber is used for buffering the gas input into the second gas leakage detection module.

The second solenoid valve is also used to adjust the opening of the second solenoid valve according to the received second gas flow rate and second pressure.

Further, the gas input device also includes: a pressure reducing valve;

The decompression valve is used to adjust the flow rate and pressure of gas input into the longitudinal chamber and the transverse chamber.

The implementation of the present invention has the following beneficial effects: the present invention discloses an air leakage detection system of an exhalation valve, which detects the airtightness of the transverse cavity and the longitudinal cavity of the exhalation valve respectively through the air leakage detection device, According to the pressure and gas flow in the horizontal cavity and the vertical cavity, the gas leakage in the horizontal cavity and the vertical cavity are detected respectively through the preset gas leakage detection model, and the gas leakage in the vertical cavity is greater than the first When the threshold value is reached, it is determined that the exhalation valve to be tested is leaking. When the gas leakage of the longitudinal cavity is less than or equal to the first threshold, the gas leakage of the horizontal cavity is judged. If the gas leakage of the horizontal cavity is greater than the second Threshold, it is determined that the exhalation valve to be tested is leaking, otherwise it is determined that the exhalation valve to be tested is not leaking; by implementing the present invention, the gas leakage of the exhalation valve to be tested can be accurately obtained, thereby judging the amount of gas to be tested according to the gas leakage. Test whether the exhalation valve leaks, so as to make an accurate judgment on the air tightness of the exhalation valve to be tested.

Description of drawings

Fig. 1 is a schematic structural diagram of an exhalation valve leakage detection system provided by an embodiment of the present invention.

Fig. 2 is a schematic structural diagram of an exhalation valve to be tested provided by an embodiment of the present invention.

Fig. 3 is an air circuit diagram of an exhalation valve air leakage detection system provided by an embodiment of the present invention.

Detailed ways

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

As shown in Figure 1, an exhalation valve leakage detection system provided by an embodiment of the present invention includes: an air leakage detection device, a gas input device, and an exhalation valve to be tested; the exhalation valve to be tested includes a transverse cavity Body and longitudinal cavity; wherein, the transverse cavity vertically intersects with the longitudinal cavity; the gas input device is respectively connected with the transverse cavity and the longitudinal cavity; the gas leakage detection device includes: a first gas leakage detection module And the second air leakage detection module; the first air leakage detection module is arranged on the gas passage formed by the gas input device and the longitudinal cavity; the second air leakage detection module is arranged on the gas input device and the horizontal cavity On the formed gas passage; the gas input device is used to transport gas to the longitudinal cavity and the horizontal cavity;

The first gas leakage detection module is used to input the first pressure in the longitudinal cavity and the first gas flow rate in the longitudinal cavity to the preset gas leakage detection model, so that the gas leakage detection model can predict the longitudinal cavity The first gas leakage amount corresponding to the body; judge whether the first gas leakage amount is greater than the first threshold value, if so, then determine whether the exhalation valve to be tested is leaking; if not, then detect whether the horizontal cavity is leaking;

The second gas leakage detection module is configured to input the second pressure in the lateral chamber and the second gas flow in the lateral chamber to the preset gas leakage amount when the first gas leakage amount is less than or equal to the first threshold Detection model, so that the gas leakage detection model predicts the second gas leakage corresponding to the horizontal cavity; judging whether the second gas leakage is greater than the second threshold; if so, judging that the exhalation valve to be tested is leaking ; If not, it is determined that the exhalation valve to be tested is not leaking.

Specifically, the exhalation valve to be tested includes a horizontal cavity and a longitudinal cavity, and the horizontal cavity and the longitudinal cavity are vertically intersected; as shown in Figure 2, it is an exhalation valve to be tested according to an embodiment of the present invention. The exhalation valve to be tested consists of a transverse cavity device 110 and a longitudinal cavity device 120; the longitudinal cavity device 120 consists of a bottom cover 10, a first gasket 11, a sealing structure 14, a rubber ring 3, a third gasket 2 And the floating core 1 is connected in sequence; the sealing structure 14 is composed of the rubber ring 12 and the second gasket 13 connected in sequence; the rubber ring 3, the third gasket 2 and the floating core 1 are connected in sequence to form a floating chip; the floating chip is arranged in the horizontal direction On the cavity device 110; the vertical cavity device 120 and the horizontal cavity device 110 are vertically intersected; the sealing structure 14 and the first gasket 11 are used to block or connect the gas between the horizontal cavity device 110 and the longitudinal cavity device 120 Passage; one side of the bottom cover 10 is provided with an air inlet, which is connected to one end of the first air leakage detection module in the air leakage detection device, and the other end of the first air leakage detection module is connected to the gas input device , the gas output by the input device is input into the longitudinal cavity device 120 from the air inlet of the bottom cover 10 through the first air leakage detection module, and pushes the first gasket 11 to move in the direction of the sealing structure 14 until the first gasket 11 And the sealing structure 14 is closely attached to the horizontal cavity device 110. At this time, the above-mentioned longitudinal cavity is formed between the bottom cover 10 and the first gasket 11; the horizontal cavity device 110 is formed by the adapter 7, the fourth gasket 6. The cavity tube 5, the conduit 15 and the check valve 16 are connected in sequence; one end of the conduit 15 is connected to the check valve 16, and the other end is built into the cavity tube 5; one end of the adapter 7 is connected to the fourth gasket 6 The other end is provided with an inlet, which is used to connect with one end of the second air leakage detection module when performing air leakage detection. The other end of the second air leakage detection module is connected to the gas input device, and the gas input device is input In the case of gas, the above-mentioned lateral cavity is formed in the lateral cavity device 110 .

When performing gas leakage detection, the gas input device supplies gas to the longitudinal cavity, and a gas passage is formed on the connection path where the gas input device, the first gas leakage detection module and the longitudinal cavity are located, and the gas is passed through the first gas leakage detection module. The gas leakage of the gas passage is detected; a gas leakage detection model is preset in the first gas leakage detection module, and the first gas flow and the first pressure in the longitudinal cavity are input into the preset gas leakage detection model, That is, the corresponding first gas leakage amount can be predicted, and it is judged whether the longitudinal cavity is leaking according to the relationship between the first gas leakage amount and the first threshold; if the longitudinal cavity is leaking, it can be determined that the exhalation valve to be tested is leaking; If there is no air leakage in the longitudinal cavity, the longitudinal cavity of the exhalation valve to be tested is not leaking. In the case of keeping the gas in the longitudinal cavity unchanged, enable the gas input device to input gas into the horizontal cavity of the exhalation valve to be tested. ;A gas passage is formed on the connection path where the gas input device, the second gas leakage detection module and the lateral cavity are located, and the gas leakage of the gas passage is detected by the second gas leakage detection module; in the second gas leakage detection module A gas leakage detection model that is consistent with the first gas leakage detection module is preset, and the second gas flow rate and second pressure in the lateral cavity are input into the above-mentioned preset gas leakage model to predict the corresponding second gas leakage. Gas leakage, according to the relationship between the second gas leakage and the second threshold to determine whether the horizontal cavity is leaking; if the horizontal cavity is not leaking, then the two cavities included in the exhalation valve to be tested are not leaking If there is air leakage in the expiratory valve to be tested, it is determined that the exhalation valve to be tested is not leaking; if the horizontal cavity is leaking, it is determined that the exhalation valve to be tested is not leaking in the longitudinal cavity and the horizontal cavity is leaking.

In a preferred embodiment, said generating a preset gas leakage detection model includes: constructing an initial gas leakage detection model; acquiring sample pressures, sample gas flow rates, and corresponding gas flow rates at several sampling moments; Leakage of sample gas: For each sampling moment, the sample pressure and sample gas flow rate at each sampling moment are used as input, and the sample gas leakage at the corresponding sampling moment is used as output to train the initial gas leakage detection model , generating the preset gas leakage detection model;

Specifically, an initial gas leakage detection model is constructed according to the neural network model, and a large number of known sample pressures, sample gas flow rates, and sample gas leakage are obtained to train the constructed initial gas leakage detection model; The sample pressure, sample gas flow, and sample gas leakage at multiple sampling moments, with the next sample pressure and sample gas flow at a certain moment as input, and the next sample gas flow at the corresponding moment as output, the initial gas leakage The detection model is trained, and the loss function in the initial gas leakage detection model is adjusted according to multiple sets of training results to minimize the output value of the loss function in the initial gas leakage model, and then the output value of the loss function is obtained during the training process The smallest case generates the above preset gas leakage model;

It should be noted that when training the neural network model (that is, the above-mentioned initial gas leakage model), when the input of sample data and the output of corresponding sample data are input into the neural network model for training, it is essentially through training The loss function (loss function) in the process regulation neural network model, the loss function can also be called the objective function (objective function); The important equation of the difference between them; for example: the higher the output value of the loss function, the greater the difference, then in the process of training the neural network model, it is necessary to make the output value of the loss function as small as possible, and then the neural network model predicts The difference between the result and the target result should be reduced as much as possible; From the perspective of a single training, after the sample pressure and sample gas flow are input into the neural network model, the neural network model will And the sample gas flow rate to obtain a predicted gas leakage, compare the predicted gas leakage with the corresponding sample gas leakage, and adjust the loss function in the neural network model according to the difference value of the comparison; from the perspective of overall training, Adjust the loss function in the overall neural network model according to multiple sets of single training data, and use the neural network model when the output value of the loss function is the smallest as the above-mentioned preset gas leakage detection model; this model can be expressed as:

y=f(x)

Among them, y represents the predicted gas leakage, x represents the pressure to be measured and the flow rate of the gas to be measured in the input model, and f(x) represents the difference between the pressure to be measured and the flow rate of the gas to be measured in the input model and the predicted gas leakage mapping relationship.

In a preferred embodiment, the preset gas leakage detection model includes: an input layer, a hidden layer, and an output layer; the input layer is used to divide the pressure to be measured and the flow rate of the gas to be measured to obtain several The pressure segment to be measured and the gas flow segment to be measured, and several pressure segments to be measured and the gas flow segment to be measured are transmitted to the hidden layer; the hidden layer is used to extract each The correlation information between the pressure segments to be measured and the gas flow segments to be measured, and the correlation information between the pressure segments to be measured and the gas flow segments to be measured are transmitted to the output layer; Correlation information between the pressure segment to be measured and the gas flow rate segment to be measured, generating a predicted gas leakage corresponding to the pressure to be measured and the gas flow rate to be measured in the input layer;

Specifically, when the pressure to be measured and the gas flow rate to be measured are input to the preset gas leakage model, the pressure to be measured and the gas flow rate to be measured first reach the input layer of the preset gas leakage model, and the input layer is to be measured The pressure and the gas flow rate to be measured are normalized, and then the pressure to be measured and the gas flow rate to be measured are divided into segments of any length, and the segmented segments are transmitted to the hidden layer; the hidden layer is based on the input layer. Segment, extract the similarity information between all possible segments, calculate it, and transmit the calculation result (ie the above correlation information) as the output data of the hidden layer to the output layer; the output data of the hidden layer can be used to represent the pressure to be measured And the spatial correlation of the extracted information in the gas flow to be measured; the output layer is used to use the output data of the hidden layer as input data to generate corresponding prediction results;

It should be noted that the gas leakage prediction model may include a long-term short-term memory model, a self-attention model, and a prediction result model; when the pressure to be measured and the flow rate of the gas to be measured are input into To the long-term short-term memory model, the long-term short-term memory model calculates the time-series results according to the input data; the time-series results are input into the self-attention model to obtain high-dimensional feature data, wherein the self-attention model is used to determine the pressure from the test and the pressure to be measured. The spatial correlation of the information extracted from the gas flow measurement; the self-attention model is also connected to a fully connected neural network, and according to the characteristics that the output of the upper network of the fully connected neural network is used as the input of the next layer of the network, the above time series results are input After entering the self-attention model, the self-attention model obtains multiple self-attention values, and then inputs the self-attention values into the fully connected neural network, and then obtains high-dimensional feature data; finally, the high-dimensional feature data is input into the prediction result model Obtain the corresponding prediction results; it should be understood that the above-mentioned long-term short-term memory model, self-attention model and prediction result model can be three independent neural network models, and can also be integrated in the same neural network model; when the long-term short-term memory model When the memory model, self-attention model, and prediction result model are integrated in the same neural network model, the long-short-term memory model, self-attention model, and prediction result model can be different layers in the neural network model, for example: long-short-term The memory model can be the above-mentioned input layer, the self-attention model can be the above-mentioned hidden layer, and the prediction result model can be the above-mentioned output layer;

What needs to be added is that when the neural network model (i.e. the above-mentioned gas leakage detection model) has multiple hidden layers, the first hidden layer is used to use the fragments output by the input layer as input data to extract the distance between all possible fragment pairs. The similarity information is calculated to obtain the output data of this layer; the second hidden layer is used to use the output data of the first hidden layer as input data to further extract similarity information to obtain the output data of this layer; the third hidden layer is used to use The output data of the second hidden layer is used as the input data, and the similarity information is further extracted to obtain the output data of this layer; and so on, until the last hidden layer is calculated to obtain the output data of this layer; and the output layer is used for the last one The output data of the hidden layer is used as input data.

In a preferred embodiment, according to the correlation information between the pressure segments to be measured and the flow segments to be measured, the predicted gas leakage corresponding to the pressure to be measured and the flow rate to be measured in the input layer is generated, including: According to the correlation information between the pressure segments to be measured and the flow segments to be measured, combined with the normalized exponential function and the logistic function, the predicted gas leakage corresponding to the pressure to be measured and the flow rate to be measured in the input layer is generated quantity;

Specifically, the input layer uses the output data of the hidden layer as input data, and uses functions such as normalized exponential function and logistic function to perform calculations to obtain the predicted gas leakage corresponding to the pressure to be measured and the flow rate of the gas to be measured.

In a preferred embodiment, the first gas leakage detection module includes: a first solenoid valve, a first flow sensor, and a first pressure sensor; the first flow sensor is used to measure the first gas in the longitudinal cavity flow; the first pressure sensor is used to measure the first pressure in the longitudinal cavity; the first solenoid valve is used to obtain the first gas flow transmitted by the first flow sensor and the first pressure transmitted by the first pressure sensor , and input the first gas flow rate and the first pressure into the preset gas leakage detection model;

Specifically, as shown in Figure 3, the first air leakage detection module includes a first electromagnetic valve V1, a first flow sensor DPS1 and a first pressure sensor PS1; the gas input device is connected to one end of the first electromagnetic valve V1, and the first One end of the solenoid valve V1 is connected to one end of the first flow sensor DPS1, one end of the first solenoid valve V1 is connected to one end of the first pressure sensor PS1, and the other end of the first flow sensor DPS1 is connected to one end of the first pressure sensor PS1, The other end of the first pressure sensor PS1 is connected to the longitudinal cavity C1; when the first gas leakage detection module performs gas leakage detection, the first flow sensor DPS1 measures the gas flow in the longitudinal cavity in real time (ie the first gas flow), The first pressure sensor PS1 measures the gas pressure in the longitudinal cavity in real time (that is, the above-mentioned first air pressure), and feeds back the measured gas flow and gas pressure to the first solenoid valve V1 in real time, and the first solenoid valve V1 The flow rate and gas pressure are input into the gas leakage detection model to predict the gas leakage in the longitudinal cavity C1 (i.e. the above-mentioned first gas leakage);

In a preferred embodiment, the first air leakage detection module further includes: a first valve, a first air block, and a first air volume cavity; the first valve is connected to one end of the first air block, and the first The other end of the air resistance is connected to the first solenoid valve, and the first pressure sensor is connected to the first gas chamber; the first valve is used to control whether the gas provided by the gas input device can enter the first air leakage detection module ; The first air resistance is used to reduce the pressure input into the first air leakage detection module; the first air volume chamber is used to buffer the gas input into the first air leakage detection module; the first solenoid valve , which is also used to adjust the opening degree of the first electromagnetic valve according to the received first gas flow rate and the first pressure;

Specifically, one end of the gas input device is connected to one end of the first valve K1, the other end of the first valve K1 is connected to the first gas resistance R1, and the other end of the first gas resistance R1 is connected to one end of the first electromagnetic valve V1, One end of the 50ml air volume chamber (i.e. the above-mentioned first air volume chamber) is connected to one end of the first pressure sensor PS1, and the other end of the 50ml air volume chamber (i.e. the above-mentioned first air volume chamber) is connected to the longitudinal chamber C1; the first When the valve K1 is in the open state, the gas output by the gas input device can enter the first gas leakage detection device through the first valve K1 and then reach the longitudinal cavity C1. When the first valve K1 is closed, the output gas of the gas input device is blocked from flowing into the first gas leakage detection device An air leakage detection device, when the first air leakage detection module performs air leakage detection, the first valve K1 is kept open; the first air resistance R1 is used to reduce the air pressure in the gas passage; due to the longitudinal chamber of the exhalation valve The body space is very small, and a 50ml gas volume chamber (i.e. the above-mentioned first gas volume chamber) is set on the gas passage to buffer the flow rate of the gas input into the longitudinal chamber C1, allowing the gas in the longitudinal chamber to be filled slowly, avoiding the If the gas pressure of C1 in the longitudinal cavity is too high in a short period of time; the first solenoid valve V1 is provided with a circuit control unit, and the circuit control unit of the first solenoid valve V1 receives the first gas feedback from the first flow sensor DPS1 The flow rate and the first pressure fed back by the first pressure sensor PS1 adjust the opening of the first solenoid valve V1, and the adjustment of the opening of the first solenoid valve V1 can be controlled by a PID algorithm;

What needs to be added is that, in order to make the above prediction of gas leakage more accurate, before inputting the first gas flow and the first pressure into the preset gas leakage model, the input data should be optimized as follows: The first gas flow rate and the first pressure measured during the inflating process of the body are not used as the input data input into the preset gas leakage model, but set a pressure for the gas channel where the longitudinal cavity C1 is located. Set the pressure value, and continuously adjust the opening of the first solenoid valve V1 until the first pressure fed back to the first solenoid valve V1 by the first pressure sensor PS1 reaches the set pressure value, and start to obtain the gas leakage for predicting the longitudinal cavity The first gas flow and the first pressure are used as the input data for predicting the first gas leakage, and are input into the preset gas leakage detection model; the input data can be the first gas flow and the first gas flow at a certain moment or at a certain time The first pressure, the output data will make corresponding response results according to the different input data; for example: the current set pressure value is 3.5L/min/2bar, after the current pressure of the gas passage reaches 3.5L/min/2bar, keep the current The pressure of the gas passage is 3.5L/min/2bar for a period of time, and the gas flow and pressure during this period of time are obtained, which are input into the gas leakage detection model as input data, and this is output according to the gas leakage detection model. The corresponding gas leakage in the time period.

In a preferred embodiment, the second gas leakage detection module includes: a second solenoid valve, a second flow sensor, and a second pressure sensor; the second flow sensor is used to measure the second gas in the horizontal cavity flow rate; the second pressure sensor is used to measure the first pressure in the transverse cavity; the first solenoid valve is used to obtain the second gas flow rate transmitted by the second flow sensor and the second pressure transmitted by the second pressure sensor , and input the second gas flow and the second pressure into the preset gas leakage detection model;

Specifically, when the detection result of the first air leakage detection module is that the longitudinal cavity of the exhalation valve is not leaking, the gas in the longitudinal cavity remains unchanged, the first valve K1 is in an open state, and the gas input device sends air to the exhalation valve. The lateral chamber of the air valve transmits gas, and the second air leakage detection module is used to detect air leakage in the lateral chamber; the second air leakage detection module includes a second solenoid valve V2, a second flow sensor DPS2 and a second pressure sensor PS2 The gas input device is connected with one end of the second solenoid valve V2, one end of the second solenoid valve V2 is connected with one end of the second flow sensor DPS2, one end of the second solenoid valve V2 is connected with one end of the second pressure sensor PS2, and the second The other end of the flow sensor DPS2 is connected to one end of the second pressure sensor PS2, and the other end of the second pressure sensor PS2 is connected to the lateral cavity C2; when the second air leakage detection module performs air leakage detection, the second flow sensor DPS2 real-time Measuring the gas flow in the horizontal chamber (i.e. the above-mentioned second gas flow), the second pressure sensor PS2 measures the gas pressure in the horizontal chamber in real time (i.e. the above-mentioned second air pressure), and feeds back the measured gas flow and gas pressure in real time For the second solenoid valve V2, the second solenoid valve V2 inputs it into the gas leakage detection model according to the received gas flow and gas pressure, and the gas leakage in the lateral cavity C2 (that is, the above-mentioned second gas leakage ) to predict;

In a preferred embodiment, the second air leakage detection module further includes: a second valve, a second air block, and a second air volume cavity; the second valve is connected to one end of the second air block, and the second The other end of the air resistance is connected to the second solenoid valve, and the second pressure sensor is connected to the second gas chamber; the second valve is used to control whether the gas provided by the gas input device can enter the second air leakage detection module ; The second air resistance is used to reduce the pressure input into the second air leakage detection module; the second air volume chamber is used to buffer the gas input into the second air leakage detection module; the second solenoid valve , which is also used to adjust the opening degree of the second electromagnetic valve according to the received second gas flow rate and second pressure;

Specifically, one end of the gas input device is connected to one end of the second valve K2, the other end of the second valve K2 is connected to the second air resistance R2, and the other end of the second air resistance R2 is connected to one end of the second solenoid valve V2, One end of the 100ml air volume chamber (i.e. the above-mentioned second air volume chamber) is connected to one end of the second pressure sensor PS2, and the other end of the 100ml air volume chamber (i.e. the above-mentioned second air volume chamber) is connected to the transverse cavity C2; the second When the valve K2 is open, the gas output from the gas input device can enter the second gas leakage detection device through the second valve K2 and then reach the lateral cavity C2. When the second valve K2 is closed, the output gas from the gas input device is blocked from flowing into the first The second air leakage detection device, when the second air leakage detection module performs air leakage detection, keeps the second valve K2 open, closes the valve K, and then opens the second valve K2; When the second gas leakage detection module detects the gas leakage of the horizontal cavity C2, the gas in the gas channel where the longitudinal cavity C1 is located is already in a saturated state, then when the gas leakage detection of the horizontal cavity C2 is performed , can keep the first valve K1 in the open state; the second air resistance R2 is used to reduce the air pressure in the gas passage where the transverse cavity C2 is located; since the space of the transverse cavity of the exhalation valve is very small, 100ml gas is set on the gas passage The cavity (namely the above-mentioned second gas cavity) can buffer the flow rate of the gas input into the lateral cavity C2, so that the gas in the horizontal cavity can be filled slowly, avoiding the excessive gas pressure of C2 in the horizontal cavity in a short time The second solenoid valve V2 is provided with a circuit control unit, and the circuit control unit of the second solenoid valve V2 receives the second gas flow rate fed back by the second flow sensor DPS2 and the second pressure fed back by the second pressure sensor PS2. , to adjust the opening degree of the second solenoid valve V2, the adjustment of the opening degree of the second solenoid valve V2 can be controlled by the PID algorithm;

What needs to be added is that, in order to make the above-mentioned predicted gas leakage more accurate, before inputting the second gas flow and second pressure into the preset gas leakage model, the input data should be optimized as follows: The second gas flow rate and the second pressure measured during the body inflating process are not used as the input data input into the preset gas leakage model, but set a pressure for the gas channel where the lateral cavity C2 is located. Set the pressure value, and continuously adjust the opening of the second solenoid valve V2 until the second pressure fed back to the second solenoid valve V2 by the second pressure sensor PS2 reaches the set pressure value, and start to obtain the gas leakage for predicting the lateral cavity The second gas flow and the second pressure are used as the input data for predicting the second gas leakage, and are input into the preset gas leakage detection model; the input data can be the second gas flow and the second gas flow at a certain moment or at a certain time The second pressure, the output data will make corresponding response results according to the different input data; for example: the current set pressure value is 3.5L/min/2bar, after the current pressure of the gas passage reaches 3.5L/min/2bar, keep the current The pressure of the gas passage is 3.5L/min/2bar for a period of time, and the gas flow and pressure during this period of time are obtained, which are input into the gas leakage detection model as input data, and this is output according to the gas leakage detection model. The corresponding gas leakage in the time period.

It should be noted that the set pressure value of the longitudinal chamber C1, the set pressure value of the transverse chamber C2, the first threshold value and the second threshold value are the result values obtained after many tests in the actual implementation process.

In a preferred embodiment, the gas input device further includes: a pressure reducing valve; the pressure reducing valve is used to adjust the flow rate and pressure of the gas input into the longitudinal cavity and the horizontal cavity;

Specifically, the gas output by the gas input device is medical gas, and the pressure reducing valve can be set as a first-stage adjustable pressure reducing valve; the gas output by the gas input device passes through the pressure reducing valve before entering the horizontal cavity C2 or the longitudinal cavity C1. After decompression, it is then input into the gas passage where the corresponding chamber to be tested is located for leak detection.

It should be noted that the device embodiments described above are only illustrative, and the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physically separated. A unit can be located in one place, or it can be distributed to multiple network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, in the drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that they have a communication connection, which can be specifically implemented as one or more communication buses or signal lines. It can be understood and implemented by those skilled in the art without creative effort.

Those skilled in the art can clearly understand that, for the sake of convenience and brevity, for the specific working process of the device described above, reference can be made to the corresponding process in the foregoing method embodiments, which will not be repeated here.

The above description is a preferred embodiment of the present invention, and it should be pointed out that for those skilled in the art, without departing from the principle of the present invention, some improvements and modifications can also be made, and these improvements and modifications are also considered Be the protection scope of the present invention.

Claims (9)

1. An air leak detection system for an exhalation valve, comprising: the device comprises an air leakage detection device, a gas input device and an exhalation valve to be detected;

the exhalation valve to be tested comprises a transverse cavity and a longitudinal cavity; wherein the transverse cavity is vertically intersected with the longitudinal cavity;

the gas input device is connected with the transverse cavity and the longitudinal cavity respectively;

the air leakage detection device includes: a first air leakage detection module and a second air leakage detection module;

the first air leakage detection module is arranged on an air passage formed by the air input device and the longitudinal cavity; the second air leakage detection module is arranged on an air passage formed by the air input device and the transverse cavity;

the gas input device is used for conveying gas to the longitudinal cavity and the transverse cavity;

the first air leakage detection module is used for inputting the first pressure in the longitudinal cavity and the first gas flow in the longitudinal cavity into a preset gas leakage amount detection model so that the gas leakage amount detection model predicts the first gas leakage amount corresponding to the longitudinal cavity; judging whether the first gas leakage amount is larger than a first threshold value, if so, judging that the to-be-detected exhalation valve is air leakage; if not, detecting whether the transverse cavity leaks air;

the second air leakage detection module is used for inputting the second pressure in the transverse cavity and the second air flow in the transverse cavity into a preset air leakage detection model when the first air leakage is smaller than or equal to a first threshold value, so that the air leakage detection model predicts the second air leakage corresponding to the transverse cavity; judging whether the second gas leakage amount is larger than a second threshold value; if yes, judging that the expiratory valve to be detected is air leakage; if not, the exhalation valve to be detected is judged to be airtight.

2. The air leakage detecting system of an exhalation valve according to claim 1, wherein the generating of the preset air leakage amount detecting model includes:

constructing an initial gas leakage amount detection model;

acquiring sample pressure, sample gas flow and corresponding sample gas leakage at each sampling time;

and for each sampling time, taking the sample pressure and the sample gas flow at each sampling time as input, and taking the sample gas leakage at the corresponding sampling time as output, training an initial gas leakage detection model, and generating the preset gas leakage detection model.

3. A gas leakage detection system for an exhalation valve according to claim 2, wherein the preset gas leakage amount detection model comprises: an input layer, a hidden layer, and an output layer;

the input layer is used for dividing the pressure to be measured and the gas flow to be measured to obtain a plurality of pressure segments to be measured and gas flow segments to be measured, and transmitting the pressure segments to be measured and the gas flow segments to be measured to the hidden layer;

the hidden layer is used for extracting the correlation information between each pressure segment to be detected and the gas flow segment to be detected according to the plurality of pressure segments to be detected and the gas flow segment to be detected, and transmitting the correlation information between each pressure segment to be detected and the gas flow segment to be detected to the output layer;

the output layer is used for generating predicted gas leakage quantity corresponding to the pressure to be detected and the gas flow to be detected in the input layer according to the correlation information between the pressure fragments to be detected and the gas flow fragments to be detected.

4. A gas leakage detecting system for an exhalation valve according to claim 3, wherein the generating of the predicted gas leakage amount corresponding to the pressure to be measured and the flow to be measured in the input layer based on the correlation information between the pressure to be measured and the flow to be measured fragments comprises:

and generating predicted gas leakage corresponding to the pressure to be measured and the flow to be measured in the input layer according to the correlation information between the pressure to be measured and the flow to be measured and combining a normalized exponential function and a logic substance function.

5. A gas leakage detection system for an exhalation valve according to claim 1, wherein said first gas leakage detection module comprises: a first solenoid valve, a first flow sensor, and a first pressure sensor;

the first flow sensor is used for measuring the first gas flow in the longitudinal cavity;

the first pressure sensor is used for measuring a first pressure in the longitudinal cavity;

the first electromagnetic valve is used for acquiring the first gas flow transmitted by the first flow sensor and the first pressure transmitted by the first pressure sensor, and inputting the first gas flow and the first pressure into a preset gas leakage detection model.

6. A gas leakage detection system for an exhalation valve according to claim 5, wherein said second gas leakage detection module comprises: a second solenoid valve, a second flow sensor, and a second pressure sensor;

the second flow sensor is used for measuring second gas flow in the transverse cavity;

the second pressure sensor is used for measuring a second pressure in the transverse cavity;

the second electromagnetic valve is used for acquiring second gas flow transmitted by the first flow sensor and second pressure transmitted by the first pressure sensor, and inputting the second gas flow and the second pressure into a preset gas leakage detection model.

7. The air leak detection system of an exhalation valve of claim 5, wherein the first air leak detection module further comprises: the first valve, the first air resistor and the first air cavity;

the first valve is connected with one end of a first air resistor, the other end of the first air resistor is connected with a first electromagnetic valve, and the first pressure sensor is connected with a first air cavity;

the first valve is used for controlling whether the gas provided by the gas input device can enter the first gas leakage detection module;

the first air resistance is used for reducing the pressure input into the first air leakage detection module;

the first gas containing cavity is used for buffering gas input into the first gas leakage detection module;

the first electromagnetic valve is also used for adjusting the opening of the first electromagnetic valve according to the received first gas flow and the first pressure.

8. The air leak detection system of an exhalation valve of claim 6, wherein the second air leak detection module further comprises: the second valve, the second air resistor and the second air containing cavity;

the second valve is connected with one end of a second air resistor, the other end of the second air resistor is connected with a second electromagnetic valve, and the second pressure sensor is connected with a second air cavity;

the second valve is used for controlling whether the gas provided by the gas input device can enter the second gas leakage detection module;

the second air resistance is used for reducing the air pressure input into the second air leakage detection module;

the second gas accommodating cavity is used for buffering the gas input into the second gas leakage detection module;

the second electromagnetic valve is also used for adjusting the opening degree of the second electromagnetic valve according to the received second gas flow and the second pressure.

9. A gas leakage detection system for an exhalation valve according to claim 1, wherein said gas input means further comprises: a pressure reducing valve;

the pressure reducing valve is used for adjusting the flow and pressure of the gas input into the longitudinal cavity and the transverse cavity.

Images