Breathing heating pipeline fault monitoring method and device and electronic equipment

Abstract

The invention provides a breathing heating pipeline fault monitoring method and device and electronic equipment, and relates to the field of fault monitoring, and the method comprises the steps: determining a lower heating wire corresponding to each heating pipeline unit according to the current heating wire heating current corresponding to each heating pipeline unit in a breathing heating pipeline; inputting the current temperature corresponding to each heating pipeline unit to a preset calibration function to obtain a temperature output value corresponding to each heating pipeline unit output by the preset calibration function; and when any temperature output value is greater than or equal to a preset temperature threshold value, determining that the temperature sensor corresponding to the heating pipeline unit has a fault. According to the invention, the method achieves the determination of the preset calibration function through the nonlinear fitting, achieves the fault monitoring of each temperature sensor through the preset calibration function, improves the accuracy of fault monitoring, reduces the calculation complexity of fault monitoring, and improves the sensitivity of fault monitoring.

Description

Technical Field

[0001] The present invention relates to the field of fault monitoring, and in particular to a method, device and electronic equipment for monitoring faults in a breathing heating pipeline. Background Art

[0002] The existing heated breathing circuit monitors the temperature of the pipeline by placing a single gas temperature acquisition device (NTC thermistor) at the output end of the pipeline. This method is limited by the accuracy and inherent error of the NTC caused by the circuit board and the ambient temperature, and is difficult to calibrate. In addition, the single-point measurement does not accurately grasp the overall temperature of the entire pipeline, which may cause a small amount of condensed water to be generated locally far away from the sensor. During use, due to the change of flow in the pipeline and the uneven temperature caused by breathing, it becomes particularly difficult to accurately control the temperature with a single sensor, while multiple sensors make precise temperature control possible. If multiple sensors are arranged on the pipeline, the values ​​of the sensors can be calibrated with each other to reduce errors, and multiple sensors can bring more stable temperature control effects, with strong generalization capabilities. Multiple sensors can also better feedback the overall temperature of the pipeline and better prevent the generation of condensed water. However, the arrangement of multiple sensors may lead to more fault points.

[0003] A Chinese invention patent with publication number CN 110501094 A discloses a fault detection and calibration method for an electrical temperature sensor, an air conditioner, and a computer-readable storage medium, which obtains the temperature values ​​detected by each temperature sensor under the same temperature environment; determines a temperature sensor whose absolute value of the difference between the detected temperature value and the average temperature value detected by each temperature sensor is greater than a preset difference as a faulty sensor; determines a temperature sensor whose absolute value of the difference between the detected temperature value and the average temperature value detected by each temperature sensor is less than or equal to a preset difference as a normal sensor; calibrates the faulty temperature value detected by the faulty sensor with reference to the average temperature value detected by each temperature sensor; or, calibrates the faulty temperature value detected by the faulty sensor with reference to the average temperature value detected by the normal sensor. However, the monitoring means are relatively simple and cannot meet the monitoring requirements of respiratory heating pipeline failures.

[0004] However, there is currently no technical solution that can solve the above technical problems, and there is no method, device or electronic equipment for monitoring faults in a breathing heating pipeline. Summary of the invention

[0005] The present invention provides a method, device and electronic equipment for monitoring faults of a respiratory heating pipeline, which realize monitoring of temperature sensor faults through a preset calibration function determined by nonlinear fitting.

[0006] In a first aspect, the present invention provides a method for monitoring a respiratory heating circuit fault, comprising:

[0007] Determine the kurtosis value corresponding to each heating pipeline unit according to the current heating current of the heating wire corresponding to each heating pipeline unit in the breathing heating pipeline, wherein the breathing heating pipeline includes a plurality of heating pipeline units arranged at intervals along the length direction of the breathing heating pipeline, and each heating pipeline unit includes a heating wire spirally fixed to the outer surface of the breathing heating pipeline, a current acquisition unit for obtaining the heating current of the heating wire, and a temperature sensor for obtaining the current temperature;

[0008] When the kurtosis value corresponding to each heating pipeline unit is less than the preset kurtosis, the current temperature corresponding to each heating pipeline unit is input into the preset calibration function to obtain the temperature output value corresponding to each heating pipeline unit output by the preset calibration function;

[0009] When any temperature output value is greater than or equal to a preset temperature threshold, it is determined that the temperature sensor corresponding to the heating pipeline unit fails;

[0010] Among them, the preset calibration function performs nonlinear fitting on the sample temperatures corresponding to all heating pipeline units and the expected sample temperatures, determines the initial logarithmic term coefficient, the initial first sine term frequency, the initial second sine term frequency, and the initial calibration function intercept, and uses the gradient descent method to update the parameters until the preset conditions are met, and then determines them according to the target logarithmic term coefficient, the target first sine term frequency, the target second sine term frequency, and the target calibration function intercept.

[0011] According to the respiratory heating pipeline fault monitoring method provided by the present invention, after determining the kurtosis value corresponding to each heating pipeline unit according to the current heating wire heating current corresponding to each heating pipeline unit in the respiratory heating pipeline, the method further includes:

[0012] When the kurtosis value corresponding to any heating pipeline unit is greater than the preset kurtosis, it is determined that the heating wire corresponding to the heating pipeline unit fails, and a first indication instruction is generated;

[0013] The first indication instruction is used to instruct to cut off the heating current of the heating wire corresponding to the heating pipeline unit, obtain the current temperature corresponding to each heating pipeline unit again, and obtain the temperature output value corresponding to each heating pipeline unit output by the preset calibration function again.

[0014] According to the respiratory heating circuit fault monitoring method provided by the present invention, before inputting the current temperature corresponding to each heating circuit unit into the preset calibration function, the method further includes:

[0015] Obtain the temperature data of each temperature sensor at different times and construct a sensor data matrix m×n, where m is the temperature data of different temperature sensors and n is different times;

[0016] The sensor data matrix is ​​determined as the sample temperatures corresponding to all heating pipeline units.

[0017] According to the respiratory heating circuit fault monitoring method provided by the present invention, the calculation formula of the preset calibration function is:

[0018] Y=Alog(X)+B sin(CX)+D

[0019] Among them, Y is the temperature output value, X is the current temperature corresponding to any heating pipeline unit, A is the target logarithmic term coefficient, B is the target first sinusoidal term frequency, C is the target second sinusoidal term frequency, and D is the target calibration function intercept.

[0020] According to the respiratory heating circuit fault monitoring method provided by the present invention, the preset conditions include that the number of times the parameter is updated using the gradient descent method reaches a preset number of iterations, or that the loss function after the parameter is updated using the gradient descent method is less than a preset threshold.

[0021] According to the respiratory heating circuit fault monitoring method provided by the present invention, when any temperature output value is greater than or equal to a preset temperature threshold, determining that the temperature sensor corresponding to the heating circuit unit has failed includes:

[0022] When any temperature output value is greater than or equal to the preset temperature threshold, repeat the following steps:

[0023] At intervals of a first preset time length, inputting the current temperature corresponding to the heating pipeline unit into a preset calibration function, obtaining a temperature output value corresponding to the heating pipeline unit output by the preset calibration function, and increasing a count value when the temperature output value is greater than or equal to a preset temperature threshold;

[0024] Until the accumulated count value exceeds the pre-designed value within the second preset time period, it is determined that the temperature sensor corresponding to the heating pipeline unit fails.

[0025] According to the respiratory heating circuit fault monitoring method provided by the present invention, after obtaining the temperature output value corresponding to each heating circuit unit output by the preset calibration function, the method further includes:

[0026] When any temperature output value is less than the preset temperature threshold, it is determined that the breathing heating circuit has no fault.

[0027] According to the respiratory heating pipeline fault monitoring method provided by the present invention, the kurtosis value corresponding to each heating pipeline unit in the respiratory heating pipeline is determined according to the current heating wire heating current corresponding to each heating pipeline unit in the respiratory heating pipeline, including:

[0028]

[0029] Where K is the kurtosis value, N is the number of heating pipeline units, I i For any current heating wire heating current, It is the average value of the current heating current of the heating wire.

[0030] In a second aspect, a respiratory heating pipeline fault monitoring device is provided, comprising:

[0031] a first determining unit, the first determining unit being used to determine a kurtosis value corresponding to each heating pipeline unit according to a current heating current of a heating wire corresponding to each heating pipeline unit in the breathing heating pipeline, the breathing heating pipeline comprising a plurality of heating pipeline units spaced apart along a length direction of the breathing heating pipeline, each heating pipeline unit comprising a heating wire spirally fixed to an outer surface of the breathing heating pipeline, a current acquisition unit for obtaining a heating current of the heating wire, and a temperature sensor for obtaining a current temperature;

[0032] An input unit, wherein when the kurtosis value corresponding to each heating pipeline unit is less than a preset kurtosis, the input unit is used to input the current temperature corresponding to each heating pipeline unit into a preset calibration function to obtain a temperature output value corresponding to each heating pipeline unit output by the preset calibration function;

[0033] a second determination unit, the second determination unit being used to determine that a temperature sensor corresponding to the heating pipeline unit fails when any temperature output value is greater than or equal to a preset temperature threshold;

[0034] Among them, the preset calibration function performs nonlinear fitting on the sample temperatures corresponding to all heating pipeline units and the expected sample temperatures, determines the initial logarithmic term coefficient, the initial first sine term frequency, the initial second sine term frequency, and the initial calibration function intercept, and uses the gradient descent method to update the parameters until the preset conditions are met, and then determines them according to the target logarithmic term coefficient, the target first sine term frequency, the target second sine term frequency, and the target calibration function intercept.

[0035] In a third aspect, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the method for monitoring a respiratory heating circuit fault is implemented when the processor executes the program.

[0036] In the process of monitoring the fault of the heating breathing circuit, since the fault monitoring of the temperature sensor needs to collect temperature data for a period of time, before troubleshooting the temperature sensor, the possibility of the heating wire failure is first eliminated, and then the temperature sensor failure monitoring is performed, thereby improving the fault monitoring efficiency and the timeliness of judgment; at the same time, the present invention uses nonlinear fitting to determine the preset calibration function, and then uses the preset calibration function to realize the fault monitoring of each temperature sensor, which not only improves the accuracy of the fault monitoring, but also reduces the calculation complexity of the fault monitoring, and improves the sensitivity of the fault monitoring;

[0037] The present invention proposes a method for monitoring faults in a heated breathing circuit based on multi-point segmented measurement of feature engineering, which has many advantages such as strong generalization ability, safety, over-temperature prevention, and accurate temperature control. The temperature control algorithm of the present invention can tolerate the uncertainty caused by the inherent error of NTC during the production process, and the temperature control is more accurate, and the generation of condensed water is better prevented. It has accurate fault diagnosis ability and can accurately identify failed sensors or heating wires. Once a sensor failure is detected, the system will stop working immediately to ensure safety. In after-sales service, only the damaged sensor needs to be replaced, thereby effectively reducing the maintenance cost and also reducing the cost of use for consumers. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0039] Figure 1 It is a flow chart of the breathing heating pipeline fault monitoring method provided by the present invention;

[0040] Figure 2 It is a structural schematic diagram of the breathing heating pipeline fault monitoring device provided by the present invention;

[0041] Figure 3 It is a functional block diagram of the breathing heating pipeline fault monitoring device provided by the present invention;

[0042] Figure 4 It is a structural schematic diagram of the breathing heating pipeline provided by the present invention;

[0043] Figure 5 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] Figure 1 : is a flow chart of a method for monitoring a breathing heating pipeline fault provided by the present invention, wherein the method for monitoring a breathing heating pipeline fault comprises:

[0046] Step 101, determining a kurtosis value corresponding to each heating pipeline unit according to a current heating current of a heating wire corresponding to each heating pipeline unit in the breathing heating pipeline, wherein the breathing heating pipeline comprises a plurality of heating pipeline units spaced apart along a length direction of the breathing heating pipeline, and each heating pipeline unit comprises a heating wire spirally fixed to an outer surface of the breathing heating pipeline, a current acquisition unit for obtaining a heating current of the heating wire, and a temperature sensor for obtaining a current temperature;

[0047] Step 102: when the kurtosis value corresponding to each heating pipeline unit is less than the preset kurtosis, input the current temperature corresponding to each heating pipeline unit into the preset calibration function to obtain the temperature output value corresponding to each heating pipeline unit output by the preset calibration function;

[0048] Step 103: When any temperature output value is greater than or equal to a preset temperature threshold, it is determined that a temperature sensor corresponding to the heating pipeline unit fails;

[0049] Among them, the preset calibration function performs nonlinear fitting on the sample temperatures corresponding to all heating pipeline units and the expected sample temperatures, determines the initial logarithmic term coefficient, the initial first sine term frequency, the initial second sine term frequency, and the initial calibration function intercept, and uses the gradient descent method to update the parameters until the preset conditions are met, and then determines them according to the target logarithmic term coefficient, the target first sine term frequency, the target second sine term frequency, and the target calibration function intercept.

[0050] In step 101, the breathing heating pipeline is used to connect the air circuit between the ventilator and the anesthesia machine and the mask or endotracheal tube. The purpose is to prevent the generation of condensed water by heating the gas in the pipeline, thereby reducing the discomfort to the patient. The fault monitoring method of the present invention is mainly used in the breathing heating pipeline matched with the ventilator, such as Figure 4 As shown, Figure 4It is a structural schematic diagram of the breathing heating pipeline provided by the present invention. Each heating pipeline unit is equipped with a heating wire spirally fixed on the outer surface of the breathing heating pipeline, and a current acquisition unit for obtaining the heating current of the heating wire. These heating wires are heated by alternating current to ensure the uniformity of heating and the accuracy of control. The current acquisition unit collects the heating current of the heating wire of each heating pipeline unit in real time and transmits it to the central processing unit (CPU) for analysis. The CPU calculates the kurtosis value of each heating pipeline unit based on the collected heating current data of the heating wire. The kurtosis is an indicator that describes the random distribution characteristics of the signal. When the heating wire fails, the kurtosis value will change, usually showing an increase. The present invention can use the preset kurtosis value as a threshold for judging whether the heating wire fails.

[0051] Optionally, determining the kurtosis value corresponding to each heating circuit unit according to the current heating wire heating current corresponding to each heating circuit unit in the breathing heating circuit includes:

[0052]

[0053] Where K is the kurtosis value, N is the number of heating pipeline units, I i For any current heating wire heating current, It is the average value of the current heating current of the heating wire.

[0054] Optionally, after determining the kurtosis value corresponding to each heating circuit unit according to the current heating wire heating current corresponding to each heating circuit unit in the breathing heating circuit, the method further includes:

[0055] When the kurtosis value corresponding to any heating pipeline unit is greater than the preset kurtosis, it is determined that the heating wire corresponding to the heating pipeline unit fails, and a first indication instruction is generated;

[0056] The first indication instruction is used to instruct to cut off the heating current of the heating wire corresponding to the heating pipeline unit, obtain the current temperature corresponding to each heating pipeline unit again, and obtain the temperature output value corresponding to each heating pipeline unit output by the preset calibration function again.

[0057] Optionally, based on the above-mentioned breathing heating circuit fault monitoring method, the present invention adds a step for handling heating wire faults. Specifically, after the kurtosis value corresponding to each heating circuit unit is determined according to the current heating wire heating current corresponding to each heating circuit unit in the breathing heating circuit, if the kurtosis value of any heating circuit unit is greater than the preset kurtosis, it is judged that the heating wire of the heating circuit unit may have failed. The abnormal increase in the kurtosis value usually indicates that the working state of the heating wire is unstable, which may be caused by aging, breakage or poor connection. The CPU generates the first indication instruction, which is used to instruct the system to cut off the heating current of the heating wire corresponding to the faulty heating circuit unit, and cut off the heating current of the faulty heating circuit unit through a relay or other control element. This is to prevent more serious faults that may be caused by the continued operation of the faulty heating wire, such as short circuit, overheating, etc. After cutting off the problematic heating wire, the current temperature data of each heating circuit unit (excluding the faulty unit whose heating current has been cut off) is obtained again, and then it is evaluated whether there is still a temperature sensor fault in other heating circuit units in the entire heating circuit system after the faulty heating wire is cut off. Specifically, the CPU inputs the re-acquired temperature data into a preset calibration function, and recalculates the temperature output value corresponding to each heating pipeline unit to verify whether the temperature data of the remaining heating pipeline units are still accurate and reliable after the faulty heating wire is cut off.

[0058] In step 102, a temperature sensor for obtaining the current temperature is installed in each heating pipeline unit. The temperature sensor collects the temperature data in the pipeline in real time and transmits it to the CPU. The CPU includes a preset calibration function, which is obtained by nonlinear fitting of the sample temperature and the expected sample temperature corresponding to all heating pipeline units. During the fitting process, the initial logarithmic term coefficient, the initial first sine term frequency, the initial second sine term frequency, the initial calibration function intercept and other parameters are determined, and the parameters are updated by the gradient descent method until the preset conditions are met (such as the loss function value is less than the specified threshold or the predetermined number of iterations is reached), and the target parameter value is obtained, thereby determining the final calibration function. After confirming that the kurtosis values ​​of all heating pipeline units are less than the preset kurtosis value, the CPU inputs the current temperature corresponding to each heating pipeline unit into the preset calibration function to obtain the temperature output value corresponding to each heating pipeline unit.

[0059] Optionally, before inputting the current temperature corresponding to each heating pipeline unit into the preset calibration function, the method further includes:

[0060] Obtain the temperature data of each temperature sensor at different times and construct a sensor data matrix m×n, where m is the temperature data of different temperature sensors and n is different times;

[0061] The sensor data matrix is ​​determined as the sample temperatures corresponding to all heating pipeline units.

[0062] Optionally, at different times (for example, every minute, every hour, etc.), each temperature sensor is used to record temperature data, and the temperature data of all temperature sensors are organized into a matrix, wherein the rows (m) of the matrix represent different temperature sensors (i.e., sensors on different heating pipeline units), and the columns (n) of the matrix represent different times (i.e., data recorded at different time points). Assuming there are m sensors and n times, the sensor data matrix is ​​an m×n matrix, and each element in the matrix represents the temperature value of a specific sensor at a specific time. This sensor data matrix is ​​determined as the sample temperature corresponding to all heating pipeline units.

[0063] Optionally, the calculation formula of the preset calibration function is:

[0064] Y=A log(X)+B sin(CX)+D

[0065] Among them, Y is the temperature output value, X is the current temperature corresponding to any heating pipeline unit, A is the target logarithmic term coefficient, B is the target first sinusoidal term frequency, C is the target second sinusoidal term frequency, and D is the target calibration function intercept.

[0066] Optionally, the preset condition includes that the number of times the parameter is updated using the gradient descent method reaches a preset number of iterations, or that the loss function after the parameter is updated using the gradient descent method is less than a preset threshold.

[0067] Optionally, the preset calibration function of the present invention is determined in the following manner:

[0068] Assuming that the breathing heating circuit has n heating circuit units equipped with heating wires and temperature sensors, then we have n groups of heating wires and n groups of temperature sensors. Generally speaking, under ideal conditions, when the heating wire has constant power, the longer the gas stays in the tube, the faster the gas heats up, but it will eventually stabilize near the heating temperature of the heating wire. Use these n groups of sensor units to collect data for a period of time, the number of points of these data is m, and this data is organized into a sensor data matrix X, where X is an m×n matrix, that is, the temperature data of each temperature sensor at different times is obtained, and a sensor data matrix m×n is constructed, where m is the temperature data of different temperature sensors and n is different times; the sensor data matrix is ​​determined as the sample temperature corresponding to all heating circuit units.

[0069] Since the model is actually very complex, a nonlinear fitting function can more effectively reflect the actual situation. Since the error of the NTC resistor used has a certain periodic relationship, adding a sine and cosine function to reflect its periodic changes can better fit the actual situation. The fitting function is as follows:

[0070] T=E log(J)+F sin(GJ)+H

[0071] Wherein, Y is the expected sample temperature, X is the sample temperature, E is the initial logarithmic term coefficient, F is the initial first sine term frequency, G is the initial second sine term frequency, and H is the initial calibration function intercept.

[0072] Optionally, nonlinear fitting is performed on the sample temperatures corresponding to all heating pipeline units and the expected sample temperatures to determine the initial logarithmic term coefficient, the initial first sinusoidal term frequency, the initial second sinusoidal term frequency, and the initial calibration function intercept, and the parameters are updated using the gradient descent method until the preset conditions are met. The preset calibration function is determined based on the target logarithmic term coefficient, the target first sinusoidal term frequency, the target second sinusoidal term frequency, and the target calibration function intercept.

[0073] The loss function is the Huber loss function, which combines the mean squared error and the mean absolute error:

[0074]

[0075] Among them, δ is a hyperparameter that controls the switching point between the mean square error and the mean absolute error. According to experience, the value of δ can be set to 0.25. i is the actual observed value, target value, or known value of the ith data point in the loss function L(A,B,C,D), which is a scalar and is compared with the observed value f(X i ,A,B,C,D) for comparison, the observed value f(X i , A, B, C, D) is based on the model parameters A, B, C, D and the input data X i The calculated predicted values, the loss function L(A,B,C,D) measures the difference between the observed values ​​and the known values. Specifically, the loss function combines the mean square error and the mean absolute error to handle outliers.

[0076] X i It is the matrix X when the time series is i. X is an m×n matrix, that is, the temperature data of each temperature sensor at different times is obtained, and the sensor data matrix m×n is constructed. i is the sequence number of the data in m. Yi in the iteration represents the value calculated in the last iteration. m is the data collected for a period of time using these n sets of sensor units. The number of these data points is m.

[0077] During the gradient calculation:

[0078]

[0079] x is the independent variable sign(X) is a function.

[0080] The gradient of the target logarithmic term coefficient A:

[0081]

[0082] Gradient of the target first sinusoidal frequency B:

[0083]

[0084] Gradient of the target second sinusoidal frequency C:

[0085]

[0086] The gradient of the target calibration function intercept D is:

[0087]

[0088] is the partial derivative of the loss function L.

[0089] Furthermore, the gradient descent method is used to update the parameters:

[0090]

[0091] Among them, η is a hyperparameter, called the learning rate.

[0092] During the iteration process, the steps of calculating gradients and updating parameters are repeated until the specified number of iterations is reached or the loss function value reaches the specified threshold. For each iteration t from t = 0 to T-1, the following steps are performed:

[0093] Calculate predicted values

[0094]

[0095] Calculate the loss function L (t) :

[0096]

[0097] Compute the gradient:

[0098]

[0099] Update parameters:

[0100]

[0101] Check the convergence condition. If the loss function L (t) If the value is less than the expected threshold ∈, or reaches the predetermined number of iterations T, the iteration is stopped. The learning rate η is generally taken as 0.1 based on experience. Denotes the loss function L (t) The partial derivative with respect to parameter A is Denotes the loss function L (t) The partial derivative with respect to parameter B is Denotes the loss function L (t) The partial derivative with respect to parameter C is Denotes the loss function L (t) The partial derivative with respect to the parameter D.

[0102] In step 103, the present invention can set a preset temperature threshold as a reference standard for determining whether the temperature sensor fails. The CPU compares the temperature output value corresponding to each heating pipeline unit with the preset temperature threshold. If any temperature output value is greater than or equal to the preset temperature threshold, it is considered that the temperature sensor corresponding to the heating pipeline unit fails. Once a temperature sensor failure is detected, the system will immediately stop the operation of the heating system to ensure safety. At the same time, the system will record the fault information and guide the user or maintenance personnel to replace the damaged temperature sensor in after-sales service.

[0103] Optionally, when any temperature output value is greater than or equal to a preset temperature threshold, determining that a temperature sensor corresponding to the heating pipeline unit fails includes:

[0104] When any temperature output value is greater than or equal to the preset temperature threshold, repeat the following steps:

[0105] At intervals of a first preset time length, inputting the current temperature corresponding to the heating pipeline unit into a preset calibration function, obtaining a temperature output value corresponding to the heating pipeline unit output by the preset calibration function, and increasing a count value when the temperature output value is greater than or equal to a preset temperature threshold;

[0106] Until the accumulated count value exceeds the pre-designed value within the second preset time period, it is determined that the temperature sensor corresponding to the heating pipeline unit fails.

[0107] Optionally, the system first obtains the temperature value output by the temperature sensor of the heating pipeline unit, and determines whether the temperature output value is greater than or equal to a preset temperature threshold. If so, the system proceeds to the next step; otherwise, the system considers that the temperature sensor is currently working normally and continues to monitor normally. When the temperature output value is greater than or equal to the preset temperature threshold for the first time, the system begins to repeat the following steps:

[0108] At intervals of a first preset time length (for example, every 1 minute or 5 minutes), the current temperature value of the heating pipeline unit is obtained, and the current temperature value is input into a preset calibration function. The function is designed based on known temperature sensor characteristics and environmental factors (such as ambient temperature, sensor aging, etc.), and is used to calibrate the original temperature value to obtain a more accurate temperature output value. It is checked again whether the temperature output value after processing by the preset calibration function is still greater than or equal to the preset temperature threshold. If so, the count value is increased to record the number of temperature anomalies.

[0109] Within the second preset time period, for example, 1 hour or 24 hours, the system continues to perform the above-mentioned repeated detection. If within this time period, the accumulated count value exceeds the pre-designed value, for example, 5 times or 10 times, the system determines that the temperature sensor corresponding to the heating pipeline unit has failed. By introducing a calibration function and a repeated detection mechanism, the present invention can more accurately identify whether the temperature sensor has failed, avoiding misjudgment caused by a single misdetection or a short-term fluctuation. By multiple detections and cumulative count values, it can effectively reduce false alarms caused by a single abnormal reading, while ensuring that an alarm can be issued in time when the sensor does fail, reducing the possibility of missed reports.

[0110] Optionally, after obtaining the temperature output value corresponding to each heating pipeline unit output by the preset calibration function, the method further includes:

[0111] When any temperature output value is less than the preset temperature threshold, it is determined that the breathing heating circuit has no fault.

[0112] Optionally, after obtaining the original temperature values ​​output by the temperature sensor of the heating circuit unit, these original temperature values ​​are first input into a preset calibration function to obtain the temperature output values ​​corresponding to each heating circuit unit output by the preset calibration function. If the temperature output values ​​of all heating circuit units are less than the preset temperature threshold, it is considered that the current breathing heating circuit is in a normal working state and no fault has occurred. By continuously monitoring and judging the calibrated temperature output values, the present invention enables the system to promptly detect and respond to possible fault conditions, thereby avoiding further development of the fault and causing system shutdown or damage.

[0113] If the fault monitoring method provided by the present invention cannot be calibrated within a period of time, it is considered that the information collected by the acquisition unit is unreliable. Since the judgment of the temperature sensor fault requires the collection of temperature data for a period of time for statistical analysis, the timeliness of the judgment is poor. This is because the present invention uses a heating wire for heating, and the heating wire transfers heat to the gas in the tube, and the gas in the tube transfers heat to the sensor, and the sensor finally performs data analysis. The whole process is relatively time-consuming, so it is necessary to introduce a prior heating wire fault judgment mechanism to improve the timeliness of the judgment.

[0114] Specifically, the present invention monitors the heating current of the heating wire of each acquisition unit and performs analysis. Compared with direct current heating, using alternating current for heating has multiple advantages. Alternating current heating can achieve uniform heating through the principle of electromagnetic induction to avoid local overheating or overcooling. The control of the alternating current heating system is relatively simple, and accurate temperature control can be achieved by adjusting the voltage, current or frequency, with high reliability and stability. The present invention uses alternating current heating to accurately control the heating amount of the acquisition unit and calculate the kurtosis. When a fault arc occurs, the kurtosis of the current signal will increase. The heating current of the heating wire of each acquisition unit is collected and the kurtosis analysis is performed on it. The kurtosis of the heating wire will increase when a fault occurs compared with that in normal use. When the heating wire is excited at a specified frequency and the kurtosis analysis is performed, the kurtosis amplitude will be very low. However, when the heating wire fails, such as poor contact or other faults, the kurtosis will increase. Whether a fault occurs is determined by judging whether the amplitude of the component of the kurtosis increase reaches a specified threshold. If a fault occurs, the heating current can be quickly cut off.

[0115] In the process of monitoring the fault of the heating breathing circuit, since the fault monitoring of the temperature sensor needs to collect temperature data for a period of time, before troubleshooting the temperature sensor, the possibility of the heating wire failure is first eliminated, and then the temperature sensor failure monitoring is performed, thereby improving the fault monitoring efficiency and the timeliness of judgment; at the same time, the present invention uses nonlinear fitting to determine the preset calibration function, and then uses the preset calibration function to realize the fault monitoring of each temperature sensor, which not only improves the accuracy of the fault monitoring, but also reduces the calculation complexity of the fault monitoring, and improves the sensitivity of the fault monitoring;

[0116] A method for monitoring faults in heated breathing circuits based on multi-point segmented measurement of feature engineering is proposed. It has many advantages such as strong generalization ability, safety, over-temperature prevention, and accurate temperature control. The temperature control algorithm of the present invention can tolerate the uncertainty caused by the inherent error of NTC during the production process, and the temperature control is more accurate, and the generation of condensed water is better prevented. It has accurate fault diagnosis capabilities and can accurately identify failed sensors or heating wires. Once a sensor failure is detected, the system will stop working immediately to ensure safety. In after-sales service, only the damaged sensor needs to be replaced, which effectively reduces the maintenance cost and also reduces the cost of use for consumers.

[0117] Figure 2It is a structural schematic diagram of the breathing heating pipeline fault monitoring device provided by the present invention, the breathing heating pipeline fault monitoring device includes a first determination unit 1, the first determination unit is used to determine the kurtosis value corresponding to each heating pipeline unit according to the current heating current of the heating wire corresponding to each heating pipeline unit in the breathing heating pipeline, the breathing heating pipeline includes a plurality of heating pipeline units arranged at intervals along the length direction of the breathing heating pipeline, each heating pipeline unit includes a heating wire spirally fixed to the outer surface of the breathing heating pipeline, a current acquisition unit for obtaining the heating current of the heating wire, and a temperature sensor for obtaining the current temperature. The working principle of the first determination unit 1 can refer to the aforementioned step 101 and will not be repeated here.

[0118] The respiratory heating circuit fault monitoring device also includes an input unit 2, which is used to input the current temperature corresponding to each heating circuit unit into a preset calibration function when the kurtosis value corresponding to each heating circuit unit is less than a preset kurtosis, so as to obtain the temperature output value corresponding to each heating circuit unit output by the preset calibration function. The working principle of the input unit 2 can refer to the aforementioned step 102 and will not be repeated here.

[0119] The respiratory heating circuit fault monitoring device also includes a second determination unit 3, which is used to determine that the temperature sensor corresponding to the heating circuit unit has failed when any temperature output value is greater than or equal to a preset temperature threshold. The working principle of the second determination unit 3 can refer to the aforementioned step 103 and will not be repeated here.

[0120] Among them, the preset calibration function performs nonlinear fitting on the sample temperatures corresponding to all heating pipeline units and the expected sample temperatures, determines the initial logarithmic term coefficient, the initial first sine term frequency, the initial second sine term frequency, and the initial calibration function intercept, and uses the gradient descent method to update the parameters until the preset conditions are met, and then determines them according to the target logarithmic term coefficient, the target first sine term frequency, the target second sine term frequency, and the target calibration function intercept.

[0121] In the process of monitoring the fault of the heating breathing circuit, since the fault monitoring of the temperature sensor needs to collect temperature data for a period of time, before troubleshooting the temperature sensor, the possibility of the heating wire failure is first eliminated, and then the temperature sensor failure monitoring is performed, thereby improving the fault monitoring efficiency and the timeliness of judgment; at the same time, the present invention uses nonlinear fitting to determine the preset calibration function, and then uses the preset calibration function to realize the fault monitoring of each temperature sensor, which not only improves the accuracy of the fault monitoring, but also reduces the calculation complexity of the fault monitoring, and improves the sensitivity of the fault monitoring;

[0122] A method for monitoring faults in heated breathing circuits based on multi-point segmented measurement of feature engineering is proposed. It has many advantages such as strong generalization ability, safety, over-temperature prevention, and accurate temperature control. The temperature control algorithm of the present invention can tolerate the uncertainty caused by the inherent error of NTC during the production process, and the temperature control is more accurate, and the generation of condensed water is better prevented. It has accurate fault diagnosis capabilities and can accurately identify failed sensors or heating wires. Once a sensor failure is detected, the system will stop working immediately to ensure safety. In after-sales service, only the damaged sensor needs to be replaced, which effectively reduces the maintenance cost and also reduces the cost of use for consumers.

[0123] Figure 3 is a functional block diagram of the breathing heating pipeline fault monitoring device provided by the present invention, such as Figure 3 As shown, after the multi-channel sensor information is collected, the temperature control algorithm is used to realize the multi-channel resistance wire heating. The more accurate algorithm can better control the temperature of the heating pipeline, improve the heating efficiency, reduce the heating power consumption of the ventilator and better prevent the generation of condensed water. In the present invention, the breathing heating pipeline is composed of a plurality of heating pipeline units equipped with heating wires and temperature sensors. The heating wires of each heating pipeline unit are independently connected to the temperature control unit, so the temperature control unit can control the temperature of each heating pipeline unit separately. During the use of the existing scheme, due to the change of flow in the pipeline and the uneven temperature caused by breathing, it becomes particularly difficult for a single sensor to accurately control the temperature. A single NTC resistor cannot be self-calibrated, but multiple temperature sensors can be referenced to each other for self-calibration. The present invention realizes accurate judgment of the state of the breathing heating pipeline by introducing a calibration function and a continuous monitoring mechanism, and improves the stability, reliability and user experience of the system.

[0124] Figure 5 Schematic diagram of the structure of the electronic device provided by the present invention. Figure 5As shown, the electronic device may include: a processor 110, a communication interface 120, a memory 130 and a communication bus 140, wherein the processor 110, the communication interface 120 and the memory 130 communicate with each other through the communication bus 140. The processor 110 may call the logic instructions in the memory 130 to execute the respiratory heating circuit fault monitoring method, the method comprising: determining the kurtosis value corresponding to each heating circuit unit according to the current heating current of the heating wire corresponding to each heating circuit unit in the respiratory heating circuit, the respiratory heating circuit comprising a plurality of heating circuit units spaced apart along the length direction of the respiratory heating circuit, each heating circuit unit comprising a heating wire spirally fixed on the outer surface of the respiratory heating circuit, a current acquisition unit for obtaining the heating current of the heating wire and a temperature sensor for obtaining the current temperature; when the kurtosis value corresponding to each heating circuit unit is less than the preset kurtosis, inputting the current value corresponding to each heating circuit unit. Temperature to a preset calibration function, and obtain the temperature output value corresponding to each heating pipeline unit output by the preset calibration function; when any temperature output value is greater than or equal to the preset temperature threshold, it is determined that the temperature sensor corresponding to the heating pipeline unit is faulty; wherein the preset calibration function is to perform nonlinear fitting on the sample temperatures corresponding to all heating pipeline units and the expected sample temperatures, and determine the initial logarithmic term coefficient, the initial first sine term frequency, the initial second sine term frequency, and the initial calibration function intercept, and use the gradient descent method to update the parameters until the preset conditions are met, and then the target logarithmic term coefficient, the target first sine term frequency, the target second sine term frequency, and the target calibration function intercept are determined.

[0125] In addition, the logic instructions in the above-mentioned memory 130 can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0126] On the other hand, the present invention further provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute a respiratory heating circuit fault monitoring method provided by the above methods, the method including: determining the kurtosis value corresponding to each heating circuit unit according to the current heating current of the heating wire corresponding to each heating circuit unit in the respiratory heating circuit, the respiratory heating circuit including a plurality of heating circuit units spaced apart along the length direction of the respiratory heating circuit, each heating circuit unit including a heating wire spirally fixed to the outer surface of the respiratory heating circuit, a current acquisition unit for obtaining the heating current of the heating wire, and a temperature sensor for obtaining the current temperature; in each heating circuit unit When the kurtosis values ​​corresponding to the elements are all less than the preset kurtosis, the current temperature corresponding to each heating pipeline unit is input into the preset calibration function to obtain the temperature output value corresponding to each heating pipeline unit output by the preset calibration function; when any temperature output value is greater than or equal to the preset temperature threshold, it is determined that the temperature sensor corresponding to the heating pipeline unit is faulty; wherein the preset calibration function performs nonlinear fitting on the sample temperatures corresponding to all heating pipeline units and the expected sample temperatures, determines the initial logarithmic term coefficient, the initial first sine term frequency, the initial second sine term frequency, and the initial calibration function intercept, and uses the gradient descent method to update the parameters until the preset conditions are met, and then the parameters are determined according to the target logarithmic term coefficient, the target first sine term frequency, the target second sine term frequency, and the target calibration function intercept.

[0127] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the respiratory heating circuit fault monitoring method provided by the above-mentioned methods, the method comprising: determining the kurtosis value corresponding to each heating circuit unit according to the current heating current of the heating wire corresponding to each heating circuit unit in the respiratory heating circuit, the respiratory heating circuit comprising a plurality of heating circuit units spaced apart along the length direction of the respiratory heating circuit, each heating circuit unit comprising a heating wire spirally fixed to the outer surface of the respiratory heating circuit, a current acquisition unit for obtaining the heating current of the heating wire, and a temperature sensor for obtaining the current temperature; in the case where the kurtosis value corresponding to each heating circuit unit is less than a preset kurtosis value, Under this condition, the current temperature corresponding to each heating pipeline unit is input into the preset calibration function to obtain the temperature output value corresponding to each heating pipeline unit output by the preset calibration function; when any temperature output value is greater than or equal to the preset temperature threshold, it is determined that the temperature sensor corresponding to the heating pipeline unit is faulty; wherein, the preset calibration function performs nonlinear fitting on the sample temperatures corresponding to all heating pipeline units and the expected sample temperatures, determines the initial logarithmic term coefficient, the initial first sine term frequency, the initial second sine term frequency, and the initial calibration function intercept, and uses the gradient descent method to update the parameters until the preset conditions are met, and then the parameters are determined according to the target logarithmic term coefficient, the target first sine term frequency, the target second sine term frequency, and the target calibration function intercept.

[0128] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0129] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0130] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for monitoring a respiratory heating circuit fault, characterized in that: include: Determine the kurtosis value corresponding to each heating pipeline unit according to the current heating current of the heating wire corresponding to each heating pipeline unit in the breathing heating pipeline, wherein the breathing heating pipeline includes a plurality of heating pipeline units arranged at intervals along the length direction of the breathing heating pipeline, and each heating pipeline unit includes a heating wire spirally fixed to the outer surface of the breathing heating pipeline, a current acquisition unit for obtaining the heating current of the heating wire, and a temperature sensor for obtaining the current temperature; When the kurtosis value corresponding to each heating pipeline unit is less than the preset kurtosis, the current temperature corresponding to each heating pipeline unit is input into the preset calibration function to obtain the temperature output value corresponding to each heating pipeline unit output by the preset calibration function; When any temperature output value is greater than or equal to a preset temperature threshold, it is determined that the temperature sensor corresponding to the heating pipeline unit fails; Among them, the preset calibration function performs nonlinear fitting on the sample temperatures corresponding to all heating pipeline units and the expected sample temperatures, determines the initial logarithmic term coefficient, the initial first sine term frequency, the initial second sine term frequency, and the initial calibration function intercept, and uses the gradient descent method to update the parameters until the preset conditions are met, and then determines them according to the target logarithmic term coefficient, the target first sine term frequency, the target second sine term frequency, and the target calibration function intercept.

2. The method for monitoring a breathing heating circuit fault according to claim 1, characterized in that: After determining the kurtosis value corresponding to each heating circuit unit according to the current heating wire heating current corresponding to each heating circuit unit in the breathing heating circuit, the method further includes: When the kurtosis value corresponding to any heating pipeline unit is greater than the preset kurtosis, it is determined that the heating wire corresponding to the heating pipeline unit is faulty, and a first indication instruction is generated; The first indication instruction is used to instruct to cut off the heating current of the heating wire corresponding to the heating pipeline unit, obtain the current temperature corresponding to each heating pipeline unit again, and obtain the temperature output value corresponding to each heating pipeline unit output by the preset calibration function again.

3. The method for monitoring a breathing heating circuit fault according to claim 1, characterized in that: Before inputting the current temperature corresponding to each heating pipeline unit into the preset calibration function, the method further includes: Obtain the temperature data of each temperature sensor at different times and construct a sensor data matrix m×n, where m is the temperature data of different temperature sensors and n is different times; The sensor data matrix is ​​determined as the sample temperatures corresponding to all heating pipeline units.

4. The method for monitoring a breathing heating circuit fault according to claim 3, characterized in that: The calculation formula of the preset calibration function is: Y=Alog(X)+B sin(CX)+D Among them, Y is the temperature output value, X is the current temperature corresponding to any heating pipeline unit, A is the target logarithmic term coefficient, B is the target first sinusoidal term frequency, C is the target second sinusoidal term frequency, and D is the target calibration function intercept.

5. The method for monitoring respiratory heating circuit fault according to claim 1, characterized in that: The preset conditions include that the number of times the parameters are updated using the gradient descent method reaches a preset number of iterations, or that the loss function after the parameters are updated using the gradient descent method is less than a preset threshold.

6. The method for monitoring a breathing heating circuit fault according to claim 1, characterized in that: The step of determining that a temperature sensor corresponding to the heating pipeline unit fails when any temperature output value is greater than or equal to a preset temperature threshold comprises: When any temperature output value is greater than or equal to the preset temperature threshold, repeat the following steps: At intervals of a first preset time length, inputting the current temperature corresponding to the heating pipeline unit into a preset calibration function, obtaining a temperature output value corresponding to the heating pipeline unit output by the preset calibration function, and increasing a count value when the temperature output value is greater than or equal to a preset temperature threshold; Until the accumulated count value exceeds the pre-designed value within the second preset time period, it is determined that the temperature sensor corresponding to the heating pipeline unit fails.

7. The method for monitoring a respiratory heating circuit fault according to claim 1, characterized in that: After obtaining the temperature output value corresponding to each heating pipeline unit output by the preset calibration function, the method further includes: When any temperature output value is less than the preset temperature threshold, it is determined that the breathing heating circuit has no fault.

8. The method for monitoring failure of a breathing heating circuit according to claim 1, characterized in that: The method of determining the kurtosis value corresponding to each heating line unit according to the current heating current of the heating wire corresponding to each heating line unit in the breathing heating line comprises: Where K is the kurtosis value, N is the number of heating pipeline units, I i For any current heating wire heating current, It is the average value of the current heating current of the heating wire.

9. A breathing heating pipeline fault monitoring device, characterized in that: include: a first determining unit, the first determining unit being used to determine a kurtosis value corresponding to each heating pipeline unit according to a current heating current of a heating wire corresponding to each heating pipeline unit in the breathing heating pipeline, the breathing heating pipeline comprising a plurality of heating pipeline units spaced apart along a length direction of the breathing heating pipeline, each heating pipeline unit comprising a heating wire spirally fixed to an outer surface of the breathing heating pipeline, a current acquisition unit for obtaining a heating current of the heating wire, and a temperature sensor for obtaining a current temperature; An input unit, wherein when the kurtosis value corresponding to each heating pipeline unit is less than a preset kurtosis, the input unit is used to input the current temperature corresponding to each heating pipeline unit into a preset calibration function to obtain a temperature output value corresponding to each heating pipeline unit output by the preset calibration function; a second determination unit, the second determination unit being used to determine that a temperature sensor corresponding to the heating pipeline unit fails when any temperature output value is greater than or equal to a preset temperature threshold; Among them, the preset calibration function performs nonlinear fitting on the sample temperatures corresponding to all heating pipeline units and the expected sample temperatures, determines the initial logarithmic term coefficient, the initial first sine term frequency, the initial second sine term frequency, and the initial calibration function intercept, and uses the gradient descent method to update the parameters until the preset conditions are met, and then determines them according to the target logarithmic term coefficient, the target first sine term frequency, the target second sine term frequency, and the target calibration function intercept.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the breathing heating circuit fault monitoring method as described in any one of claims 1 to 8 is implemented.

Images