A method and system for predicting failure of an emergency transport ventilator

Abstract

The application discloses an emergency transport respirator fault prediction method and system, and the method comprises the following steps: a fault prediction system acquires a group of operation and maintenance records of emergency transport respirators, obtains a fault prediction data set and a fault local data set; the fault prediction data set is used to calculate the damage value and the fault value of the emergency transport respirator; the fault local data set is used to calculate the prediction parameter of the emergency transport respirator; the consistency of the secondary fault prediction data of the suspected fault module is judged; and the fault prediction result is output. When the fault prediction of the emergency transport respirator is performed, not only the operation and maintenance records of the current emergency transport respirator are used, but also the operation and maintenance records of other emergency transport respirators of the same type are used; through the first-level prediction of the prediction parameter and the second-level prediction of the consistency, multi-dimensional prediction of multi-dimensional information is realized, and the accuracy of the fault prediction is effectively improved.

Description

Technical Field

[0001] This invention belongs to the field of fault prediction technology, and in particular relates to a method and system for predicting faults in emergency transport ventilators. Background Technology

[0002] In modern medical systems, emergency transport ventilators, as crucial emergency equipment, play an irreplaceable role in resuscitation and patient transport. Therefore, ensuring that emergency transport ventilators are in normal working order before use is of paramount importance.

[0003] Currently, fault diagnosis of emergency transport ventilators often only reveals abnormalities after the instrument triggers a self-test alarm or even after shutdown. This not only severely impacts rescue efficiency but also increases the risk of serious accidents, posing a significant threat to patients' lives and health. Existing fault prediction methods rely solely on the device's own operational data, failing to fully utilize the vast amount of data from similar emergency ventilators, which is often crucial for accurate fault prediction. Furthermore, some methods do not achieve ideal prediction accuracy when dealing with complex and ever-changing real-world situations, making it difficult to meet the high-precision clinical requirements for ventilator fault prediction. Therefore, there is a need to develop a new fault prediction method for emergency transport ventilators. Summary of the Invention

[0004] This invention provides a method and system for predicting faults in emergency transport ventilators, in order to solve the technical problems of current emergency transport ventilator fault prediction data being singular, lacking sufficient dimensions for fault prediction, and having low accuracy in fault prediction.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0006] On one hand, the present invention provides a method for predicting the failure of an emergency transport ventilator, the method comprising:

[0007] S1: The fault prediction system acquires a set of operation and maintenance records of emergency transport ventilators, normalizes the operation and maintenance records, re-divides the functional modules of the emergency transport ventilators, calculates the secondary fault prediction data of each functional module, and obtains the fault prediction dataset and the fault local dataset.

[0008] S2: A multidimensional diagnostic model is established based on the fault prediction dataset, and the damage value and fault value of the emergency transport ventilator are calculated based on the secondary fault prediction data of each functional module.

[0009] S3: Calculate the remaining life coefficient and maintenance coefficient of the emergency transport ventilator from the fault local dataset, set the prediction parameter threshold, calculate the prediction parameter, if the prediction parameter is greater than the prediction threshold, the functional module is a suspected fault module, and perform consistency judgment on the functional parameters corresponding to the prediction parameter; otherwise, directly output the prediction result.

[0010] S4: Perform consistency judgment on the secondary fault prediction data in the suspected fault module functional module, calculate the consistency coefficient, and set the consistency coefficient threshold. If the consistency coefficient is greater than the consistency coefficient threshold, the suspected fault module has a fault; otherwise, the suspected fault module does not have a fault.

[0011] S5: Output fault prediction results.

[0012] Further, step S1 includes:

[0013] The fault prediction system obtains a set of maintenance records for emergency transport ventilators. These maintenance records come from two sources: one is the maintenance records of similar emergency transport ventilators obtained from the network using big data, and the other is the maintenance records of the emergency transport ventilators currently in use.

[0014] In the operation and maintenance records of the emergency transport ventilator, the operation and maintenance records are normalized and the functional modules of the emergency transport ventilator are re-divided into gas source and power module, respiratory control and monitoring module, safety and power module, and comprehensive module.

[0015] In the gas source and power module, the functional parameters gas source pressure P and gas humidity W are extracted as primary fault prediction data.

[0016] In the respiratory control and monitoring module, the functional parameters respiratory rate H and oxygen concentration S are extracted as primary fault prediction data.

[0017] In the safety and power module, the functional parameters power supply voltage U and power supply current I are extracted as first-level fault prediction data.

[0018] In the integrated module, the functional parameters service life Q and module maintenance frequency R are extracted as primary fault prediction data.

[0019] The primary fault prediction data of the gas source and power module, respiratory control and monitoring module, safety and power module, and integrated module of the emergency transport ventilator were optimized to obtain secondary fault prediction data.

[0020] After normalization processing, the operation and maintenance records of similar emergency transport ventilators obtained from the network using big data are used to generate a fault prediction dataset. The fault prediction dataset mainly uses secondary fault prediction data from the gas source and power module, respiratory control and monitoring module, and safety assurance and power supply module.

[0021] The maintenance records of the currently used emergency transport ventilators are normalized to generate a local fault dataset, which mainly uses the secondary fault prediction data from the comprehensive module.

[0022] Furthermore, the optimization of the primary fault prediction data within the gas source and power module, respiratory control and monitoring module, safety and power module, and integrated module of the emergency transport ventilator yields secondary fault prediction data, including:

[0023] In the air source and power module, the number of air source pressures P acquired is n. P Calculate n P Average pressure of each gas source The specific calculation formula for the secondary fault prediction data is as follows:

[0024]

[0025] Where i is an integer, P i Let ∑ represent the pressure of the i-th gas source, and let ∑ be the summation function;

[0026] The number n of gas humidity W obtained W Calculate n W Average humidity of each gas The specific calculation formula for the secondary fault prediction data is as follows:

[0027]

[0028] Among them, W i Indicates the humidity of the i-th air source;

[0029] In the respiratory control and monitoring module, obtain n H The mode of respiratory rates H As data for secondary fault prediction;

[0030] Get n S The mode of oxygen concentration S As data for secondary fault prediction;

[0031] In the safety and power module, obtain n U The maximum value of the power supply voltage U and minimum value difference The specific calculation formula for the secondary fault prediction data is as follows:

[0032]

[0033] Where U1 represents the first power supply voltage, U iThis represents the voltage of the i-th power supply. Indicates the nth U There are several power supply voltages, where max{} represents the maximum value operation and min{} represents the minimum value operation, and n... U Indicates the quantity of power supply voltage;

[0034] Get n I The maximum value of the power supply current I and minimum value difference The specific calculation formula for the secondary fault prediction data is as follows:

[0035]

[0036] Where I1 represents the first power supply current, I i This represents the current of the i-th power source. Indicates the nth I A power supply current, n I Indicates the amount of power supply current;

[0037] In the integrated module, the first-level fault prediction data is directly used as the second-level fault prediction data.

[0038] Further, step S2 includes:

[0039] A multidimensional diagnostic model was established based on the fault prediction dataset. The damage value and fault value of the emergency transport ventilator were calculated based on the secondary fault prediction data of each functional module of the emergency transport ventilator.

[0040] The damage value F is divided into 5 levels, namely 1, 2, 3, 4, and 5. Different damage values ​​are assigned based on the deviations of the secondary fault prediction data in each functional module and the functional parameters of the current emergency transport ventilator before use.

[0041] In the air source and power module, calculations are performed. The absolute value ΔP of the pressure difference between the ventilator and the current emergency transport ventilator. The damage value corresponding to the gas source pressure is 1; if The damage value corresponding to the gas source pressure is 2; if The damage value corresponding to the gas source pressure is 3; if The damage value corresponding to the gas source pressure is 4; if The damage value corresponding to the gas source pressure is 5;

[0042] calculate The absolute value ΔW of the humidity difference between the gas used in the current emergency transport ventilator and the gas used in the ventilator, if The damage value corresponding to gas humidity is 1; if The damage value corresponding to gas humidity is 2; if The damage value corresponding to gas humidity is 3; if The damage value corresponding to gas humidity is 4; if The damage value corresponding to gas humidity is 5;

[0043] The average of the air source pressure and gas humidity damage values ​​is taken as the damage value F between the air source and the power module. 1 ;

[0044] In the respiratory control and monitoring module, calculations are performed. The absolute value ΔH of the difference between the current respiratory rate and the respiratory rate of the current emergency transport ventilator, if Respiratory rate corresponds to a damage value of 1; if The respiratory rate corresponds to a damage value of 2; if The respiratory rate corresponds to a damage value of 3; if The respiratory rate corresponds to a damage value of 4; if The respiratory rate corresponds to a damage value of 5;

[0045] calculate The absolute value ΔS of the difference between the oxygen concentration of the current emergency transport ventilator and the oxygen concentration of the current ventilator, if The damage value corresponding to the oxygen concentration is 1; if The damage value corresponding to the oxygen concentration is 2; if The damage value corresponding to the oxygen concentration is 3; if The damage value corresponding to the oxygen concentration is 4; if The damage value corresponding to the oxygen concentration is 5;

[0046] The average of the respiratory rate and oxygen concentration impairment values ​​is taken as the impairment value F of the respiratory control and monitoring module. 2 ;

[0047] In the safety and power modules, computing The ratio of the current emergency transport ventilator power supply voltage ΔU to the value of the ventilator voltage ΔU, if The damage value corresponding to the power supply voltage is 1; if The damage value corresponding to the power supply voltage is 2; if The damage value corresponding to the power supply voltage is 3; if The damage value corresponding to the power supply voltage is 4; if The damage value corresponding to the power supply voltage is 5;

[0048] calculate The ratio of the current emergency transport ventilator power supply current ΔI to the current, if The damage value corresponding to the power supply current is 1; if The damage value corresponding to the power supply current is 2; if The damage value corresponding to the power supply current is 3; if The corresponding damage value for the power supply current is 4; if The damage value corresponding to the power supply current is 5;

[0049] The average of the power supply voltage and power supply current damage values ​​is taken as the safety guarantee and the damage value F of the power supply module. 3 .

[0050] Furthermore, the calculation of the damage value and failure value of the emergency transport ventilator includes:

[0051] The fault value X represents the likelihood of a fault occurring, with 5 levels as follows: almost impossible, fault value 1; low probability, fault value 2; medium probability, fault value 3; high probability, fault value 4; almost possible, fault value 5.

[0052] For each functional module, the ratio of the number of times a fault occurs in the secondary fault prediction data of that functional module to the total number of secondary fault prediction data of that functional module is the probability value, and the fault value is calculated from the probability value.

[0053] If the probability value of each functional module is less than or equal to 0.0002, it is considered almost impossible, and the fault value is 1; if the probability value of each functional module is greater than 0.0002 and less than or equal to 0.005, it is considered low probability, and the fault value is 2; if the probability value of each functional module is greater than 0.005 and less than or equal to 0.1, it is considered medium probability, and the fault value is 3; if the probability value of each functional module is greater than 0.1 and less than or equal to 0.5, it is considered high probability, and the fault value is 4; if the probability value of each functional module is greater than 0.5, it is considered almost possible, and the fault value is 5.

[0054] The average of the fault values ​​for gas source pressure and gas humidity is taken as the fault value X for the gas source and power module. 1 The average of the respiratory rate and oxygen concentration fault values ​​is taken as the fault value X of the respiratory control and monitoring module. 2 The average of the power supply voltage and power supply current fault values ​​is taken as the safety guarantee and the fault value X of the power supply module. 3 .

[0055] Further, S3 includes:

[0056] The remaining life factor and maintenance factor of the emergency transport ventilator are calculated from the fault local dataset;

[0057] Set the total lifespan of the current emergency transport ventilator function module to [value]. The remaining life coefficient Where ΔQ iThis represents the remaining lifespan coefficient of the i-th functional module. Q represents the total lifespan of the i-th functional module. i This indicates the service life of the i-th functional module. The gas source and power module is the first functional module, the breathing control and monitoring module is the second functional module, and the safety and power module is the third functional module.

[0058] The maintenance coefficient is set to ΔR. If the number of maintenance times of a functional module is less than or equal to 2, the maintenance coefficient is 1. If the number of maintenance times of a functional module is greater than 2 and less than or equal to 5, the maintenance coefficient is 2. If the number of maintenance times of a functional module is greater than 5, the maintenance coefficient is 4.

[0059] Set the prediction parameter threshold ΔA, and calculate the prediction parameter A using the following formula:

[0060]

[0061] Where, ΔR i A represents the maintenance coefficient of the i-th functional module; i X represents the prediction parameters of the i-th functional module; i This represents the fault value of the i-th functional module;

[0062] Iterate through the prediction parameters of each functional module. If the prediction parameter is greater than the prediction threshold, the module is a suspected fault module. Perform a consistency judgment on the functional parameters corresponding to the prediction parameter. Otherwise, the functional module corresponding to the prediction parameter is normal, and the prediction result is output directly.

[0063] Further, S4 includes:

[0064] The consistency of the secondary fault prediction data of suspected faulty modules is assessed, and the consistency coefficient is calculated. The specific steps are as follows:

[0065] S41: Hardware consistency coefficient K 1 Check the hardware of the suspected faulty module. If there is no obvious damage, the hardware consistency coefficient is 1. If there is obvious damage, the hardware consistency coefficient is 0.

[0066] S42: Functional consistency coefficient K 2 For the air source and power module, select the consistency coefficient of the air source pressure calculation function; for the breathing control and monitoring module, select the consistency coefficient of the breathing rate calculation function; for the safety assurance and power supply, select the consistency coefficient of the power supply voltage calculation function.

[0067] The setting value for the suspected fault module is g, and the collected values ​​of the functional parameters of the suspected fault module are selected within 30 seconds. There are n g A numerical value, functional consistency coefficient K2 The calculation formula is as follows:

[0068]

[0069] Where j represents an integer, Let n represent the functional consistency coefficient of the i-th suspected faulty functional module. g This indicates the number of functional parameter values ​​collected from the suspected faulty module within 30 seconds.

[0070] S43: Calculate the consistency coefficient K;

[0071]

[0072] Among them, K i Let represent the consistency coefficient of the i-th suspected faulty module. Represents the hardware consistency coefficient of the i-th suspected faulty module;

[0073] Set the consistency coefficient threshold ΔK i If the consistency coefficient is greater than the consistency coefficient threshold, then the suspected faulty module is faulty; otherwise, the suspected faulty module is not faulty.

[0074] On the other hand, the present invention also provides an emergency transport ventilator failure prediction system, which includes:

[0075] The data acquisition module and the fault prediction system acquire a set of operation and maintenance records of emergency transport ventilators, normalize the operation and maintenance records, redivide the functional modules of the emergency transport ventilators, calculate the secondary fault prediction data of each functional module, and obtain the fault prediction dataset and the fault local dataset.

[0076] The multidimensional diagnostic model module establishes a multidimensional diagnostic model based on the fault prediction dataset, and calculates the damage value and fault value of the emergency transport ventilator based on the secondary fault prediction data of each functional module.

[0077] The prediction parameter calculation module calculates the remaining life coefficient and maintenance coefficient of the emergency transport ventilator from the local fault dataset, sets the prediction parameter threshold, calculates the prediction parameter, and if the prediction parameter is greater than the prediction threshold, the module is a suspected fault module. The module performs a consistency judgment on the functional parameters corresponding to the prediction parameter; otherwise, it directly outputs the prediction result.

[0078] The consistency judgment and output module performs consistency judgment on the secondary fault prediction data of the suspected fault module, calculates the consistency coefficient, sets the consistency coefficient threshold, and if the consistency coefficient is greater than the consistency coefficient threshold, the suspected fault module has a fault; otherwise, the suspected fault module does not have a fault, and outputs the fault prediction result.

[0079] The beneficial effects of the technical solution provided by this invention include at least the following:

[0080] 1. This invention employs a multi-level computational approach to predict the failure of emergency transport ventilators. By normalizing the emergency transport ventilator and dividing it into multiple functional modules, it uses maintenance records to extract feature parameters from these modules for failure prediction. This not only enhances the richness and dimension of information during failure prediction but also improves the accuracy of fault diagnosis, enabling timely detection of faulty modules and achieving grid-based management of failures. Furthermore, this solution utilizes not only the maintenance records of the current emergency transport ventilator but also those of other similar emergency transport ventilators. Compared to other failure prediction solutions, this solution provides more targeted data, and the use of large datasets is convenient, fast, and computationally efficient.

[0081] 2. When calculating the prediction parameters, this invention first uses the fault prediction dataset to calculate the damage value and fault value. The directly extracted parameters are not characteristic. Therefore, for the different functional parameters, the secondary fault prediction data is calculated from the primary fault prediction data. Then, the prediction parameters are calculated using the local fault dataset. By using the service life and maintenance frequency of the current emergency transport ventilator, the degree of damage to different functional modules can be obtained. Compared with the traditional solution, this solution extracts more representative features when calculating the data, thus further improving the accuracy of fault prediction.

[0082] 3. When performing consistency judgment, this solution not only calculates the hardware consistency coefficient and checks the basic condition of suspected faulty modules to reduce the computational cost of software, but also further calculates the functional consistency coefficient. By monitoring the fluctuation of functional parameters over a period of time, it determines whether the suspected faulty module has actually failed. Through consistency judgment, the results of the predicted parameters are further verified, thereby improving the accuracy of fault prediction. Attached Figure Description

[0083] Figure 1 This is a flowchart illustrating a method for predicting ventilator malfunctions during emergency transport, provided by an embodiment of the present invention. Detailed Implementation

[0084] The present invention will be further described below with reference to the accompanying drawings, but this is not intended to limit the present invention in any way. Any modifications or substitutions made based on the teachings of the present invention shall fall within the protection scope of the present invention.

[0085] Example 1

[0086] This embodiment provides a method for predicting ventilator malfunctions during emergency transport, such as... Figure 1As shown, the method includes the following steps:

[0087] The fault prediction system acquires a set of maintenance records for emergency transport ventilators. These records come from two sources: one is the maintenance records of similar emergency transport ventilators obtained from the network using big data, and the other is the maintenance records of the currently used emergency transport ventilators. The maintenance records are normalized, the functional modules of the emergency transport ventilators are re-divided, and the secondary fault prediction data of each functional module are calculated to obtain the fault prediction dataset and the local fault dataset.

[0088] It should be noted that this invention utilizes multidimensional data to improve information richness, thereby increasing the accuracy of fault prediction. In the operation and maintenance records of emergency transport ventilators, the operation and maintenance records are normalized. The purpose of normalization is not only to effectively utilize the effective information in the operation and maintenance records, but also to accurately locate the faulty module during fault prediction, thereby realizing grid-based fault management.

[0089] Based on the operation and maintenance records, the functional modules of the emergency transport ventilator are reclassified into gas source and power module, respiratory control and monitoring module, safety and power module, and comprehensive module. It should be further noted that in this embodiment, the calculation is mainly performed on the first three functional modules, and the fourth module mainly refers to the current operation and maintenance records of the emergency transport ventilator.

[0090] In the gas source and power module, the functional parameters gas source pressure P and gas humidity W are extracted as primary fault prediction data.

[0091] In the respiratory control and monitoring module, the functional parameters respiratory rate H and oxygen concentration S are extracted as primary fault prediction data.

[0092] In the safety and power module, the functional parameters power supply voltage U and power supply current I are extracted as first-level fault prediction data.

[0093] In the integrated module, the functional parameters service life Q and module maintenance frequency R are extracted as primary fault prediction data.

[0094] The primary fault prediction data of the gas source and power module, respiratory control and monitoring module, safety and power module, and integrated module of the emergency transport ventilator were optimized to obtain secondary fault prediction data.

[0095] The optimization process mainly involves the following steps:

[0096] In the air source and power module, the number of air source pressures P acquired is n. P Calculate n P Average pressure of each gas source The specific calculation formula for the secondary fault prediction data is as follows:

[0097]

[0098] Where i is an integer, P i Let ∑ represent the pressure of the i-th gas source, and let ∑ be the summation function;

[0099] The number n of gas humidity W obtained W Calculate n W Average humidity of each gas The specific calculation formula for the secondary fault prediction data is as follows:

[0100]

[0101] Among them, W i Indicates the humidity of the i-th air source;

[0102] In the respiratory control and monitoring module, obtain n H The mode of respiratory rates H As data for secondary fault prediction;

[0103] Get n S The mode of oxygen concentration S As data for secondary fault prediction;

[0104] In the safety and power module, obtain n U The maximum value of the power supply voltage U and minimum value difference The specific calculation formula for the secondary fault prediction data is as follows:

[0105]

[0106] Where U1 represents the first power supply voltage, U i This represents the voltage of the i-th power supply. Indicates the nth U There are two power supply voltages; max{} represents the maximum value operation and min{} represents the minimum value operation.

[0107] Get n I The maximum value of the power supply current I and minimum value difference The specific calculation formula for the secondary fault prediction data is as follows:

[0108]

[0109] Where I1 represents the first power supply current, I i This represents the current of the i-th power source. Indicates the nthI One power supply current;

[0110] In the integrated module, the first-level fault prediction data is directly used as the second-level fault prediction data;

[0111] After normalization processing, the operation and maintenance records of similar emergency transport ventilators obtained from the network using big data are used to generate a fault prediction dataset. The fault prediction dataset mainly uses secondary fault prediction data from the gas source and power module, respiratory control and monitoring module, and safety assurance and power supply module.

[0112] The maintenance records of the currently used emergency transport ventilators are normalized to generate a local fault dataset, which mainly uses the secondary fault prediction data from the comprehensive module.

[0113] After obtaining the fault prediction dataset and the local fault dataset, the dataset needs to be processed. A multidimensional diagnostic model is established from the fault prediction dataset, and the damage value and fault value of the emergency transport ventilator are calculated from the secondary fault prediction data of each functional module. It should be noted that the so-called multidimensional diagnostic model refers to the multidimensional information used in calculating the damage value and fault value, and also to the multi-level calculation.

[0114] The damage value F is divided into 5 levels, namely 1, 2, 3, 4, and 5. Different damage values ​​are assigned based on the deviations of the secondary fault prediction data in each functional module and the functional parameters of the current emergency transport ventilator before use.

[0115] In the air source and power module, calculations are performed. The absolute value ΔP of the pressure difference between the ventilator and the current emergency transport ventilator. The damage value corresponding to the gas source pressure is 1. The damage value corresponding to the gas source pressure is 2. The damage value corresponding to the gas source pressure is 3. The damage value corresponding to the gas source pressure is 4. The damage value corresponding to the gas source pressure is 5;

[0116] calculate The absolute value ΔW of the humidity difference between the gas used in the current emergency transport ventilator and the gas used in the ventilator, if The damage value corresponding to gas humidity is 1. The damage value corresponding to gas humidity is 2. The damage value corresponding to gas humidity is 3. The damage value corresponding to gas humidity is 4. The damage value corresponding to gas humidity is 5;

[0117] The average of the air source pressure and gas humidity damage values ​​is taken as the damage value F between the air source and the power module. 1 ;

[0118] If the damage value of the gas source pressure is 3 and the damage value of the gas humidity is 2, then the damage value of the gas source and power module is 2.5. All calculations are rounded to one decimal place.

[0119] In the respiratory control and monitoring module, calculations are performed. The absolute value ΔH of the difference between the current respiratory rate and the respiratory rate of the current emergency transport ventilator, if Respiratory rate corresponds to a damage value of 1, if The respiratory rate corresponds to a damage value of 2, if The respiratory rate corresponds to a damage value of 3. The respiratory rate corresponds to a damage value of 4. The respiratory rate corresponds to a damage value of 5;

[0120] calculate The absolute value ΔS of the difference between the oxygen concentration of the current emergency transport ventilator and the oxygen concentration of the current ventilator, if The oxygen concentration corresponds to a damage value of 1, if The damage value corresponding to the oxygen concentration is 2. The oxygen concentration corresponds to a damage value of 3. The damage value corresponding to the oxygen concentration is 4. The damage value corresponding to the oxygen concentration is 5;

[0121] The average of the respiratory rate and oxygen concentration impairment values ​​is taken as the impairment value F of the respiratory control and monitoring module. 2 If the impairment value for respiratory rate is 4 and the impairment value for oxygen concentration is 3, then the impairment value for the respiratory control and monitoring module is 3.5.

[0122] In the safety and power modules, computing The ratio of the current emergency transport ventilator power supply voltage ΔU to the value of the ventilator voltage ΔU, if The damage value corresponding to the power supply voltage is 1. The damage value corresponding to the power supply voltage is 2. The damage value corresponding to the power supply voltage is 3. The damage value corresponding to the power supply voltage is 4. The damage value corresponding to the power supply voltage is 5;

[0123] calculate The ratio of the current emergency transport ventilator power supply current ΔI to the current, if The damage value corresponding to the power supply current is 1. The damage value corresponding to the power supply current is 2. If The damage value corresponding to the power supply current is 3. If The power supply current corresponds to a damage value of 4. If The damage value corresponding to the power supply current is 5;

[0124] The average of the power supply voltage and power supply current damage values ​​is taken as the safety guarantee and the damage value U of the power supply module. 3 ;

[0125] If the damage value of the power supply voltage is 1 and the damage value of the power supply current is 2, then the damage value of the safety protection and power module is 1.5.

[0126] Furthermore, the fault value X represents the likelihood of a fault occurring, with 5 levels as follows: almost impossible (fault value 1), low probability of occurrence (fault value 2), medium probability of occurrence (fault value 3), high probability of occurrence (fault value 4), almost possible (fault value 5).

[0127] For each functional module, the ratio of the number of times a fault occurs in the secondary fault prediction data of that functional module to the total number of secondary fault prediction data of that functional module is the probability value, and the fault value is calculated from the probability value.

[0128] If the probability value of each functional module is less than or equal to 0.0002, it is considered almost impossible, with a fault value of 1; if the probability value is greater than 0.0002 and less than or equal to 0.005, it is considered low probability, with a fault value of 2; if the probability value is greater than 0.005 and less than or equal to 0.1, it is considered medium probability, with a fault value of 3; if the probability value is greater than 0.1 and less than or equal to 0.5, it is considered high probability, with a fault value of 4; and if the probability value is greater than 0.5, it is considered almost possible, with a fault value of 5. To further illustrate, consider the safety and power supply module. If there are 10 data points for power voltage and 2 failures, and 30 data points for power current and 5 failures, then the total number of secondary fault prediction data points for this functional module is 40. The number of failures in the secondary fault prediction data points for this functional module is 7, and the probability value is 7 / 40 = 0.175. This is just an example; for a large amount of data, the probability value will vary.

[0129] The average of the fault values ​​for gas source pressure and gas humidity is taken as the fault value X for the gas source and power module. 1 The average of the respiratory rate and oxygen concentration fault values ​​is taken as the fault value X of the respiratory control and monitoring module. 2 The average of the power supply voltage and power supply current fault values ​​is taken as the safety guarantee and the power module fault value X. 3 ;

[0130] In this embodiment, the fault value for gas source pressure is 2, and the fault value for gas humidity is 1. Therefore, the fault value X for the gas source and power module is... 1 The fault value is 1.5, the respiratory rate fault value is 2, the oxygen concentration fault value is 3, and the respiratory control and monitoring module fault value is X. 2 The fault value is 2.5, the power supply voltage fault value is 3, the power supply current fault value is 2, and the safety protection and power module fault value is X. 3 It is 2.5;

[0131] After calculating the damage value and failure value, the remaining life coefficient and maintenance coefficient of the emergency transport ventilator are then calculated from the local failure dataset.

[0132] Set the total lifespan of the emergency transport ventilator function module to: The remaining life coefficient Where ΔQ i This represents the remaining lifespan coefficient of the i-th functional module. Q represents the total lifespan of the i-th functional module. i This indicates the lifespan of the i-th functional module. The gas source and power module is the first functional module, the breathing control and monitoring module is the second functional module, and the safety and power module is the third functional module.

[0133] In this embodiment, the remaining lifespan coefficient ΔQ of the air source and power module is assumed. 1 The remaining lifespan coefficient ΔQ of the respiratory control and monitoring module is 4. 2 The safety guarantee and the remaining life coefficient ΔQ of the power module are both 1. 3 The value is 2.

[0134] The maintenance coefficient is set to ΔR. If the number of maintenance times of a functional module is less than or equal to 2, the maintenance coefficient is 1. If the number of maintenance times of a functional module is greater than 2 and less than or equal to 5, the maintenance coefficient is 2. If the number of maintenance times of a functional module is greater than 5, the maintenance coefficient is 4.

[0135] In this embodiment, the maintenance coefficient ΔR for the air source and power module is assumed to be... 1 The maintenance factor ΔR of the respiratory control and monitoring module is 2. 2 The maintenance coefficient ΔR for safety assurance and power module is 2. 3 The value is 1.

[0136] Set the prediction parameter threshold ΔA. In this embodiment, ΔA is 10. Calculate the prediction parameter A using the following formula:

[0137]

[0138] Where, ΔR i A represents the maintenance coefficient of the i-th functional module;i X represents the prediction parameters of the i-th functional module; i This represents the fault value of the i-th functional module;

[0139] We can obtain A 1 A is 1.875. 2 It is 17.5, A 3 It is 1.875.

[0140] Iterate through the prediction parameters of each functional module. If the prediction parameter is greater than the prediction threshold, the module is a suspected fault module. Perform a consistency judgment on the functional parameters corresponding to the prediction parameter. Otherwise, the functional module corresponding to the prediction parameter is normal, and the prediction result is output directly.

[0141] Consistency assessment is performed on the secondary fault prediction data of the suspected faulty module, and the consistency coefficient is calculated. Based on the above, the respiratory control and monitoring module is a suspected faulty module and requires consistency assessment. The specific steps for assessing the consistency of the secondary fault prediction data in the respiratory control and monitoring module and calculating the consistency coefficient are as follows:

[0142] (1) Hardware consistency coefficient K 1 Inspect the hardware of the suspected faulty module. If there is no obvious damage, the hardware consistency coefficient is 1; if there is obvious damage, the hardware consistency coefficient is 0; if no obvious damage is found, then K... 1 =1;

[0143] (2) Functional consistency coefficient K 2 For the air source and power module, select the consistency coefficient of the air source pressure calculation function; for the breathing control and monitoring module, select the consistency coefficient of the breathing rate calculation function; for the safety assurance and power supply, select the consistency coefficient of the power supply voltage calculation function.

[0144] The setting value for the suspected fault module is g, and the collected values ​​of the functional parameters of the suspected fault module are selected within 30 seconds. There are n g A numerical value, functional consistency coefficient K 2 The calculation formula is as follows:

[0145]

[0146] Where j represents an integer, This represents the functional consistency coefficient of the i-th suspected faulty functional module, assuming that K is calculated. 2 It is 0.4;

[0147] (3) Consistency coefficient: Calculate the consistency coefficient K, where Therefore, K is 0.4. i Let represent the consistency coefficient of the i-th suspected faulty module. Represents the hardware consistency coefficient of the i-th suspected faulty module;

[0148] Set a consistency coefficient threshold ΔK, where ΔK i This represents the i-th consistency coefficient threshold, which corresponds to a functional module. If the consistency coefficient is greater than the consistency coefficient threshold, the suspected faulty module is faulty; otherwise, the suspected faulty module is not faulty. In this case, ΔK2 is 0.05, and the consistency coefficient is greater than the consistency coefficient threshold. Therefore, the suspected faulty module is faulty, and the respiratory control and monitoring module can be predicted to be faulty.

[0149] Output the fault prediction results.

[0150] Example 2

[0151] This embodiment provides an emergency transport ventilator malfunction prediction system, which includes the following modules:

[0152] The data acquisition module and the fault prediction system acquire a set of operation and maintenance records of emergency transport ventilators, normalize the operation and maintenance records, redivide the functional modules of the emergency transport ventilators, calculate the secondary fault prediction data of each functional module, and obtain the fault prediction dataset and the fault local dataset.

[0153] The multidimensional diagnostic model module establishes a multidimensional diagnostic model based on the fault prediction dataset, and calculates the damage value and fault value of the emergency transport ventilator based on the secondary fault prediction data of each functional module.

[0154] The prediction parameter calculation module calculates the remaining life coefficient and maintenance coefficient of the emergency transport ventilator from the local fault dataset, sets the prediction parameter threshold, calculates the prediction parameter, and if the prediction parameter is greater than the prediction threshold, the module is a suspected fault module. The module performs a consistency judgment on the functional parameters corresponding to the prediction parameter; otherwise, it directly outputs the prediction result.

[0155] The consistency judgment and output module performs consistency judgment on the secondary fault prediction data of the suspected fault module, calculates the consistency coefficient, sets the consistency coefficient threshold, and if the consistency coefficient is greater than the consistency coefficient threshold, the suspected fault module has a fault; otherwise, the suspected fault module does not have a fault, and outputs the fault prediction result.

[0156] The functional units in this invention embodiment can be integrated into a processing module, or each unit can exist physically separately, or multiple units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. The aforementioned devices or systems can execute the storage methods in the corresponding method embodiments.

[0157] In summary, the above embodiments are one implementation of the present invention, but the implementation of the present invention is not limited to the embodiments described above. Any changes, modifications, substitutions, combinations, or simplifications made that deviate from the spirit and principle of the present invention should be considered equivalent substitutions and are included within the protection scope of the present invention.

Claims

1. A method for predicting malfunctions of emergency transport ventilators, characterized in that, Includes the following steps: S1: The fault prediction system acquires a set of operation and maintenance records of emergency transport ventilators, normalizes the operation and maintenance records, re-divides the functional modules of the emergency transport ventilators, calculates the secondary fault prediction data of each functional module, and obtains the fault prediction dataset and the fault local dataset. Step S1 includes: The fault prediction system obtains a set of maintenance records for emergency transport ventilators. These maintenance records come from two sources: one is the maintenance records of similar emergency transport ventilators obtained from the network using big data, and the other is the maintenance records of the emergency transport ventilators currently in use. In the operation and maintenance records of the emergency transport ventilator, the operation and maintenance records are normalized and the functional modules of the emergency transport ventilator are re-divided into gas source and power module, respiratory control and monitoring module, safety and power module, and comprehensive module. In the gas source and power module, the functional parameters gas source pressure P and gas humidity W are extracted as primary fault prediction data. In the respiratory control and monitoring module, the functional parameters respiratory rate H and oxygen concentration S are extracted as primary fault prediction data. In the safety and power module, the functional parameters power supply voltage U and power supply current I are extracted as first-level fault prediction data. In the integrated module, the functional parameters service life Q and module maintenance frequency R are extracted as primary fault prediction data. The primary fault prediction data of the gas source and power module, respiratory control and monitoring module, safety and power module, and integrated module of the emergency transport ventilator were optimized to obtain secondary fault prediction data. After normalization processing, the operation and maintenance records of similar emergency transport ventilators obtained from the network using big data are used to generate a fault prediction dataset. The fault prediction dataset uses secondary fault prediction data from the gas source and power module, respiratory control and monitoring module, and safety assurance and power module. The maintenance records of the currently used emergency transport ventilators are normalized to generate a local fault dataset, which uses the secondary fault prediction data in the comprehensive module. The optimization of the primary fault prediction data within the gas source and power module, respiratory control and monitoring module, safety and power module, and integrated module of the emergency transport ventilator yields secondary fault prediction data, including: In the air source and power module, the number of air source pressures P acquired is: ,calculate Average pressure of each gas source The specific calculation formula for the secondary fault prediction data is as follows: ； Where i is an integer. This represents the pressure of the i-th gas source. For summation functions; The amount of gas humidity W obtained ,calculate Average humidity of each gas The specific calculation formula for the secondary fault prediction data is as follows: ； in, Indicates the humidity of the i-th air source; In the respiratory control and monitoring module, obtain The mode of respiratory rates H As data for secondary fault prediction; Get The mode of oxygen concentration S As data for secondary fault prediction; In the safety and power modules, obtain The maximum value of the power supply voltage U is max{ } and minimum value min{ } difference The specific calculation formula for the secondary fault prediction data is as follows: ； in, This indicates the first power supply voltage. This represents the voltage of the i-th power supply. Indicates the first Given two power supply voltages, max{} represents the maximum value operation and min{} represents the minimum value operation. Indicates the quantity of power supply voltage; Get The maximum value of the power supply current I is max{ } and minimum value min{ } difference The specific calculation formula for the secondary fault prediction data is as follows: ； in, This indicates the first power supply current. This represents the current of the i-th power source. Indicates the first One power supply current, Indicates the amount of power supply current; In the integrated module, the first-level fault prediction data is directly used as the second-level fault prediction data; S2: A multidimensional diagnostic model is established based on the fault prediction dataset, and the damage value and fault value of the emergency transport ventilator are calculated based on the secondary fault prediction data of each functional module. S3: Calculate the remaining life coefficient and maintenance coefficient of the emergency transport ventilator from the fault local dataset, set the prediction parameter threshold, calculate the prediction parameter, if the prediction parameter is greater than the prediction threshold, the functional module is a suspected fault module, and perform consistency judgment on the functional parameters corresponding to the prediction parameter; otherwise, directly output the prediction result. S4: Perform consistency judgment on the secondary fault prediction data of suspected fault modules, calculate the consistency coefficient, and set the consistency coefficient threshold. If the consistency coefficient is greater than the consistency coefficient threshold, the suspected fault module has a fault; otherwise, the suspected fault module does not have a fault. S5: Output fault prediction results.

2. The emergency transport ventilator malfunction prediction method according to claim 1, characterized in that, Step S2 includes: A multidimensional diagnostic model was established based on the fault prediction dataset. The damage value and fault value of the emergency transport ventilator were calculated based on the secondary fault prediction data of each functional module of the emergency transport ventilator. For damage value It is divided into 5 levels, namely 1, 2, 3, 4 and 5, which are assigned different damage values ​​based on the deviation of the secondary fault prediction data in each functional module and the functional parameters of the current emergency transport ventilator before use; In the air source and power module, calculations are performed. The absolute value of the pressure difference between the ventilator and the current emergency transport ventilator. ,if The damage value corresponding to the gas source pressure is 1; if 0.1 The damage value corresponding to the gas source pressure is 2; if 0.2 The damage value corresponding to the gas source pressure is 3; if 0.3 The damage value corresponding to the gas source pressure is 4; if 0.4 The damage value corresponding to the gas source pressure is 5; calculate The absolute value of the humidity difference between the gas used in the current emergency transport ventilator and the gas used in the ventilator. ,if The damage value corresponding to gas humidity is 1; if 0.05 The damage value corresponding to gas humidity is 2; if 0.1 The damage value corresponding to gas humidity is 3; if 0.15 The damage value corresponding to gas humidity is 4; if 0.2 The damage value corresponding to gas humidity is 5; The average of the air source pressure and gas humidity damage values ​​is taken as the damage value of the air source and power module. ; In the respiratory control and monitoring module, calculations are performed. The absolute value of the difference between the respiratory rate and the current emergency transport ventilator ,if The respiratory rate corresponds to a damage value of 1; if 0.02 The respiratory rate corresponds to a damage value of 2; if 0.08 The respiratory rate corresponds to a damage value of 3; if 0.14 The respiratory rate corresponds to a damage value of 4; if 0.2 The respiratory rate corresponds to a damage value of 5; calculate The absolute value of the difference between the oxygen concentration of the current emergency transport ventilator and the oxygen concentration of the current ventilator. ,if The oxygen concentration corresponds to a damage value of 1; if 0.2 The oxygen concentration corresponds to a damage value of 2; if 0.4 The oxygen concentration corresponds to a damage value of 3; if 0.6 The oxygen concentration corresponds to a damage value of 4; if 0.8 The damage value corresponding to the oxygen concentration is 5; The average of the respiratory rate and oxygen concentration impairment values ​​was taken as the impairment value of the respiratory control and monitoring module. ; In the safety and power modules, computing and the current power voltage of the emergency transport ventilator The ratio, if The damage value corresponding to the power supply voltage is 1; if 0.05 The damage value corresponding to the power supply voltage is 2; if 0.1 The power supply voltage corresponds to a damage value of 3; if 0.15 The power supply voltage corresponds to a damage value of 4; if 0.2 The damage value corresponding to the power supply voltage is 5; calculate and the current power supply current of the emergency transport ventilator The ratio, if The damage value corresponding to the power supply current is 1; if 0.05 The damage value corresponding to the power supply current is 2; if 0.1 The damage value corresponding to the power supply current is 3; if 0.15 The power supply current corresponds to a damage value of 4; if 0.2 The damage value corresponding to the power supply current is 5; The average of the power supply voltage and power supply current damage values ​​is taken as the safety guarantee and the damage value of the power supply module. .

3. The emergency transport ventilator malfunction prediction method according to claim 1, characterized in that, The calculation of damage and malfunction values ​​for emergency transport ventilators includes: The fault value X represents the likelihood of a fault occurring, with 5 levels as follows: almost impossible, fault value 1; low probability, fault value 2; medium probability, fault value 3; high probability, fault value 4; almost possible, fault value 5. For each functional module, the ratio of the number of times a fault occurs in the secondary fault prediction data of that functional module to the total number of secondary fault prediction data of that functional module is the probability value, and the fault value is calculated from the probability value. If the probability value of each functional module is less than or equal to 0.0002, it is considered almost impossible, and the fault value is 1; if the probability value of each functional module is greater than 0.0002 and less than or equal to 0.005, it is considered low probability, and the fault value is 2; if the probability value of each functional module is greater than 0.005 and less than or equal to 0.1, it is considered medium probability, and the fault value is 3; if the probability value of each functional module is greater than 0.1 and less than or equal to 0.5, it is considered high probability, and the fault value is 4; if the probability value of each functional module is greater than 0.5, it is considered almost possible, and the fault value is 5. The average of the fault values ​​for gas source pressure and gas humidity is taken as the fault value for the gas source and power module. The average of the respiratory rate and oxygen concentration fault values ​​is taken as the fault value of the respiratory control and monitoring module. The average of the power supply voltage and power supply current fault values ​​is taken as the safety guarantee and the fault value of the power supply module. .

4. The emergency transport ventilator malfunction prediction method according to claim 1, characterized in that, S3 includes: The remaining life factor and maintenance factor of the emergency transport ventilator are calculated from the fault local dataset; Set the total lifespan of the current emergency transport ventilator function module to [value]. Then the remaining life factor ,in This represents the remaining lifespan coefficient of the i-th functional module. This indicates the total lifespan of the i-th functional module. This indicates the service life of the i-th functional module. The gas source and power module is the first functional module, the breathing control and monitoring module is the second functional module, and the safety and power module is the third functional module. Set the maintenance factor to If the number of repairs to a functional module is less than or equal to 2, the repair coefficient is 1; if the number of repairs to a functional module is greater than 2 and less than or equal to 5, the repair coefficient is 2; if the number of repairs to a functional module is greater than 5, the repair coefficient is 4. Set prediction parameter threshold The prediction parameter A is calculated using the following formula: ； in, This represents the maintenance coefficient of the i-th functional module; This represents the prediction parameters for the i-th functional module; This represents the fault value of the i-th functional module; Iterate through the prediction parameters of each functional module. If the prediction parameter is greater than the prediction threshold, the module is a suspected fault module. Perform a consistency judgment on the functional parameters corresponding to the prediction parameter. Otherwise, the functional module corresponding to the prediction parameter is normal, and the prediction result is output directly.

5. The emergency transport ventilator malfunction prediction method according to claim 1, characterized in that, S4 includes: The consistency of the secondary fault prediction data of suspected faulty modules is assessed, and the consistency coefficient is calculated. The specific steps are as follows: S41: Hardware Consistency Coefficient Check the hardware of the suspected faulty module. If there is no obvious damage, the hardware consistency coefficient is 1. If there is obvious damage, the hardware consistency coefficient is 0. S42: Functional Consistency Coefficient For the air source and power module, select the consistency coefficient of the air source pressure calculation function; for the breathing control and monitoring module, select the consistency coefficient of the breathing rate calculation function; for the safety assurance and power supply, select the consistency coefficient of the power supply voltage calculation function. The setting value for the suspected fault module is g, and the collected values ​​of the functional parameters of the suspected fault module are selected within 30 seconds. There are a total of A numerical value, functional consistency coefficient The calculation formula is as follows: ； Where j represents an integer, This represents the functional consistency coefficient of the i-th suspected faulty functional module. This indicates the number of functional parameter values ​​collected from the suspected faulty module within 30 seconds; S43: Calculate the consistency coefficient K; ； in, Let represent the consistency coefficient of the i-th suspected faulty module. Represents the hardware consistency coefficient of the i-th suspected faulty module; Set consistency coefficient threshold ,in This represents the i-th consistency coefficient threshold. The consistency coefficient threshold corresponds to the functional module. If the consistency coefficient is greater than the consistency coefficient threshold, the suspected faulty module has a fault; otherwise, the suspected faulty module does not have a fault.

6. A fault prediction system for emergency transport ventilators, characterized in that, include: The data acquisition module and the fault prediction system acquire a set of operation and maintenance records of emergency transport ventilators, normalize the operation and maintenance records, redivide the functional modules of the emergency transport ventilators, calculate the secondary fault prediction data of each functional module, and obtain the fault prediction dataset and the fault local dataset. The multidimensional diagnostic model module establishes a multidimensional diagnostic model based on the fault prediction dataset, and calculates the damage value and fault value of the emergency transport ventilator based on the secondary fault prediction data of each functional module. The prediction parameter calculation module calculates the remaining life coefficient and maintenance coefficient of the emergency transport ventilator from the local fault dataset, sets the prediction parameter threshold, calculates the prediction parameter, and if the prediction parameter is greater than the prediction threshold, the module is a suspected fault module. The module performs a consistency judgment on the functional parameters corresponding to the prediction parameter; otherwise, it directly outputs the prediction result. The consistency judgment and output module performs consistency judgment on the secondary fault prediction data of the suspected fault module, calculates the consistency coefficient, sets the consistency coefficient threshold, and if the consistency coefficient is greater than the consistency coefficient threshold, the suspected fault module has a fault; otherwise, the suspected fault module does not have a fault, and outputs the fault prediction result. The method for predicting ventilator malfunctions during emergency transport, as described in any one of claims 1-5, is implemented.

Images