Health management method based on aviation power source

Abstract

The invention discloses a health management method based on an aviation power supply, relates to the technical field of aviation power supply intelligent management, and solves the problems that the existing aviation power supply management mostly depends on a traditional management mode, equipment operation parameters are updated in a manual mode, the parameter protection range is fixed, the power supply performance cannot be effectively released, and the maintenance cost is low. And the aviation power supply health management efficiency is low. Generating a power supply health state through each detection item data in the operation data and the standard detection item data; judging whether the power supply health state is an abnormal state; if yes, generating a fault type and an alarm signal according to the operation data; generating a parameter adjustment value according to the fault type; if not, whether the parameter adjustment value is updated is judged, if yes, the corresponding project parameter value is adjusted according to the parameter adjustment value and the corresponding parameter protection value, the equipment operation parameters are intelligently calibrated, the parameter protection range is adjusted, and the efficiency and the intelligent level of aviation power source health management are improved.

Description

Technical Field

[0001] The present application belongs to the technical field of intelligent management of aviation power supplies, and specifically is a health management method based on aviation power supplies. Background Art

[0002] The aviation power supply system is an important guarantee for flight safety. It is composed of key components such as the main power supply, emergency power supply, auxiliary power supply and secondary power supply. This system provides stable and reliable power support for various electronic equipment and systems on the aircraft through efficient energy conversion and distribution mechanisms. The main power supply is usually provided by the engine-driven generator to ensure the main power demand during the flight; while the emergency power supply takes over the power supply task in an emergency to ensure the safe operation of key equipment. In addition, the auxiliary power supply and secondary power supply also play an important role respectively, and jointly maintain the stable operation of the aviation power supply system. The high reliability and safety design of this system ensure that the aircraft can maintain the continuity and stability of power supply in various complex environments, and escort the safety of passengers and crew members.

[0003] Existing aviation power management mostly relies on traditional management modes, and updates equipment operating parameters manually. However, a large number of parameters need to be considered in the health management of aviation power, which is difficult to handle. At the same time, the parameter protection range is fixed, so when adjusting the parameters, the parameters are limited to a fixed range, resulting in power performance not achieving the expected effect. This makes it impossible to effectively release the power performance, resulting in low efficiency in aviation power health management. Therefore, there is still room for further improvement in the health management method of aviation power. Summary of the invention

[0004] The present application aims to solve at least one of the technical problems existing in the prior art; to this end, the present application proposes a health management method based on aviation power supply, which is used to solve the technical problem that the existing aviation power supply management mostly relies on the traditional management mode, updates the equipment operation parameters manually, and the parameter protection range is fixed, so that the power supply performance cannot be effectively released, resulting in low efficiency of aviation power supply health management. The present application generates a power supply health status by acquiring the power supply data of the aviation power supply and its corresponding standard operation data; determines whether the power supply health status is an abnormal state; if yes, generates a fault type and an alarm signal according to the operation data; constructs a multivariate linear regression model under the fault type according to several parameter adjustment values ​​under the fault type and the corresponding calibration detection item data; generates a parameter adjustment value according to the standard detection item data and the multivariate linear regression model; if no, determines whether the parameter adjustment value is updated: if yes, adjusts the corresponding item parameter value according to the parameter adjustment value and its corresponding parameter protection value, thereby solving the above problems.

[0005] To achieve the above objectives, the first aspect of the present application provides a health management method based on aviation power supply, comprising:

[0006] S1: Acquire power data of an aviation power supply and its corresponding standard operation data; the power data includes power supply ID, power supply performance parameters, equipment operation parameters and operation data;

[0007] S2: pre-processing the operation data in the power supply data;

[0008] S3: Acquire pre-processed operation data; generate a power health status according to each detection item data in the operation data and the corresponding standard detection item data in the standard operation data; determine whether the power health status is an abnormal state; if yes, proceed to S4; if no, proceed to S5;

[0009] S4: Generate a fault type and an alarm signal according to the operating data; generate a parameter adjustment value according to the fault type;

[0010] S5: Determine whether the parameter adjustment value is updated; if yes, adjust the corresponding project parameter value according to the parameter adjustment value and its corresponding parameter protection value; if no, enter S6;

[0011] S6: Integrate the power performance, device operating parameters, power health status, operating data and parameter adjustment values ​​corresponding to the same power ID into a power log of the power ID.

[0012] The present application obtains power supply data of an aviation power supply and its corresponding standard operating data; pre-processes the operating data in the power supply data; obtains the pre-processed operating data; generates a power supply health status according to each detection item data in the operating data and the corresponding standard detection item data in the standard operating data; determines whether fault identification is required according to the power supply health status; if so, identifies the fault; if not, directly enters the subsequent process; when a parameter adjustment value update is detected; adjusts the corresponding project parameter value according to the parameter adjustment value and its corresponding parameter protection value, intelligently calibrates the equipment operating parameters and adjusts the parameter protection range, thereby improving the efficiency and intelligence level of aviation power supply health management.

[0013] Furthermore, the preprocessing includes data cleaning and recursive average filtering.

[0014] Further, generating the power health status according to each detection item data in the operation data and the corresponding standard detection item data in the standard operation data includes:

[0015] Obtaining data of several test items in the operating data and corresponding standard test item data in the standard operating data;

[0016] Calculate the absolute value of the difference between each test item data and its corresponding standard test item data in sequence;

[0017] It is determined whether the absolute value of the difference is greater than the corresponding health threshold: if yes, the power health state is set to an abnormal state; if no, the power health state is set to a normal state.

[0018] Further, generating the fault type and alarm signal according to the operation data includes:

[0019] Acquire operation data, and input the operation data into a fault diagnosis model to obtain a number of fault categories and their corresponding confidence levels; the fault diagnosis model is obtained by training a convolutional neural network VGGNet;

[0020] When the maximum value among several confidence levels is greater than the alarm threshold, a fault alarm signal is generated and the fault type is recorded;

[0021] Otherwise, when the maximum value among the several confidence levels is greater than the warning threshold, the fault types corresponding to the first two confidence levels among the several confidence levels are selected, the two fault types are integrated into a mixed fault type, a mixed fault alarm signal is generated, and the mixed fault alarm type is recorded;

[0022] Otherwise, the fault type corresponding to the maximum value among several confidence levels is obtained, a potential risk fault warning signal is generated, and the fault warning type is recorded.

[0023] Furthermore, the fault diagnosis model is obtained by training a convolutional neural network, including:

[0024] Obtain some historical operation data and their corresponding fault type labels;

[0025] Divide a number of historical operation data and their corresponding fault type labels into training data, verification data, and test data;

[0026] Preprocess the training data, verification data and test data to obtain the training set, verification set and test set;

[0027] Select the convolutional neural network model VGGNet as the basic model;

[0028] The basic model is trained using the training set, the parameters are adjusted on the validation set, and the model is tested on the test data. Finally, a fault diagnosis model is obtained whose input is the operating data and whose output is several confidence levels and corresponding fault types.

[0029] Furthermore, a parameter adjustment value is generated according to the fault type, including:

[0030] Obtain several parameter adjustment values ​​of the equipment operating parameters under each fault type and their corresponding calibration test item data;

[0031] The objective function of each calibration test item data and several parameter adjustment values ​​is constructed by the least square method;

[0032] By bringing each calibration test item data and several parameter adjustment values ​​into the objective function to solve, a multivariate linear regression model under each fault type is obtained;

[0033] Select the corresponding multivariate linear regression model according to the fault type;

[0034] Substitute the standard test item data into the corresponding multivariate linear regression model to obtain several parameter adjustment values.

[0035] Furthermore, generating corresponding project parameter values ​​according to the parameter adjustment values ​​and their corresponding parameter protection values ​​includes:

[0036] Obtain project parameter values ​​and their corresponding parameter protection values ​​in equipment operation parameters; obtain several parameter adjustment values;

[0037] Calculate the absolute value of the difference between the corresponding parameter adjustment value and the project parameter value;

[0038] Determine whether the absolute value of the difference is less than the corresponding parameter protection value: if yes, set the project parameter value to the parameter adjustment value;

[0039] Otherwise, obtain the maximum adjustment step of each parameter protection value; calculate the product of the parameter protection value and the maximum adjustment step as the maximum parameter protection value; determine whether the absolute value of the difference is less than the corresponding maximum parameter protection value: if yes, calculate the quotient between the absolute value of the difference and the parameter protection value as the adjustment step; adjust the project parameter value in stages according to the adjustment step until the parameter adjustment value is reached; otherwise, adjust the project parameter value in stages according to the maximum adjustment step until the maximum parameter protection value is reached.

[0040] Furthermore, it also includes S7: intelligent software update through IAP online upgrade, including:

[0041] The bootloader obtains new program code through the external communication interface of the power supply; CRC check is performed to ensure the integrity and correctness of the data;

[0042] The bootloader writes the received program code into the non-volatile memory inside the device, while ensuring that the bootloader itself or other important data is not destroyed;

[0043] After writing is complete, the bootloader reboots the device; initializes according to the new program code and jumps to the new application entry point for execution;

[0044] Determine whether the update is successful by checking the update log: if yes, no operation is performed; otherwise, the program rolls back to the old version.

[0045] Compared with the prior art, the beneficial effects of this application are:

[0046] 1. The present application obtains power supply data of an aviation power supply and its corresponding standard operating data; preprocesses the operating data in the power supply data; obtains the preprocessed operating data; generates a power supply health status according to the data of each detection item in the operating data and the corresponding standard detection item data in the standard operating data; determines whether fault identification is required according to the power supply health status; if yes, identifies the fault; if not, directly enters the subsequent process; when a parameter adjustment value update is detected; adjusts the corresponding project parameter value according to the parameter adjustment value and its corresponding parameter protection value, intelligently calibrates the equipment operating parameters and adjusts the parameter protection range, thereby improving the efficiency and intelligence level of aviation power supply health management. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0048] Figure 1 This is a schematic diagram of a method flow chart of a health management method based on aviation power supply in this application;

[0049] Figure 2 This is a parameter adjustment flow chart of a health management method based on aviation power supply in this application. DETAILED DESCRIPTION

[0050] The technical solution of the present application will be described clearly and completely below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0051] See also Figure 1-Figure 2 The first aspect of the present application provides a health management method based on an aviation power supply, including:

[0052] S1: Obtain the power data of the aviation power supply and its corresponding standard operation data; the power data includes the power ID, power performance parameters, equipment operation parameters and operation data; the operation data includes voltage, current and temperature, etc.; the equipment operation parameters include input voltage, output voltage, output current, power parameters and ripple voltage, etc.; the data type of standard operation data is the same as that of operation data and also includes voltage, current and temperature, etc.;

[0053] S2: pre-processing the operation data in the power supply data;

[0054] S3: Obtain pre-processed operation data; generate power health status according to each test item data in the operation data and the corresponding standard test item data in the standard operation data; the battery health status includes normal status and abnormal status; the normal status indicates that the aviation power supply is operating normally; the abnormal status indicates that the aviation power supply may be faulty; determine whether the power health status is an abnormal state; if yes, enter S4; if no, enter S5;

[0055] S4: Generate a fault type and an alarm signal according to the operating data; generate a parameter adjustment value according to the fault type; the parameter adjustment value is an adjustment value of the corresponding operating parameter generated according to the fault type, and the project parameter value of the corresponding operating parameter is adjusted to the corresponding parameter adjustment value so that the aviation power supply returns to a normal working state; the alarm signal includes a fault alarm signal, a mixed fault alarm signal and a potential risk fault warning signal; the fault alarm signal is an alarm signal issued by an abnormal fault of the aviation power supply that can determine a single fault type; the mixed fault alarm signal is an alarm signal issued by an abnormal fault of the aviation power supply caused by two fault types; the potential risk fault warning signal is a warning signal issued by an abnormal fault of the aviation power supply that may be caused by one fault type;

[0056] S5: Determine whether the parameter adjustment value is updated; the parameter adjustment value update includes updating the parameter adjustment value generated according to the fault or the parameter adjustment value input manually; if yes, adjust the corresponding project parameter value according to the parameter adjustment value and its corresponding parameter protection value; if no, enter S6; the parameter protection value is the standard range within which the corresponding parameter can be adjusted; the project parameter value is the value corresponding to each parameter in the current equipment operation parameter;

[0057] S6: Integrate the power performance, device operating parameters, power health status, operating data, and parameter adjustment values ​​corresponding to the same power ID into a power log of the power ID.

[0058] This embodiment obtains power data of the aviation power supply and its corresponding standard operation data; pre-processes the operation data in the power supply data; obtains the pre-processed operation data; generates a power health status according to each detection item data in the operation data and the corresponding standard detection item data in the standard operation data; determines whether fault identification is required according to the power health status; if yes, identifies the fault; if no, directly enters the subsequent process; when it is detected that the parameter adjustment value is updated; adjusts the corresponding project parameter value according to the parameter adjustment value and its corresponding parameter protection value, intelligently calibrates the equipment operation parameters and adjusts the parameter protection range, thereby improving the efficiency and intelligence level of aviation power health management.

[0059] Preprocessing includes data cleaning and recursive average filtering, including:

[0060] Data cleaning: Delete empty values, abnormal values ​​or duplicate values ​​in the running data to obtain cleaned data;

[0061] Recursive average filtering method: obtain several groups of continuous cleaning data, set the queue length to N, put N continuous cleaning data into the queue in order, and the current cleaning data is the end of the queue; calculate the arithmetic mean of the data in the current queue, and use the arithmetic mean as the preprocessing data.

[0062] In this embodiment, the queue adopts a first-in-first-out method. Each time a new data is sampled, it is put into the end of the queue and the measurement value of the original head of the queue is discarded. The N data in the queue are arithmetic averaged and the arithmetic mean is calculated by the formula: The larger N is, the better the effect of the arithmetic mean. In this embodiment, the current temperature is 24°C, and the queue length is set according to experience. In this embodiment, the queue length is set to 7. The queue in the previous stage is obtained as [25, 23, 24, 22, 26, 25, 24]°C. The current queue is [23, 24, 22, 26, 25, 24, 24]°C according to the first-in-first-out principle. The arithmetic mean is 24°C calculated by the formula, that is, the preprocessed data of the current temperature is 24°C.

[0063] Generating a power health status according to each test item data in the operation data and the corresponding standard test item data in the standard operation data, including: obtaining a number of test item data of the operation data and the corresponding standard test item data in the standard operation data; sequentially calculating the absolute value of the difference between each test item data and its corresponding standard test item data;

[0064] Determine whether the absolute value of the difference is greater than the corresponding health threshold: if yes, set the power health state to an abnormal state; if no, set the power health state to a normal state; the health threshold is set based on experience.

[0065] Generate fault type and alarm signals based on operating data, including:

[0066] Obtaining operation data, and inputting the operation data into the fault diagnosis model to obtain several fault categories and their corresponding confidence levels; the fault diagnosis model is obtained by training the convolutional neural network VGGNet;

[0067] When the maximum value among several confidence levels is greater than the alarm threshold, a fault alarm signal is generated and the fault type is recorded; the alarm threshold is set based on experience, and in this embodiment, the alarm threshold is set to 0.8;

[0068] Otherwise, when the maximum value among the several confidence levels is greater than the warning threshold, the fault types corresponding to the first two confidence levels among the several confidence levels are selected, the two fault types are integrated into a mixed fault type, a mixed fault alarm signal is generated, and the mixed fault alarm type is recorded; the warning threshold is set based on experience, and in this embodiment, the warning threshold is set to 0.5;

[0069] Otherwise, the fault type corresponding to the maximum value among several confidence levels is obtained, a potential risk fault warning signal is generated, and the fault warning type is recorded.

[0070] In this embodiment, the fault types are set to three types {power failure, undervoltage, module heating}, and a set of confidence levels and their corresponding fault types are obtained, namely {0.1 power failure, 0.6 undervoltage, 0.3 module heating}, where the maximum confidence level is 0.6, which is between the warning threshold and the alarm threshold. The fault types with the top two confidence levels are integrated into a mixed fault type, i.e., undervoltage + module heating, to generate a mixed fault alarm signal; other embodiments may have more fault types, and the judgment method is the same as that of this embodiment.

[0071] The fault diagnosis model is trained by convolutional neural network, including:

[0072] Obtain some historical operation data and their corresponding fault type labels;

[0073] Divide a number of historical operation data and their corresponding fault type labels into training data, verification data and test data; the training data, verification data and test data are divided in a ratio of 7:2:1;

[0074] Preprocess the training data, verification data and test data to obtain the training set, verification set and test set;

[0075] Select the convolutional neural network model VGGNet as the basic model;

[0076] The basic model is trained using the training set, the parameters are adjusted on the validation set, and the model is tested on the test data. Finally, a fault diagnosis model is obtained whose input is the operating data and whose output is a number of confidence levels and corresponding fault types. The goal of the training phase is to enable the model to capture the characteristic relationship between the operating data and the fault type, so that the confidence level and the fault type can be accurately output in the subsequent phase. The validation phase is to evaluate and optimize the performance of the model, and adjust the parameters or structure of the model according to the validation results, such as modifying the learning rate, adding regularization terms, adjusting the number of network layers, etc. The fault diagnosis model can efficiently process the input operating data and give corresponding confidence levels and fault types.

[0077] Generate parameter adjustment values ​​based on the fault type, including:

[0078] Obtaining several parameter adjustment values ​​of the equipment operating parameters under various fault types and their corresponding calibration test item data; the calibration test item data is the test item data corresponding to several parameter adjustment values ​​under specific fault types;

[0079] The objective function of each calibration test item data and several parameter adjustment values ​​is constructed by the least square method;

[0080] By bringing each calibration test item data and several parameter adjustment values ​​into the objective function to solve, a multivariate linear regression model under each fault type is obtained;

[0081] Select the corresponding multivariate linear regression model according to the fault type;

[0082] Substitute the standard test item data into the corresponding multivariate linear regression model to obtain several parameter adjustment values; the data type of the standard test item data is the same as that of the test item data.

[0083] Generate corresponding project parameter values ​​according to the parameter adjustment values ​​and their corresponding parameter protection values, including:

[0084] Obtain project parameter values ​​and their corresponding parameter protection values ​​in equipment operation parameters; obtain several parameter adjustment values;

[0085] Calculate the absolute value of the difference between the corresponding parameter adjustment value and the project parameter value;

[0086] Determine whether the absolute value of the difference is less than the corresponding parameter protection value: if yes, set the project parameter value to the parameter adjustment value; the parameter protection value is set based on experience;

[0087] Otherwise, obtain the maximum adjustment step of each parameter protection value; the maximum adjustment step is the maximum adjustment range of the parameter protection value; calculate the product of the parameter protection value and the maximum adjustment step as the maximum parameter protection value; determine whether the absolute value of the difference is less than the corresponding maximum parameter protection value: if yes, calculate the quotient between the absolute value of the difference and the parameter protection value as the adjustment step; the adjustment step is the number of times the parameter adjustment value needs to be adjusted; adjust the project parameter value in stages according to the adjustment step until the parameter adjustment value is reached; otherwise, adjust the project parameter value in stages according to the maximum adjustment step until the maximum parameter protection value is reached; the maximum adjustment step of each parameter protection value is set according to experience.

[0088] In this embodiment, the project parameter value of the input voltage in the equipment operation parameters is 115V, the parameter adjustment value of the input voltage is 130V, the parameter protection value of the input voltage is 10V, and the maximum adjustment step is set according to experience. In this embodiment, the maximum adjustment step is set to 2; the absolute value of the difference between the parameter adjustment value and the project parameter value is 15V, and the absolute value of the difference is greater than the parameter protection value; the maximum parameter protection value is 20V, and the absolute value of the difference is less than the maximum parameter protection value, and the adjustment step is 1.5 at this time; the adjustment is divided into two stages according to the adjustment step. In the first stage, the project parameter value is adjusted to 125V, and in the second stage, the project parameter value is adjusted to 130V. At this time, the project parameter value is equal to the parameter adjustment value.

[0089] Also includes S7: Intelligent software update through IAP online upgrade, including:

[0090] The bootloader obtains new program code through the external communication interface of the power supply; CRC check is performed to ensure the integrity and correctness of the data;

[0091] The bootloader writes the received program code into the non-volatile memory inside the device, while ensuring that the bootloader itself or other important data is not destroyed;

[0092] After writing is complete, the bootloader reboots the device; initializes according to the new program code and jumps to the new application entry point for execution;

[0093] Determine whether the update is successful by checking the update log: if yes, no operation is performed; otherwise, the program rolls back to the old version.

[0094] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0095] How this application works:

[0096] By obtaining the power data of the aviation power supply and its corresponding standard operating data; preprocessing the operating data in the power supply data; obtaining the preprocessed operating data; generating the power health status according to the data of each detection item in the operating data and the corresponding standard detection item data in the standard operating data; judging whether the power health status is an abnormal state; if yes, generating the fault type and alarm signal according to the operating data; generating the parameter adjustment value according to the fault type; if no, judging whether the parameter adjustment value is updated: if yes, adjusting the corresponding project parameter value according to the parameter adjustment value and its corresponding parameter protection value; building a power log with the power ID; realizing intelligent software update through IAP online upgrade, intelligent calibration of equipment operating parameters and adjustment of parameter protection range, improving the efficiency and intelligence level of aviation power health management, avoiding the existing aviation power management relying on the traditional management mode, manually updating the equipment operating parameters, and fixing the parameter protection range, so that the power performance cannot be effectively released, resulting in the technical problem of low efficiency of aviation power health management.

[0097] The above embodiments are only used to illustrate the technical method of the present application and are not intended to limit it. Although the present application has been described in detail with reference to the preferred embodiments, a person of ordinary skill in the art should understand that the technical method of the present application may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present application.

Claims

1. A health management method based on aviation power supply, characterized in that: include: S1: Acquire power data of an aviation power supply and its corresponding standard operation data; the power data includes power supply ID, power supply performance parameters, equipment operation parameters and operation data; S2: pre-processing the operation data in the power supply data; S3: Obtain preprocessed operation data; Generate a power health status according to each test item data in the operation data and corresponding standard test item data in the standard operation data; Determine whether the power health status is abnormal; If yes, then go to S4; If not, then go to S5; S4: Generate a fault type and an alarm signal according to the operating data; generate a parameter adjustment value according to the fault type; S5: Determine whether the parameter adjustment value is updated; if yes, adjust the corresponding project parameter value according to the parameter adjustment value and its corresponding parameter protection value; If not, then go to S6; S6: Integrate the power performance, device operating parameters, power health status, operating data and parameter adjustment values ​​corresponding to the same power ID into a power log of the power ID.

2. The health management method based on aviation power supply according to claim 1, characterized in that: The preprocessing includes data cleaning and recursive average filtering.

3. The health management method based on aviation power supply according to claim 1, characterized in that: Generating the power health status according to each detection item data in the operation data and the corresponding standard detection item data in the standard operation data includes: Obtaining data of several test items in the operating data and corresponding standard test item data in the standard operating data; Calculate the absolute value of the difference between each test item data and its corresponding standard test item data in sequence; It is determined whether the absolute value of the difference is greater than the corresponding health threshold: if yes, the power health state is set to an abnormal state; if no, the power health state is set to a normal state.

4. The health management method based on aviation power supply according to claim 1, characterized in that: The generating of the fault type and the alarm signal according to the operation data comprises: Acquire operation data, and input the operation data into a fault diagnosis model to obtain a number of fault categories and their corresponding confidence levels; the fault diagnosis model is obtained by training a convolutional neural network VGGNet; When the maximum value among several confidence levels is greater than the alarm threshold, a fault alarm signal is generated and the fault type is recorded; Otherwise, when the maximum value among the several confidence levels is greater than the warning threshold, the fault types corresponding to the first two confidence levels among the several confidence levels are selected, the two fault types are integrated into a mixed fault type, a mixed fault alarm signal is generated, and the mixed fault alarm type is recorded; Otherwise, the fault type corresponding to the maximum value among several confidence levels is obtained, a potential risk fault warning signal is generated, and the fault warning type is recorded.

5. The health management method based on aviation power supply according to claim 4 is characterized in that: The fault diagnosis model is obtained by training a convolutional neural network, and includes: Obtain some historical operation data and their corresponding fault type labels; Divide a number of historical operation data and their corresponding fault type labels into training data, verification data, and test data; Preprocess the training data, validation data and test data to obtain the training set, validation set and test set; Select the convolutional neural network model VGGNet as the basic model; The basic model is trained using the training set, the parameters are adjusted on the validation set, and the model is tested on the test data. Finally, a fault diagnosis model is obtained whose input is the preprocessed data and whose output is several confidence levels and corresponding fault types.

6. The health management method based on aviation power supply according to claim 1, characterized in that: Generating the parameter adjustment value according to the fault type includes: Obtain several parameter adjustment values ​​of the equipment operating parameters under each fault type and their corresponding calibration test item data; The objective function of each calibration test item data and several parameter adjustment values ​​is constructed by the least square method; By substituting the data of each calibration test item and several parameter adjustment values ​​into the objective function, a multivariate linear regression model under each fault type is obtained; Select the multivariate linear regression model based on the fault type; Substitute the standard test item data into the corresponding multivariate linear regression model to obtain several parameter adjustment values.

7. The health management method based on aviation power supply according to claim 1, characterized in that: Generating the corresponding project parameter value according to the parameter adjustment value and its corresponding parameter protection value includes: Obtain project parameter values ​​and their corresponding parameter protection values ​​in equipment operation parameters; obtain several parameter adjustment values; Calculate the absolute value of the difference between the corresponding parameter adjustment value and the project parameter value; Determine whether the absolute value of the difference is less than the corresponding parameter protection value: if yes, set the project parameter value to the parameter adjustment value; Otherwise, obtain the maximum adjustment step of each parameter protection value; calculate the product of the parameter protection value and the maximum adjustment step as the maximum parameter protection value; determine whether the absolute value of the difference is less than the corresponding maximum parameter protection value: if yes, calculate the quotient between the absolute value of the difference and the parameter protection value as the adjustment step; adjust the project parameter value in stages according to the adjustment step until the parameter adjustment value is reached; otherwise, adjust the project parameter value in stages according to the maximum adjustment step until the maximum parameter protection value is reached.

8. The health management method based on aviation power supply according to claim 1, characterized in that: The step S6 also includes S7: intelligently updating software through IAP online upgrade, including: The bootloader obtains new program code through the external communication interface of the power supply; CRC check is performed to ensure the integrity and correctness of the data; The bootloader writes the received program code into the non-volatile memory inside the device, while ensuring that the bootloader itself or other important data is not destroyed; After writing is complete, the bootloader reboots the device; initializes according to the new program code and jumps to the new application entry point for execution; Determine whether the update is successful by checking the update log: if yes, no operation is performed; otherwise, the program rolls back to the old version.

Images