Fault diagnosis method and system

Abstract

A fault diagnosis method includes: dividing a target system into multiple macro modules according to a fault diagnosis requirement analysis result of the target system; determining typical faults of the macro module and corresponding test points and test parameters for any macro module, and obtaining test data of the macro module using the test points and test parameters; and performing fault diagnosis on the macro module using a machine learning fault diagnosis method according to the obtained test data of the macro module.

Description

Technical Field

[0001] This article relates to but is not limited to the field of fault diagnosis technology, and in particular to a fault diagnosis method and system. Background Art

[0002] Fault diagnosis methods can generally be divided into two categories: model-based fault diagnosis methods and data-driven fault diagnosis methods. Model-based fault diagnosis methods were the focus of early research on classical fault diagnosis methods, and they were mostly used to diagnose systems that are easy to model; data-driven fault diagnosis methods are generally used to diagnose systems that are difficult to model or where the model accuracy is not high. Summary of the invention

[0003] The following is a summary of the subject matter described in detail herein. This summary is not intended to limit the scope of the claims.

[0004] The embodiments of the present disclosure provide a fault diagnosis method and system.

[0005] On the one hand, an embodiment of the present disclosure provides a fault diagnosis method, comprising: dividing the target system into multiple macro modules according to the fault diagnosis requirements analysis results of the target system; for any macro module, determining the typical faults of the macro module and the corresponding test points and test parameters, and obtaining the test data of the macro module using the test points and the test parameters; for any macro module, performing fault diagnosis using a machine learning fault diagnosis method based on the obtained test data.

[0006] In some exemplary embodiments, for any macro module, determining the test point corresponding to the typical fault of the macro module includes: when the target system is divided into multiple macro modules and there is no coupling relationship between different macro modules, selecting the test point corresponding to the typical fault from multiple signal output terminals of each macro module; when the target system is divided into multiple macro modules and there is a coupling relationship between different macro modules, selecting the test point corresponding to the typical fault from multiple signal output terminals and internal nodes of each macro module.

[0007] In some exemplary embodiments, the test data of the macro module obtained using the test points and the test parameters include: first test data corresponding to the test points and the test parameters of the macro module in a fault-free state, and second test data corresponding to the test points and the test parameters of the macro module in a faulty state.

[0008] In some exemplary embodiments, when the target system is divided into multiple macro modules and there is a coupling relationship between different macro modules, the fault judgment rules for determining the macro module in which the fault occurs in the target system include: for any macro module whose test points include a signal output terminal and an internal node, when the output data of the internal node is normal and the output data of the signal output terminal is abnormal, the fault occurs in the macro module, or the subsequent N-level macro modules connected to the signal output terminal of the macro module, N is an integer greater than 0; when the output data of the internal node is abnormal and the output data of the signal output terminal is abnormal, the fault occurs in the macro module; when the output data of the internal node is abnormal and the output data of the signal output terminal is normal, the fault occurs in the macro module.

[0009] In some exemplary embodiments, the test data of the macro module includes: response signal data of a test point of the macro module, or characteristic data obtained by processing the response signal data.

[0010] In some exemplary embodiments, the fault diagnosis method further includes: using test data of multiple combinations of the macro module to respectively train multiple machine learning fault diagnosis methods, wherein the signal types of the test data in different combinations are different or partially the same; and selecting the optimal machine learning fault diagnosis method from the trained multiple machine learning fault diagnosis methods to perform fault diagnosis on the macro module.

[0011] On the other hand, an embodiment of the present disclosure provides a fault diagnosis system, including: a division module, configured to divide the target system into multiple macro modules according to the fault diagnosis requirements analysis results of the target system; a testing module, configured to determine the typical faults of any macro module and the corresponding test points and test parameters, and obtain the test data of the macro module using the test points and the test parameters; a diagnosis module, configured to perform fault diagnosis on any macro module using a machine learning fault diagnosis method according to the obtained test data.

[0012] In some exemplary embodiments, the test module is configured to determine the test point corresponding to the typical fault of any macro module in the following manner: when the target system is divided into multiple macro modules and there is no coupling relationship between different macro modules, the test point corresponding to the typical fault is selected from multiple signal output terminals of each macro module; when the target system is divided into multiple macro modules and there is a coupling relationship between different macro modules, the test point corresponding to the typical fault is selected from multiple signal output terminals and internal nodes of each macro module.

[0013] On the other hand, an embodiment of the present disclosure provides a fault diagnosis system, including: a memory and a processor, wherein a computer program is stored in the memory, and when the computer program in the memory is executed by the processor, the fault diagnosis method as described above is performed.

[0014] On the other hand, an embodiment of the present disclosure provides a computer-readable storage medium storing computer-executable commands, wherein the computer-executable commands are used to execute the fault diagnosis method as described above.

[0015] The fault diagnosis method provided in the embodiment of the present disclosure can simplify data processing and improve the recognition accuracy of fault diagnosis by dividing the target system into macro modules and using a machine learning fault diagnosis method to perform fault diagnosis on the divided macro modules respectively.

[0016] Other features and advantages of the present disclosure will be described in the following description, and partly become apparent from the description, or be understood by implementing the present disclosure. Other advantages of the present disclosure can be realized and obtained by the schemes described in the description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings are used to provide an understanding of the technical solution of the present disclosure and constitute a part of the specification. Together with the embodiments of the present disclosure, they are used to explain the technical solution of the present disclosure and do not constitute a limitation on the technical solution of the present disclosure.

[0018] Figure 1 A flowchart of a fault diagnosis method provided by at least one embodiment of the present disclosure;

[0019] Figure 2 An example diagram of a macro module of a target system according to at least one embodiment of the present disclosure;

[0020] Figure 3 A schematic diagram of the structure of a fault diagnosis system provided by at least one embodiment of the present disclosure;

[0021] Figure 4 A schematic diagram of the structure of another fault diagnosis system provided by at least one embodiment of the present disclosure. DETAILED DESCRIPTION

[0022] The present disclosure describes multiple embodiments, but the description is exemplary rather than restrictive, and it is obvious to those skilled in the art that there may be more embodiments and implementations within the scope of the embodiments described in the present disclosure. Although many possible feature combinations are shown in the drawings and discussed in the embodiments, many other combinations of the disclosed features are also possible. Unless specifically limited, any feature or element of any embodiment may be used in combination with any other feature or element in any other embodiment, or may replace any other feature or element in any other embodiment.

[0023] The present disclosure includes and contemplates combinations of features and elements known to those of ordinary skill in the art. The embodiments, features, and elements disclosed in the present application may also be combined with any conventional features or elements to form a unique invention scheme defined by the claims. Any features or elements of any embodiment may also be combined with features or elements from other invention schemes to form another unique invention scheme defined by the claims. Therefore, it should be understood that any feature shown or discussed in the present disclosure may be implemented individually or in any appropriate combination. Therefore, except for the limitations made according to the attached claims and their equivalents, the embodiments are not subject to other restrictions. In addition, various modifications and changes may be made within the scope of protection of the attached claims.

[0024] In addition, when describing representative embodiments, the specification may have presented the method or process as a specific sequence of steps. However, to the extent that the method or process does not rely on the specific order of the steps described herein, the method or process should not be limited to the steps of the specific order described. As will be understood by those of ordinary skill in the art, other sequences of steps are also possible. Therefore, the specific sequence of the steps set forth in the specification should not be interpreted as a limitation to the claims. In addition, the claims for the method or process should not be limited to the steps of performing them in the order written, and those skilled in the art can easily understand that these sequences can be changed and still remain within the spirit and scope of the embodiments of the present application.

[0025] In the present disclosure, "plurality" means two or more. Ordinal numbers such as "first" and "second" in the present disclosure are provided to avoid confusion of constituent elements, rather than to limit the quantity.

[0026] At least one embodiment of the present disclosure provides a fault diagnosis method and system, which can improve the problem that the large scale of the target system leads to an excessively large amount of data, which affects the accuracy of fault diagnosis, thereby improving the recognition accuracy of fault diagnosis.

[0027] Figure 1A flowchart of a fault diagnosis method provided by at least one embodiment of the present disclosure. Figure 1 As shown, the fault diagnosis method of this embodiment includes the following steps:

[0028] Step 101: Divide the target system into multiple macro modules according to the fault diagnosis requirement analysis result of the target system;

[0029] Step 102: for any macro module, determine the typical faults of the macro module and the corresponding test points and test parameters, and obtain the test data of the macro module using the test points and test parameters;

[0030] Step 103: For any macromodule, a machine learning fault diagnosis method is used to perform fault diagnosis based on the obtained test data.

[0031] The fault diagnosis method provided in this embodiment can improve the situation where the classification dimension surges and the fault response characteristics are similar by dividing the target system into macro modules, thereby simplifying data processing and improving the recognition accuracy of fault diagnosis.

[0032] The fault diagnosis method provided in this embodiment can be applied to different types of target systems, for example, electronic systems, mechanical systems, electromechanical hybrid systems, etc. However, the present disclosure is not limited thereto.

[0033] In some exemplary embodiments, the hardware units involved in implementing the same function in the target system are regarded as a macro module. A macro module can be regarded as a functional unit. In the fault diagnosis method of this embodiment, only the following two states of the macro module are distinguished: a fault-free state and a faulty state. The fault diagnosis method of this embodiment is used to determine whether there is a fault in the target system. If there is a fault, which macro module has the fault, so that targeted repairs can be performed later. By dividing the states of the macro modules into two types: fault-free and faulty, there is no need to distinguish the detailed fault types, which can simplify the data processing process and facilitate practical use. By dividing the entire target system into multiple macro modules, the fault diagnosis of the entire target system can be decomposed into the fault diagnosis of multiple macro modules, which can reduce the data processing complexity of large-scale target systems.

[0034] In some examples, the fault diagnosis requirement analysis results of the target system include at least one of the following: the composition of the target system (for example, the type and number of electronic components included in the target system, etc.), the structure (for example, the coupling relationship between the electronic components included in the target system, etc.), the working principle (for example, information such as the working mode of the target system), and the user's fault location requirements (for example, the fault location range (for example, the need to locate the faulty components or assemblies), the real-time requirements of fault location (for example, the time required to obtain the fault diagnosis results), etc.). However, this embodiment is not limited to this. In some examples, the division method and number of macro modules of the target system can be determined based on various information of the target system after comprehensive analysis and trade-offs.

[0035] In some examples, the target system is divided into multiple macro modules according to the principle of facilitating the isolation and positioning of faults and meeting the user's positioning range needs. For example, when it is necessary to obtain a diagnostic result of whether the target system has a fault in the shortest time, the entire target system can be divided into a macro module. For example, when it is necessary to diagnose a component failure in the target system, the entire target system can be divided into multiple macro modules, and each macro module can include a component (for example, a resistor, a capacitor, or a diode). For example, when it is necessary to diagnose a component or part failure in the target system, the entire target system can be divided into multiple macro modules, and each macro module can include components or parts (for example, circuit boards) involved in implementing the same function. The division method of the macro modules of this embodiment is flexible and diverse, and can be fully adapted to a variety of practical application scenarios. For example, when it is necessary to diagnose faults in an armored vehicle system during a military exercise, the armored vehicle system can be divided into macro modules at the subsystem level or the field replaceable unit level to facilitate rapid fault location and repair (for example, locating the faulty circuit board and replacing the entire circuit board); in the daily maintenance of the armored vehicle system, the armored vehicle system can be divided into macro modules at the component level to facilitate accurate fault location and repair (for example, locating the faulty components and replacing the specific components).

[0036] In some examples, when the target system is divided into multiple macro modules, the hardware units included in different macro modules may be completely different or partially the same, which is not limited in this embodiment.

[0037] In some exemplary embodiments, typical failures of any macromodule and corresponding test points and test parameters can be determined through Failure Mode, Effects and Criticality Analysis (FMECA). FMECA is an analysis technique summarized in engineering practice, based on failure modes and targeting failure effects or consequences. For example, for any macromodule, information such as typical failures and failure causes of the macromodule can be obtained through FMECA.

[0038] In some exemplary embodiments, when the target system is divided into multiple macro modules and there is no coupling relationship between different macro modules, the test points corresponding to the typical faults can be selected from the multiple signal output terminals of each macro module. When the target system is divided into multiple macro modules and there is a coupling relationship between different macro modules, the test points corresponding to the typical faults can be selected from the multiple signal output terminals and multiple internal nodes of each macro module. When the target system is divided into one macro module, the test points corresponding to the typical faults can be selected from the multiple internal nodes of the macro module. However, this embodiment is not limited to this.

[0039] In some exemplary embodiments, the test data of the macro module obtained by using the test points and test parameters of the macro module include: first test data corresponding to the test points and test parameters of the macro module in a fault-free state, and second test data corresponding to the test points and test parameters of the macro module in a faulty state. Among them, the second test data corresponding to the test points and test parameters of the macro module in a faulty state include: test data corresponding to the test points and test parameters of the macro module under various typical faults analyzed according to FMECA. In some examples, the first test data corresponding to the test points and test parameters of the macro module in a fault-free state are obtained by software simulation or simulated running experiments, bench experiments, etc.; the second test data corresponding to the test points and test parameters of the macro module in a typical fault state are obtained by simulating typical faults analyzed according to FMECA by software simulation or simulated running experiments, bench experiments, etc. However, this embodiment does not limit the method for obtaining test data.

[0040] In some exemplary embodiments, when the target system is divided into multiple macro modules and there is a coupling relationship between different macro modules, the fault judgment rule for determining the macro module in which the fault occurs in the target system may include: for any macro module whose test point includes a signal output terminal and an internal node, when the output data of the internal node is normal, but the output data of the signal output terminal is abnormal, the fault occurs in the macro module, or the subsequent N-level macro modules connected to the signal output terminal of the macro module, where N is an integer greater than 0; when the output data of the internal node is abnormal, and the output data of the signal output terminal is abnormal, the fault occurs in the macro module; when the output data of the internal node is abnormal, but the output data of the signal output terminal is normal, the fault occurs in the macro module. Among them, when the fault is masked by specific logic and is not sensitized and propagated, the output data of the internal node of the macro module is abnormal, the output data of the signal output terminal is normal, but the macro module fails.

[0041] In some exemplary embodiments, the test data of the macro module includes: the response signal data of the test point of the macro module, or the characteristic data obtained by processing the response signal data. In some examples, the response signal data of the test point of the macro module may include: real-time current value, real-time voltage value, real-time temperature value, etc. The characteristic data may include: the maximum value, minimum value, mean value, variance of the output voltage, the frequency spectrum and power spectrum of mechanical vibration, etc. In some examples, when the macro module is an electronic information and control module, the test data of the macro module may include DC signal data, AC signal data, etc., and the characteristic data of the macro module may include the average value, maximum value, minimum value, variance of the DC signal, the frequency, amplitude and effective value of the AC signal, etc. When the macro module is a mechanical module, the test data of the macro module may include: pressure, displacement, temperature, vibration acceleration and other data, and the characteristic data of the macro module may include: the mean value, variance, maximum value, minimum value of temperature and pressure, the mean value, variance, maximum value, minimum value of the vibration signal, or the frequency spectrum and power spectrum of the vibration signal, etc. However, this embodiment is not limited to this. In this exemplary embodiment, more recognizable feature data can be obtained by performing secondary processing on the response signal data, thereby better reflecting the different natures of the macro module in a non-fault state and a fault state.

[0042] In some examples, when the target system is a DC electronic system, during the fault diagnosis process of the target system, response signal data can be collected from the test points of the macro module, and the response signal data (e.g., current value, voltage value, etc.) collected from the test points can be directly used for fault diagnosis. In some examples, when the target system is an AC electronic system, during the fault diagnosis process of the target system, response signal data can be collected from the test points of the macro module, the directly collected response signal data (e.g., current value, voltage value, etc.) can be processed to obtain characteristic data (e.g., current average value, voltage average value, etc.), and the characteristic data can be used for fault diagnosis.

[0043] In some exemplary embodiments, the fault diagnosis method of this embodiment further includes: using test data of multiple combinations of macro modules to train multiple machine learning fault diagnosis methods respectively, wherein the signal types (e.g., current, voltage, temperature, pressure, amplitude, etc.) of the test data in different combinations are different or partially the same; selecting the best machine learning fault diagnosis method from the trained multiple machine learning fault diagnosis methods to perform fault diagnosis on the macro module. For example, if the target system is an electronic system, and the signal types of the test data of the macro module include: voltage, current, and temperature, then the test values ​​of voltage and current can be used as a combination of test data to train a machine learning fault diagnosis method, and the test values ​​of voltage, current, and temperature can be used as another combination of test data to train another machine learning fault diagnosis method, and according to the diagnostic accuracy of the two machine learning fault diagnosis methods, a machine learning fault diagnosis method with a high diagnostic accuracy is selected to perform fault diagnosis on the macro module of the target system.

[0044] In this exemplary embodiment, the test data includes the response signal data or feature data of the macromodule in different states (including the fault-free state and the faulty state). The feature data is the signal data after secondary processing of the response signal data of the macromodule in different states (for example, mean, variance, effective value, peak value, kurtosis, frequency component, power spectrum, etc.). The machine learning fault diagnosis method is essentially the correct binary classification of the response signal data or feature data (i.e., distinguishing between normal and abnormal data). The response signal of the macromodule generally has multiple types, such as vibration, pressure, temperature, voltage and current signals. As for which signal can better distinguish different states, it depends on the situation, and it is generally determined by comparing the actual fault detection accuracy. Assuming that a neural network classifier is designed, the test data of a macromodule in a fault-free state is used as one type of input data, and the test data under various typical fault states is used as another type of input data, and a binary classifier is designed. Assuming that the output state corresponds to [0, 1] (for example, output 0 means no fault has occurred, that is, the data is normal, and output 1 means a fault has occurred, that is, the output is abnormal), the input data of the neural network input layer can be selected as the test data of the macro module of the target system (that is, the original response signal data), or it can be selected as the feature data after secondary processing of the test data. Through training with a certain amount of data, the connection weights of different hidden layers and output layers of the neural network are adjusted and solidified. Finally, by selecting some test data under typical faults as test samples, the output classification accuracy of the neural network classifier under different input data is statistically analyzed. When the classification accuracy reaches a relatively acceptable level (such as 90%), the neural network classifier can be considered feasible.

[0045] In some exemplary embodiments, the fault diagnosis method of this embodiment further includes: training multiple different types of machine learning fault diagnosis methods, and selecting the optimal machine learning fault diagnosis method from the multiple machine learning fault diagnosis methods to perform fault diagnosis on the macro modules of the target system.

[0046] In some examples, the machine learning fault diagnosis method can adopt a supervised learning classification method, such as artificial neural network, support vector machine method, logistic regression classification method, etc. Artificial neural networks can have different models and different numbers of hidden layers. The diagnostic recognition accuracy of various machine learning algorithms needs to be obtained by the experimental and simulation data training and testing of the specific target system to be diagnosed. By using different test data or feature data as input, various factors (for example, recognition accuracy, training time, algorithm convergence stability, etc.) are considered to select the best method. On the basis of selecting the best method, the accuracy of classification judgment can be further improved by further improving the details such as the division method of the macro model, the selection of test points and test parameters, and the improvement of the feature extraction of test data, until the judgment accuracy reaches the ideal value.

[0047] The fault diagnosis method provided in this embodiment divides the target system into multiple macro modules and performs fault diagnosis based on artificial intelligence data-driven fault diagnosis method, which can simplify the data processing process, improve practicality, and is applicable to a variety of practical scenarios.

[0048] The technical solution of this embodiment is described below by taking an exemplary embodiment as an example.

[0049] Figure 2 The present invention is an example diagram of a macro module of a target system according to at least one embodiment of the present disclosure. Figure 2 Take a control system hardware circuit as an example, each macro module is a different functional block composed of different discrete components or integrated chip circuits, and the connection relationship between different macro modules is the connection sequence relationship of their respective signal output and input. Figure 2 As shown, in this exemplary embodiment, according to the fault requirement analysis result of the target system, the target system can be divided into 8 macro modules (ie, macro modules 21, 22, 23, 24, 25, 26, 27 and 28).

[0050] In this exemplary embodiment, Figure 2 As shown, there is a coupling relationship between the eight macro modules divided by the target system. Macro module 21 is coupled with macro modules 22 and 23, macro module 22 is coupled with macro module 21, macro module 24 and macro module 25, macro module 23 is coupled with macro modules 21 and 26, macro module 24 is coupled with macro modules 22 and 28, macro module 25 is coupled with macro modules 22, 26 and 27, and macro module 27 is coupled with macro modules 25 and 28. For any macro module, the typical fault of the macro module can be determined according to FMECA, and the test point corresponding to the typical fault can be selected from the signal output end of the macro module, and the corresponding test parameters can be determined. Figure 2 As shown, the test points of the macro module 21 include the internal node T1 and the signal output terminal T M1 and T M2 The test point of the macro module 22 includes a signal output terminal T M3 and T M4 The test point of the macro module 23 includes a signal output terminal T M5 The test point of the macro module 24 includes a signal output terminal T M6 The test point of the macro module 25 includes a signal output terminal T M7 and T M8 The test point of the macro module 26 includes a signal output terminal T M9 The test point of the macro module 27 includes a signal output terminal T M10 The test point of the macro module 28 includes a signal output terminal T M11 .

[0051] In this exemplary embodiment, the trained machine learning fault diagnosis method for the above-mentioned target system is used to identify the test data of the test points of each macro module to obtain an identification result of whether the test data of the test points of each macro module is normal. Then, based on the signal, energy transfer and connection relationship of each macro module, it is determined whether each macro module has a fault.

[0052] In this exemplary embodiment, Figure 2 There is coupling between the eight macromodules shown. In this example, since there is coupling between the eight macromodules, the fault judgment rule of the target system may include: for any macromodule whose test point includes a signal output terminal and an internal node, when the output data of the internal node is normal, but the output data of the signal output terminal is abnormal, the fault occurs in the macromodule, or the subsequent N-level macromodules connected to the signal output terminal of the macromodule, where N is an integer greater than 0; when the output data of the internal node is abnormal, and the output data of the signal output terminal is abnormal, the fault occurs in the macromodule; when the output data of the internal node is abnormal, but the output data of the signal output terminal is normal, the fault occurs in the macromodule.

[0053] Figure 3 FIG. 1 is a schematic diagram of the structure of a fault diagnosis system according to at least one embodiment of the present disclosure. Figure 3 As shown, the fault diagnosis system of this embodiment includes: a division module 301, a test module 302 and a diagnosis module 303. The division module 301 is configured to divide the target system into multiple macro modules according to the fault diagnosis requirement analysis results of the target system; the test module 302 is configured to determine the typical faults of any macro module and the corresponding test points and test parameters, and obtain the test data of the macro module by using the test points and the test parameters; the diagnosis module 303 is configured to perform fault diagnosis by using a machine learning fault diagnosis method according to the obtained test data for any macro module.

[0054] In some exemplary embodiments, the test module 302 is configured to determine the test point corresponding to the typical fault of any macro module in the following manner: when the target system is divided into multiple macro modules and there is no coupling relationship between different macro modules, the test point corresponding to the typical fault is selected from multiple signal output terminals of each macro module; when the target system is divided into multiple macro modules and there is a coupling relationship between different macro modules, the test point corresponding to the typical fault is selected from multiple signal output terminals and internal nodes of each macro module.

[0055] In some exemplary embodiments, the test data of the macro module obtained using the test points and the test parameters include: first test data corresponding to the test points and the test parameters of the macro module in a fault-free state, and second test data corresponding to the test points and the test parameters of the macro module in a faulty state.

[0056] In some exemplary embodiments, when the target system is divided into multiple macro modules and there is a coupling relationship between different macro modules, the fault judgment rules for determining the macro module in which the fault occurs in the target system include: for any macro module whose test points include a signal output terminal and an internal node, when the output data of the internal node is normal and the output data of the signal output terminal is abnormal, the fault occurs in the macro module, or the subsequent N-level macro modules connected to the signal output terminal of the macro module, N is an integer greater than 0; when the output data of the internal node is abnormal and the output data of the signal output terminal is abnormal, the fault occurs in the macro module; when the output data of the internal node is abnormal and the output data of the signal output terminal is normal, the fault occurs in the macro module.

[0057] In some exemplary embodiments, the test data of the macro module includes: response signal data of a test point of the macro module, or characteristic data obtained by processing the response signal data.

[0058] In some exemplary embodiments, the fault diagnosis system may further include: a training module and an optimization module. The training module is configured to respectively train multiple machine learning fault diagnosis methods using test data of multiple combinations of the macro module, wherein the signal types of the test data in different combinations are different or partially the same; and the optimization module is configured to select the best machine learning fault diagnosis method from the trained multiple machine learning fault diagnosis methods to perform fault diagnosis on the macro module.

[0059] Figure 4 FIG. 1 is a schematic diagram of the structure of another fault diagnosis system according to at least one embodiment of the present disclosure. Figure 4 As shown, it includes: a memory 401 and a processor 402, wherein the memory 401 stores a computer program, and when the computer program in the memory 401 is executed by the processor 402, the fault diagnosis method described in any of the above embodiments is executed.

[0060] An embodiment of the present disclosure further provides a computer-readable storage medium, on which computer-executable commands are stored, and the computer-executable commands are used to execute the fault diagnosis method described in any of the above embodiments.

[0061] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In hardware implementations, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all components may be implemented as software executed by a processor, such as a digital signal processor or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those of ordinary skill in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0062] Those skilled in the art should understand that the technical solutions of the present disclosure may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present disclosure, and all should be included in the scope of the claims of the present disclosure.

Claims

1. A fault diagnosis method, characterized in that: include: According to the fault diagnosis requirement analysis result of the target system, the target system is divided into a plurality of macro modules; For any macromodule, determine the typical faults of the macromodule and the corresponding test points and test parameters, and obtain test data of the macromodule using the test points and the test parameters; For any macromodule, the machine learning fault diagnosis method is used to perform fault diagnosis based on the obtained test data; For any macro module, determining a test point corresponding to a typical fault of the macro module includes: When the target system is divided into a plurality of macro modules and there is no coupling relationship between different macro modules, a test point corresponding to a typical fault is selected from a plurality of signal output ends of each macro module; When the target system is divided into a plurality of macro modules and different macro modules are coupled with each other, test points corresponding to typical faults are selected from a plurality of signal output terminals and internal nodes of each macro module; When the target system is divided into multiple macro modules and there is a coupling relationship between different macro modules, the fault judgment rule for determining the macro module in which the fault occurs in the target system includes: for any macro module whose test point includes a signal output terminal and an internal node, when the output data of the internal node is normal and the output data of the signal output terminal is abnormal, the fault occurs in the macro module, or the subsequent N-level macro modules connected to the signal output terminal of the macro module, N is an integer greater than 0; when the output data of the internal node is abnormal and the output data of the signal output terminal is abnormal, the fault occurs in the macro module; when the output data of the internal node is abnormal and the output data of the signal output terminal is normal, the fault occurs in the macro module.

2. The fault diagnosis method according to claim 1, characterized in that: The test data of the macro module obtained by using the test points and the test parameters include: first test data corresponding to the test points and the test parameters of the macro module in a fault-free state, and second test data corresponding to the test points and the test parameters of the macro module in a faulty state.

3. The fault diagnosis method according to claim 1, characterized in that: The test data of the macro module includes: response signal data of the test point of the macro module, or characteristic data obtained by processing the response signal data.

4. The fault diagnosis method according to claim 1, characterized in that: The fault diagnosis method further comprises: Using test data of multiple combinations of the macromodules to respectively train multiple machine learning fault diagnosis methods, wherein the signal types of the test data in different combinations are different or partially the same; An optimal machine learning fault diagnosis method is selected from multiple trained machine learning fault diagnosis methods to perform fault diagnosis on the macro module.

5. A fault diagnosis system, characterized in that: include: A division module, configured to divide the target system into a plurality of macro modules according to a fault diagnosis requirement analysis result of the target system; A test module configured to determine, for any macro module, typical faults of the macro module and corresponding test points and test parameters, and obtain test data of the macro module using the test points and the test parameters; A diagnosis module configured to perform fault diagnosis for any macromodule using a machine learning fault diagnosis method based on the obtained test data; The test module is specifically configured as follows: When the target system is divided into a plurality of macro modules and there is no coupling relationship between different macro modules, a test point corresponding to a typical fault is selected from a plurality of signal output ends of each macro module; When the target system is divided into a plurality of macro modules and different macro modules are coupled with each other, test points corresponding to typical faults are selected from a plurality of signal output terminals and internal nodes of each macro module; When the target system is divided into multiple macro modules and there is a coupling relationship between different macro modules, the fault judgment rule for determining the macro module in which the fault occurs in the target system includes: for any macro module whose test point includes a signal output terminal and an internal node, when the output data of the internal node is normal and the output data of the signal output terminal is abnormal, the fault occurs in the macro module, or the subsequent N-level macro modules connected to the signal output terminal of the macro module, N is an integer greater than 0; when the output data of the internal node is abnormal and the output data of the signal output terminal is abnormal, the fault occurs in the macro module; when the output data of the internal node is abnormal and the output data of the signal output terminal is normal, the fault occurs in the macro module.

6. A fault diagnosis system, characterized in that: include: A memory and a processor, wherein a computer program is stored in the memory, and when the computer program in the memory is executed by the processor, the fault diagnosis method according to any one of claims 1 to 4 is performed.

7. A computer-readable storage medium, characterized in that: The storage medium stores computer executable commands, and the computer executable commands are used to execute the fault diagnosis method according to any one of claims 1 to 4.

Images