Fault case database generation method, and apparatus

Abstract

Provided in the present disclosure is a fault case database generation method, the method comprising: acquiring fault state data and repair state data of a diagnosis object, the fault state data and the repair state data respectively corresponding to data obtained when the diagnosis object has a fault and data obtained when the fault is repaired; comparing the fault state data and the repair state data of the fault, and extracting fault features of the fault; and receiving a user-input fault cause tagged for the fault features, associating the fault features with the fault cause, storing the associated fault features and fault cause as a fault case, and storing the fault case into a fault case database.

Description

Method and device for generating fault case library Technical Field

[0001] The present disclosure mainly relates to the field of industrial digitalization, and in particular to a method and device for generating a fault case library. Background Art

[0002] With the increasing requirements for equipment reliability and safety, fault diagnosis and health maintenance have become a very important direction in the manufacturing industry.

[0003] Existing technologies use artificial intelligence for fault diagnosis. However, AI methods require the pre-establishment of neural network models. Training these models requires the collection of large amounts of data, a process that consumes considerable time and resources. Given the high cost of AI, fault diagnosis currently relies primarily on human effort, with field engineers relying on their experience. However, experienced field engineers are limited, and imparting fault diagnosis knowledge to field engineers is time-consuming and inefficient. Furthermore, the fault diagnosis knowledge acquired by field engineers is highly subjective and specific to the specific fault phenomenon, resulting in low accuracy and inability to be reused in other application scenarios.

[0004] Summary of the Invention

[0005] In order to solve the above technical problems, the present disclosure provides a method and device for generating a fault case library, so as to provide a standard fault case library and realize the standardization of fault diagnosis knowledge.

[0006] To achieve the above objectives, the present disclosure proposes a method for generating a fault case library, the method comprising:

[0007] Acquiring fault status data and repair status data of a diagnosis object, wherein the fault status data and the repair status data correspond to data when the diagnosis object fails and when the fault is repaired, respectively;

[0008] Comparing the fault status data and the repair status data of the fault to extract the fault characteristics of the fault;

[0009] The fault cause marked with the fault feature input by the user is received, the fault feature and the fault cause are associated, and the associated fault feature and the fault cause are saved as a fault case in a fault case library.

[0010] To this end, the fault characteristics of the fault are extracted by comparing the fault status data and repair status data. The fault characteristics are described by data, realizing the definition of fault characteristics at the mechanism level. The field engineers mark the fault causes corresponding to the fault characteristics based on their experience. Combining the objectivity of the fault characteristic definition and the fault diagnosis experience of the field engineers, a standard fault case library is provided to realize the standardization of fault diagnosis knowledge, thereby improving the accuracy of fault diagnosis.

[0011] Optionally, the method further includes determining statistical features from a statistical feature library, calculating the statistical features and using the calculated results as thresholds, and extracting fault features of the fault based on the fault status data, the repair status data, and the thresholds. Therefore, extracting fault features using statistical features improves the accuracy of fault feature extraction.

[0012] Optionally, the fault status data and the repair status data are time-domain data, and the method further includes: determining a time-domain statistical feature from a statistical feature library, calculating the time-domain statistical feature and using the calculated result as a threshold, and extracting a time-domain fault feature of the fault based on the time-domain fault status data, the repair status data, and the threshold. Therefore, extracting the time-domain fault feature using the time-domain statistical feature improves the accuracy of the time-domain fault feature extraction.

[0013] Optionally, the fault status data and the repair status data are time domain data, and the method further includes: performing a time-frequency conversion on the time domain fault status data and the repair status data to obtain frequency domain fault status data and the repair status data, determining frequency domain statistical features from a statistical feature library, calculating the frequency domain statistical features and using the calculated results as a threshold, and extracting frequency domain fault features of the fault based on the frequency domain fault status data, the repair status data, and the threshold. To this end, frequency domain fault features are extracted using frequency domain statistical features, and the frequency domain data includes periodic variation information of the diagnosed object, thereby improving the accuracy of frequency domain fault feature extraction.

[0014] Optionally, the method further includes searching for a frequency multiplier and / or sideband of a frequency corresponding to the extracted frequency-domain fault feature, and extracting the frequency multiplier and / or sideband as the frequency-domain fault feature if the amplitude of the frequency multiplier and / or sideband is greater than a threshold. Therefore, fault features can be quickly extracted using the frequency multiplier and / or sideband, thereby improving the efficiency of fault feature extraction.

[0015] Optionally, the method further comprises: deploying the fault case library to a cloud server. To this end, by deploying the fault case library to the cloud server, the scalability of the fault case library is improved.

[0016] The present disclosure also proposes a device for generating a fault case library, the device comprising:

[0017] An acquisition module, which acquires fault status data and repair status data of a diagnosis object, wherein the fault status data and the repair status data correspond to data when the diagnosis object fails and when the fault is repaired, respectively;

[0018] An extraction module compares the fault status data and the repair status data of the fault to extract the fault characteristics of the fault;

[0019] The generating module receives the fault cause marked with the fault feature input by the user, associates the fault feature with the fault cause, and saves the associated fault feature and fault cause as a fault case in a fault case library.

[0020] The present disclosure also provides an electronic device, comprising a processor, a memory, and instructions stored in the memory, wherein the instructions implement the above-mentioned method when executed by the processor.

[0021] The present disclosure also provides a computer-readable storage medium having computer instructions stored thereon. When the computer instructions are executed, the method described above is executed.

[0022] The present disclosure also provides a computer program product, comprising a computer program, which implements the method described above when executed by a processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The following figures are only intended to illustrate and explain the present disclosure, and do not limit the scope of the present disclosure.

[0024] FIG1 is a flow chart of a generating method according to an embodiment of the present disclosure;

[0025] FIG2 is a schematic diagram of a method for generating a fault case library according to an embodiment of the present disclosure;

[0026] FIG3 is a schematic diagram of extracting frequency fault features according to an embodiment of the present disclosure;

[0027] FIG4 is a schematic diagram of a generating device according to an embodiment of the present disclosure;

[0028] FIG5 is a schematic diagram of an electronic device according to an embodiment of the present disclosure.

[0029] DESCRIPTION OF REFERENCE NUMERALS 100 Generation method 110-130 Step 21 Extraction system 201 Fault status data acquisition module 202 Repair status data acquisition module 203 Feature extraction module 204 Statistical feature library 205 Case generator 22 Fault case library 400 Generation device 410 Acquisition module 420 Extraction module 430 Generation module 500 Electronic device 510 Processor 520 Memory DETAILED DESCRIPTION

[0030] In order to have a clearer understanding of the technical features, purposes and effects of the present disclosure, specific embodiments of the present disclosure are now described with reference to the accompanying drawings.

[0031] In the following description, many specific details are set forth to facilitate a full understanding of the present disclosure. However, the present disclosure may also be implemented in other ways different from those described herein. Therefore, the present disclosure is not limited to the specific embodiments disclosed below.

[0032] As used in this application and the claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" are not intended to refer to the singular but may include the plural. Generally speaking, the terms "comprises" and "include" only indicate the inclusion of the steps and elements specifically identified, and these steps and elements do not constitute an exclusive list. A method or apparatus may also include other steps or elements.

[0033] This application proposes a method for generating a fault case library. FIG1 is a flowchart of a generation method 100 according to an embodiment of the present disclosure. As shown in FIG1 , the method 100 includes:

[0034] Step 110, acquiring fault status data and repair status data of the diagnosis object, wherein the fault status data and the repair status data correspond to data when the diagnosis object fails and when the fault is repaired, respectively;

[0035] The diagnostic object may be a transmission system, which may include a drive, a transmission chain, and a motor. The method for generating a fault case library in the embodiments of the present application can be deployed in edge devices, drives, or PLCs to enable on-site fault diagnosis of the transmission system.

[0036] Fault status data and repair status data can be system data of the diagnostic object, such as temperature, voltage, speed, and torque data within the drive. Fault status data and repair status data can also be data collected by sensors installed on the diagnostic object, such as vibration data collected by a vibration sensor or audio data collected by a sound pressure sensor. By collecting fault status data and repair status data at a preset sampling period, time-domain fault status data and repair status data can be obtained.

[0037] The fault status data and the status repair data have the same data type, and the fault status data and the repair status data correspond to the data at the time the fault of the diagnostic object occurs and the data at the time the fault is repaired, respectively. For example, the fault status data may be the speed data of the motor when the fault occurs, and the repair status data may be the speed data of the motor when the fault is repaired. The fault status data and the repair status data may include multiple types of data, for example, the fault status data may include temperature data, voltage data, and speed data at the time of the fault, and the repair status data may include temperature data, voltage data, and speed data when the fault is repaired.

[0038] Step 120 , comparing the fault status data and the repair status data of the fault to extract the fault characteristics;

[0039] By comparing the fault status data and repair status data, the fault signature is extracted. Fault signatures are described through data, enabling a mechanism-level definition of fault signatures. This replaces the traditional phenomenon-level description of faults by field engineers, improving the objectivity, accuracy, and standardization of fault signature definitions. For example, if the fault status data for a motor indicates a rotor temperature that remains at 60°C, while the repair status data indicates a rotor temperature that remains at 50°C, the fault signature can be extracted as a rotor temperature that remains at 60°C.

[0040] In some embodiments, the method further includes: determining statistical features from a statistical feature library, calculating the statistical features and using the calculated results as a threshold, and extracting the fault features of the fault based on the fault status data, the repair status data and the threshold. Specifically, the statistical feature library may include multiple statistical features, and the statistical features are statistical descriptions of the status data. The statistical features may be time domain statistical features or frequency domain statistical features. The time domain statistical features may include mean, root mean square, kurtosis, etc. The frequency domain statistical features may include frequency mean, frequency center, third-order center distance, etc. One or more statistical features may be selected and multiple statistical features may be calculated separately. The fault features of the fault are extracted based on the fault status data, the repair status data and the threshold, that is, if the gap between the fault status data and the repair status data is greater than the threshold, the fault status data is extracted as the fault feature. To this end, the fault features are extracted through statistical features, thereby improving the accuracy of the fault feature extraction.

[0041] For example, the fault status data of the motor is that the rotor temperature remains at 60°C, and the repair status data of the motor is that the rotor temperature remains at 50°C. The difference between the two is 10°C. The determined statistical feature is the root mean square, and the result of the root mean square calculation is 5°C. 5°C is set as the threshold, that is, the difference between the fault status data and the repair status data is greater than the threshold, then the rotor temperature remains at 60°C, which is extracted as the fault status feature.

[0042] In some embodiments, the fault status data and repair status data are time-domain data. The method further includes determining time-domain statistical features from a statistical feature library, calculating the time-domain statistical features and using the calculated results as thresholds, and extracting time-domain fault features of the fault based on the time-domain fault status data, repair status data, and thresholds. Specifically, the time-domain statistical features may include mean, root mean square (RMS), kurtosis, etc. The time-domain data may be temperature data, voltage data, sound pressure data, etc. Therefore, extracting time-domain fault features using time-domain statistical features improves the accuracy of time-domain fault feature extraction.

[0043] In some embodiments, the fault status data and the repair status data are time domain data, and the method further includes: performing time-frequency conversion on the fault status data and the repair status data in the time domain to obtain fault status data and the repair status data in the frequency domain, determining the frequency domain statistical features from the statistical feature library, calculating the frequency domain statistical features and using the calculation results as the threshold, and extracting the frequency domain fault features of the fault based on the frequency domain fault status data, the repair status data and the threshold. Specifically, the frequency domain statistical features may include the frequency mean, the frequency center, the third-order center distance, etc. The frequency domain data may be speed data, torque data, vibration data, etc., and the frequency domain data includes periodic variation information of the diagnosis object. To this end, the frequency domain fault features are extracted through the frequency domain statistical features, and the frequency domain data includes periodic variation information of the diagnosis object, thereby improving the accuracy of the frequency domain fault feature extraction.

[0044] In some embodiments, the method further includes searching for multiples and / or side frequencies of the frequency corresponding to the extracted frequency-domain fault feature, and if the amplitude of the multiples and / or side frequencies is greater than a threshold, extracting the multiples and / or side frequencies as frequency-domain fault features. For example, the frequency of 50 Hz has the largest amplitude, and the frequency point of 50 Hz is extracted as the frequency-domain fault feature. Subsequently, searching for multiples of 50 Hz, 100 Hz and 150 Hz, and if their amplitudes are greater than a threshold, extracting 100 Hz and 150 Hz as frequency-domain fault features. To this end, fault features can be quickly extracted using multiples and / or side frequencies, thereby improving the efficiency of fault feature extraction.

[0045] FIG3 is a schematic diagram of extracting frequency fault features according to an embodiment of the present disclosure. As shown in FIG3 , the amplitude of frequency point f1 in spectrum a exceeds threshold T and is the largest, and frequency point f1 is extracted as a characteristic frequency point (spectrum b). The multiples f2 (2f1) and f3 (3f1) of f1 are searched, and their amplitudes are also greater than threshold T. Frequency points f2 and f3 are also extracted as characteristic frequency points (spectrum c). Frequency points f1, f2, and f3 are removed from the spectrum, and frequency domain fault features are continuously extracted in the remaining spectrum d until all frequency domain fault features are extracted.

[0046] Step 130 : receiving the fault cause marked with the fault feature input by the user, associating the fault feature with the fault cause, and saving the associated fault feature and fault cause as a fault case in a fault case library.

[0047] The user can input the fault cause for the fault feature mark through a graphical user interface, voice, etc. For example, for the fault feature that the rotor temperature remains at 60°C, the user marks the fault cause as rotor imbalance. It can be understood that in the embodiment of the present application, the user can be a field engineer who has rich knowledge of fault diagnosis. The fault feature "rotor temperature remains at 60°C" and the corresponding fault cause "rotor imbalance" are saved as fault cases in the fault case library. To this end, the fault feature of the fault is extracted by comparing the fault status data and the repair status data of the fault. The fault feature is described by data, which realizes the definition of fault features at the mechanism level. The field engineer marks the fault cause corresponding to the fault feature based on experience. Combining the objectivity of the fault feature definition and the fault diagnosis experience of the field engineer, a standard fault case library is provided to achieve the standardization of fault diagnosis knowledge, thereby improving the accuracy of fault diagnosis.

[0048] In some embodiments, the method further includes: deploying the fault case library to a cloud server. The cloud server can provide a rich set of algorithms and massive computing power. Therefore, by deploying the fault case library to the cloud server, the scalability of the fault case library is improved.

[0049] FIG2 is a schematic diagram of a method for generating a fault case library according to an embodiment of the present disclosure. As shown in FIG2 , the extraction system 21 includes a fault status data acquisition module 201, a repair status data acquisition module 202, a feature extraction module 203, a statistical feature library 204, and a case generator 205. The fault status data acquisition module 201 is used to acquire fault status data, and the repair status data acquisition module 202 is used to acquire repair status data. The feature extraction module 203 receives the fault status data sent by the fault status data acquisition module 201 and the repair status data sent by the repair status data acquisition module 202, determines statistical features from the statistical feature library 204, and extracts fault features based on the fault status data, repair status data, and statistical features. The case generator 205 receives the fault features sent by the feature extraction module 203 and generates a fault case based on the fault features and the fault cause marked for the fault features by the user. The fault case library 22 receives the fault case sent by the case generator 205.

[0050] An embodiment of the present application provides a method for generating a fault case library, which extracts the fault characteristics of the fault by comparing the fault status data and repair status data of the fault. The fault characteristics are described by data, realizing the definition of fault characteristics at the mechanism level. Field engineers mark the fault causes corresponding to the fault characteristics based on their experience. Combining the objectivity of the fault characteristic definition and the fault diagnosis experience of field engineers, a standard fault case library is provided, which realizes the standardization of fault diagnosis knowledge and thereby improves the accuracy of fault diagnosis.

[0051] The present application also proposes a device for generating a fault case library. FIG4 is a schematic diagram of a generating device 400 according to an embodiment of the present disclosure. As shown in FIG4 , the device 400 includes:

[0052] An acquisition module 410 acquires fault status data and repair status data of a diagnosis object, wherein the fault status data and the repair status data correspond to data when the diagnosis object fails and when the fault is repaired, respectively;

[0053] An extraction module 420 compares the fault status data and the repair status data of the fault to extract the fault characteristics of the fault;

[0054] The generation module 430 receives the fault cause marked with the fault feature input by the user, associates the fault feature with the fault cause, and saves the associated fault feature and fault cause as a fault case in the fault case library.

[0055] The present disclosure also provides an electronic device 500. FIG5 is a schematic diagram of an electronic device 500 according to an embodiment of the present disclosure. As shown in FIG5 , the electronic device 500 includes a processor 510 and a memory 520. The memory 520 stores instructions, wherein the instructions, when executed by the processor 510, implement the method 100 described above.

[0056] The present disclosure further provides a computer-readable storage medium having computer instructions stored thereon. When the computer instructions are executed, the method 100 described above is executed.

[0057] The present disclosure also provides a computer program product, including a computer program, which implements the method 100 described above when the computer program is executed by a processor.

[0058] Some aspects of the methods and apparatus of the present disclosure may be performed entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. The above hardware or software may be referred to as "data blocks," "modules," "engines," "units," "components," or "systems." The processor may be one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DAPDs), programmable logic devices (PLCs), field programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, or combinations thereof. In addition, various aspects of the present disclosure may be expressed as computer products located in one or more computer-readable media, the product including computer-readable program code. For example, computer-readable media may include, but are not limited to, magnetic storage devices (e.g., hard disks, floppy disks, tapes, etc.), optical disks (e.g., compact disks (CDs), digital versatile disks (DVDs), etc.), smart cards, and flash memory devices (e.g., cards, sticks, key drives, etc.).

[0059] Flowcharts are used herein to illustrate the operations performed by the methods according to the embodiments of the present application. It should be understood that the preceding operations are not necessarily performed in exact order. Instead, the various steps may be performed in reverse order or simultaneously. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0060] It should be understood that although this specification is described according to various embodiments, not every embodiment contains only one independent technical solution. This narrative method of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.

[0061] The above description is only an illustrative embodiment of the present disclosure and is not intended to limit the scope of the present disclosure. Any equivalent changes, modifications and combinations made by any person skilled in the art without departing from the concept and principle of the present disclosure shall fall within the scope of protection of the present disclosure.

[0062] Nouns and pronouns referring to persons in this patent application are not limited to a specific gender.

Claims

1. A method (100) for generating a fault case library, characterized in that: The method comprises: Acquiring fault status data and repair status data of a diagnosis object, wherein the fault status data and the repair status data correspond to data when the diagnosis object fails and when the fault is repaired, respectively (110); Comparing the fault state data and the repair state data of the fault to extract the fault characteristics of the fault (120); Receive the fault cause marked by the fault feature input by the user, associate the fault feature with the fault cause, and save the associated fault feature and fault cause as a fault case in a fault case library (130).

2. The method (100) according to claim 1, characterized in that The method (100) further comprises: determining statistical features from a statistical feature library, calculating the statistical features and using the calculation result as a threshold, and extracting the fault feature of the fault based on the fault status data, the repair status data and the threshold.

3. The method (100) according to claim 2, characterized in that The fault status data and the repair status data are time domain data. The method (100) further comprises: determining a time domain statistical feature from a statistical feature library, calculating the time domain statistical feature and using the calculation result as a threshold, and extracting the time domain fault feature of the fault based on the time domain fault status data, the repair status data and the threshold.

4. The method (100) according to claim 2, characterized in that The fault status data and the repair status data are time domain data, and the method (100) further comprises: performing time-frequency conversion on the fault status data and the repair status data in the time domain to obtain the fault status data and the repair status data in the frequency domain, determining frequency domain statistical features from a statistical feature library, calculating the frequency domain statistical features and using the calculation result as a threshold, and extracting the frequency domain fault features of the fault based on the fault status data, the repair status data and the threshold in the frequency domain.

5. The method (100) according to claim 4, characterized in that The method (100) further comprises: searching for the frequency multiples and / or side frequencies of the frequency corresponding to the extracted frequency domain fault feature, and if the amplitude of the frequency multiples and / or side frequencies is greater than the threshold, extracting the frequency multiples and / or side frequencies as the frequency domain fault feature.

6. The method (100) according to claim 1, characterized in that The method (100) further includes: deploying the fault case library to a cloud server.

7. A device (400) for generating a fault case library, characterized in that: The device (400) comprises: An acquisition module (410) acquires fault status data and repair status data of a diagnosis object, wherein the fault status data and the repair status data correspond to data when the diagnosis object fails and when the fault is repaired, respectively; An extraction module (420) compares the fault status data and the repair status data of the fault to extract the fault characteristics of the fault; A generation module (430) receives a fault cause input by a user for the fault feature mark, and associates the fault The fault characteristics and the fault causes are stored as fault cases in a fault case library.

8. An electronic device (500), comprising a processor (510), a memory (520), and instructions stored in the memory (520), wherein the instructions, when executed by the processor (510), implement the method (100) according to any one of claims 1 to 6.

9. A computer-readable storage medium having computer instructions stored thereon, wherein the computer instructions, when executed, perform the method (100) according to any one of claims 1 to 6.

10. A computer program product, characterized in that The invention comprises a computer program for implementing the method (100) according to any one of claims 1 to 6 when the computer program is executed by a processor.

Images