Fault early warning methods
By analyzing real-time operation logs and historical data of electronic devices, and using a fault analysis library for feature extraction and clustering, the problem of irreversible damage caused by maintenance after electronic device failures is solved, enabling early warning and prediction of faults, and improving the maintenance efficiency and reliability of equipment.
Patent Information
- Application Number
- CN202510580923.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-07
- Publication Date
- 2026-11-10
AI Technical Summary
Existing electronic devices are usually only repaired after a malfunction occurs, resulting in irreversible damage to the equipment and affecting its long-term effective use.
By acquiring real-time operating logs of electronic devices, feature extraction and clustering are performed using a fault analysis library to analyze trends and compare similarity with a pre-acquired set of fault type evolution trends to predict upcoming fault types.
It enables early warning of electronic equipment failures, reduces equipment loss and damage, and improves equipment maintenance efficiency and reliability.
Smart Images

Figure CN123069094A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of fault early warning technology, and more specifically, to a fault early warning method. Background Technology
[0002] Electronic equipment refers to electrical devices composed of microelectronic components. Specifically, it comprises electronic components such as integrated circuits, transistors, and vacuum tubes, and functions using electronic technology and software. This includes computers and computer-controlled robots, numerical control (NC) or programmable control (CNC) systems.
[0003] Currently, electronic devices are often only repaired after a malfunction occurs. Once a malfunction occurs, it can easily cause irreversible damage to the electronic device, which is very detrimental to its long-term effective use. Therefore, how to predict the malfunction of electronic devices in advance is a technical problem that urgently needs to be solved. Summary of the Invention
[0004] In view of this, the present invention proposes a fault early warning method that can predict the faults of electronic devices in advance.
[0005] In a first aspect, the present invention proposes a fault early warning method, comprising:
[0006] Obtain the real-time operation logs of the electronic device; extract features from the real-time operation logs according to the fault analysis library to obtain real-time features; cluster the real-time features according to the fault features corresponding to different fault types to obtain the feature set to be analyzed; obtain the trend set of the feature set to be analyzed, and compare the similarity of the trend set with the evolution trend set corresponding to different fault types. Based on the comparison results, determine whether the electronic device has a predicted fault and determine the fault type of the predicted fault.
[0007] In this way, since the evolution trend set corresponding to different fault types can characterize the changing trend of different types of faults, the feature set to be analyzed, which includes fault characteristics of different fault types, can be extracted from the real-time operation log of electronic equipment. Then, the similarity between the changing trend of the feature set to be analyzed and the pre-acquired evolution trend set of different fault types can be compared. Based on the comparison results, it is possible to predict the fault that the electronic equipment will soon experience. The fault type of the predicted fault can be determined according to the comparison results, which can provide early warning of the fault and reduce the loss or damage caused by the fault of electronic equipment.
[0008] In one possible approach, the different types of historical faults of the electronic device and the corresponding historical operation logs of the historical faults are obtained;
[0009] Fault features are extracted from historical operation logs to obtain a fault analysis library.
[0010] In one possible approach, the operation logs of different fault types of electronic devices at the time of the fault occurrence and the operation logs within a preset time period before the fault occurrence are obtained as the initial data corresponding to different fault types.
[0011] Perform parameter analysis on the initial data to filter out features with trend change characteristics as screening features;
[0012] For the initial data corresponding to each fault type, the data of the filtering features within a preset time period are sorted according to the time series to obtain the evolution trend of the filtering features within the preset time period; one fault type corresponds to at least one filtering feature, and one filtering feature corresponds to one evolution trend.
[0013] For each fault type among different fault types, the evolution trend of all the screening features corresponding to the fault type is determined as the evolution trend set of the fault type.
[0014] In one possible approach, the fault features corresponding to the fault type are used as the screening features corresponding to the fault type; based on the fault features corresponding to different fault types, the real-time features are clustered to obtain the feature set to be analyzed, including:
[0015] Identify the fault type of all faults in the fault analysis library;
[0016] For each fault type among all fault types, the filtering features in the evolution trend set corresponding to the fault type are determined as the filtering feature set corresponding to the fault type.
[0017] Deduplication is performed on all the screening features in the screening feature set corresponding to all fault types to obtain the target screening features;
[0018] Target features are selected from real-time features to complete the clustering of real-time features and obtain the feature set to be analyzed.
[0019] In one possible approach, for each target screening feature in the feature set to be analyzed, the data of the target screening feature within a preset time period are sorted according to the time series to obtain the changing trend of the target screening feature;
[0020] The changing trends of various target screening features in the feature set to be analyzed are reorganized according to the screening feature set corresponding to different fault types to obtain the changing trend set of the feature set to be analyzed.
[0021] One possible approach is to compare the set of changing trends with the sets of evolution trends corresponding to different fault types, including:
[0022] For each fault type, the change trend of each target selected feature in the change trend set is correlated with the change trend of each selected feature in the evolution trend set to calculate the similarity. The calculation formula includes:
[0023]
[0024] Where, x i For a target, select the i-th observation of the changing trend of features, y i Let i be the i-th observation representing the evolution trend of a selected feature. The mean of all observations in the trend of change. Let be the mean of all observations of the trend, n be the number of observations, and r be the similarity between the trend and the evolution trend.
[0025] In one possible approach, when r > 0, the changing trend of a target screening feature is similar to the evolution trend of a screening feature; based on the comparison results, it is determined that the electronic device has a predicted fault and the fault type of the predicted fault is determined, including:
[0026] If the change trend of each target screening feature in the change trend set is similar to the evolution trend of a screening feature in the target evolution trend set, then the electronic device is determined to have a predicted fault and the fault type corresponding to the target evolution trend set is the predicted fault type.
[0027] One possible approach is to obtain the fault operation log when the electronic device malfunctions.
[0028] Extract fault features from the fault operation log to obtain the current fault features;
[0029] The current fault characteristics are compared with the fault characteristics in the fault analysis database to determine the current fault type of the electronic device.
[0030] In one possible approach, the current fault characteristics are compared with fault characteristics in a fault analysis database to determine the current fault type of the electronic device, including:
[0031] Calculate the similarity between the fault features of each fault type in the fault analysis library and the current fault features to obtain multiple similarity scores;
[0032] The fault type corresponding to the smallest similarity among multiple similarities is determined as the current fault type of the electronic device; wherein the similarity is determined according to the following formula:
[0033]
[0034] In the formula, x jFor the j-th current fault feature, y j Let be the j-th fault feature of the target fault type in the fault analysis library, m be the number of all fault features in the target fault type, d be the similarity, and the target fault type be any fault type.
[0035] In one possible approach, if the current fault type of the electronic device is determined to be different from the actual fault type of the electronic device, the fault operation log, the operation log within a preset time period before the fault occurred, and the actual fault type of the electronic device are stored in the fault analysis database based on the maintenance results of the electronic device.
[0036] Secondly, the present invention proposes a fault early warning device, comprising: an acquisition unit, an extraction unit, a clustering unit, a comparison unit, and a determination unit.
[0037] The acquisition unit is used to acquire the real-time operating logs of electronic devices.
[0038] The extraction unit is used to extract features from real-time operation logs according to the fault analysis library to obtain real-time features.
[0039] Clustering units are used to cluster real-time features according to the fault characteristics corresponding to different fault types to obtain the feature set to be analyzed.
[0040] The acquisition unit is also used to acquire the set of changing trends of the feature set to be analyzed.
[0041] The comparison unit is used to compare the similarity between the set of change trends and the set of evolution trends corresponding to different fault types.
[0042] The determination unit is used to determine whether the electronic device has a predicted fault and to determine the type of the predicted fault based on the comparison results.
[0043] Thirdly, a fault warning device is provided, including a memory and a processor; the memory is used to store computer execution instructions, and the processor is connected to the memory via a bus; when the fault warning device is running, the processor executes the computer execution instructions stored in the memory, so that the fault warning device performs the fault warning method described in the first aspect.
[0044] The fault warning device can be a network device or a component of a network device, such as a chip system within the network device. This chip system supports the network device in implementing the functions involved in the first aspect and any of its possible implementations, such as acquiring, determining, and transmitting the data and / or information involved in the aforementioned fault warning method. The chip system includes a chip, but may also include other discrete devices or circuit structures.
[0045] Fourthly, a computer-readable storage medium is provided, comprising computer-executable instructions that, when executed on a computer, cause the computer to perform the fault warning method described in the first aspect.
[0046] Fifthly, a computer program product is also provided, which includes computer instructions that, when executed on a fault warning device, cause the fault warning device to perform the fault warning method as described in the first aspect above.
[0047] It should be noted that the aforementioned computer instructions may be stored, in whole or in part, on a computer-readable storage medium. This computer-readable storage medium may be packaged together with the processor of the fault warning device, or it may be packaged separately from the processor of the fault warning device; this application does not limit this.
[0048] The descriptions of the second, third, fourth, and fifth aspects of this application can be found in the detailed description of the first aspect.
[0049] In the embodiments of this application, the name of the aforementioned fault warning device does not limit the device or functional module itself. In actual implementation, these devices or functional modules may appear under other names. For example, the receiving unit may also be called a receiving module, receiver, etc. As long as the function of each device or functional module is similar to that of this application, it falls within the scope of the claims of this application and its equivalents. Attached Figure Description
[0050] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:
[0051] Figure 1 A system architecture diagram provided for an embodiment of the present invention;
[0052] Figure 2 A hardware structure diagram for fault early warning provided in an embodiment of the present invention;
[0053] Figure 3 A flowchart illustrating a fault early warning method provided in an embodiment of the present invention;
[0054] Figure 4 A flowchart illustrating another fault early warning method provided in an embodiment of the present invention;
[0055] Figure 5 A flowchart illustrating another fault early warning method provided in an embodiment of the present invention;
[0056] Figure 6 A flowchart illustrating another fault early warning method provided in an embodiment of the present invention;
[0057] Figure 7 A flowchart illustrating another fault early warning method provided in an embodiment of the present invention;
[0058] Figure 8 A flowchart illustrating another fault early warning method provided in an embodiment of the present invention;
[0059] Figure 9 This is a schematic diagram of a fault early warning device provided in an embodiment of the present invention. Detailed Implementation
[0060] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, embodiments and features in the embodiments of the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.
[0061] It should be noted that in the embodiments of this application, the words "exemplary" or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.
[0062] To facilitate a clear description of the technical solutions of the embodiments of this application, the terms "first" and "second" are used in the embodiments of this application to distinguish the same or similar items with essentially the same function and effect. Those skilled in the art can understand that the terms "first" and "second" are not intended to limit the quantity or execution order.
[0063] Electronic equipment refers to electrical devices composed of microelectronic components. Specifically, it comprises electronic components such as integrated circuits, transistors, and vacuum tubes, and functions using electronic technology and software. This includes computers and computer-controlled robots, numerical control (NC) or programmable control (CNC) systems.
[0064] Currently, electronic devices are often only repaired after a malfunction occurs. Once a malfunction occurs, it can easily cause irreversible damage to the device, which is highly detrimental to its long-term effective use. Therefore, how to predict electronic device malfunctions is a pressing technical problem that needs to be solved. Furthermore, locating the fault after it occurs is quite difficult, hindering rapid repair of electronic devices.
[0065] In response, this application provides a fault early warning method, which involves obtaining the real-time operation log of an electronic device; extracting features from the real-time operation log according to a fault analysis library to obtain real-time features; clustering the real-time features according to the fault features corresponding to different fault types to obtain a feature set to be analyzed; obtaining the trend set of the feature set to be analyzed, and comparing the similarity between the trend set and the evolution trend set corresponding to different fault types, and determining the type of the predicted fault based on the comparison results.
[0066] In this way, since the evolution trend set corresponding to different fault types can characterize the changing trend of different types of faults, the feature set to be analyzed, which includes fault characteristics of different fault types, can be extracted from the real-time operation log of electronic equipment. Then, the similarity between the changing trend of the feature set to be analyzed and the pre-acquired evolution trend set of different fault types can be compared. Based on the comparison results, it is possible to predict the fault that the electronic equipment will soon experience. The fault type of the predicted fault can be determined according to the comparison results, which can provide early warning of the fault and reduce the loss or damage caused by the fault of electronic equipment.
[0067] This application provides a fault early warning system, such as Figure 1 As shown, the system includes: electronic device 101 and fault warning device 102.
[0068] The electronic device 101 and the fault warning device 102 can be communicatively connected, either via wired or wireless communication. In one possible implementation, the electronic device 101 and the fault warning device 102 can be the same device, or the fault warning device 102 can be a part of the electronic device 101.
[0069] In this embodiment, the fault warning device 102 can acquire the real-time operation log of the electronic device 101 and extract features from the real-time operation log to obtain real-time features. Then, the fault warning device 102 can cluster the real-time features according to the fault features of different fault types to obtain a feature set to be analyzed. Subsequently, the fault warning device 102 can acquire the change trend set of the feature set to be analyzed and compare the similarity of the change trend set with the evolution trend sets corresponding to different fault types. Based on the comparison results, it can determine that the electronic device has a fault (i.e., a predicted fault) and determine the fault type of the predicted fault.
[0070] In this way, since the evolution trend set corresponding to different fault types can characterize the changing trend of different types of faults, the feature set to be analyzed, which includes fault characteristics of different fault types, can be extracted from the real-time operation log of electronic equipment. Then, the similarity between the changing trend of the feature set to be analyzed and the pre-acquired evolution trend set of different fault types can be compared. Based on the comparison results, it is possible to predict the fault that the electronic equipment will soon experience. The fault type of the predicted fault can be determined according to the comparison results, which can provide early warning of the fault and reduce the loss or damage caused by the fault of electronic equipment.
[0071] In one possible implementation, the electronic device 101 and the fault warning device 102 can be a terminal, a server, or other types of electronic devices.
[0072] Optionally, the aforementioned terminal may be at least one of the following devices: smartphone, smartwatch, desktop computer, laptop, virtual reality terminal, augmented reality terminal, wireless terminal, and laptop computer.
[0073] Optionally, the server mentioned above can be one of the servers in a server cluster (composed of multiple servers), a chip in the server, a system-on-a-chip in the server, or a virtual machine (VM) deployed on a physical machine. This application embodiment does not limit this.
[0074] The basic hardware structure of electronic device 101 and fault warning device 102 includes Figure 2 The fault warning device shown includes the following components. Figure 2 Taking the fault warning device shown as an example, the hardware structure of electronic device 101 and fault warning device 102 will be introduced.
[0075] like Figure 2The diagram shown is a hardware structure schematic of a fault early warning device provided in an embodiment of this application. The fault early warning device includes a processor 21, a memory 22, a communication interface 23, and a bus 24. The processor 21, memory 22, and communication interface 23 are connected via the bus 24.
[0076] Processor 21 is the control center of the fault warning device. It can be a single processor or a collective term for multiple processing elements. For example, processor 21 can be a general-purpose central processing unit (CPU) or other general-purpose processors. Among them, the general-purpose processor can be a microprocessor or any conventional processor.
[0077] As one embodiment, processor 21 may include one or more CPUs, for example Figure 2 CPU 0 and CPU 1 are shown in the diagram.
[0078] The memory 22 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto.
[0079] In one possible implementation, the memory 22 can exist independently of the processor 21. The memory 22 can be connected to the processor 21 via a bus 24 and is used to store instructions or program code. When the processor 21 calls and executes the instructions or program code stored in the memory 22, it can implement the fault warning method provided in the following embodiments of this application.
[0080] In this embodiment, the software programs stored in the memory 22 are different for the electronic device 101 and the fault warning device 102, so the functions implemented by the electronic device 101 and the fault warning device 102 are different. The functions performed by each device will be described with reference to the following flowchart.
[0081] In another possible implementation, the memory 22 can also be integrated with the processor 21.
[0082] Communication interface 23 is used for connecting the fault warning device to other devices via a communication network, such as Ethernet, wireless access network, or wireless local area network (WLAN). Communication interface 23 may include a receiving unit for receiving data and a sending unit for sending data.
[0083] Bus 24 can be an industry standard architecture (ISA) bus, a peripheral component interconnect (PCI) bus, or an extended industry standard architecture (EISA) bus, etc. This bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 2 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.
[0084] It should be pointed out that, Figure 2 The structure shown does not constitute a limitation on the fault warning device, except Figure 2 In addition to the components shown, the fault warning device may include more or fewer components than those shown, or combine certain components, or have different component arrangements.
[0085] The fault early warning method provided in the embodiments of this application will be described in detail below with reference to the accompanying drawings.
[0086] The fault early warning method provided in this application embodiment is applied to Figure 1 The fault warning device 102 in the fault warning system shown. For example... Figure 3 As shown, the fault early warning method includes:
[0087] S301. Obtain the real-time operation log of the electronic device.
[0088] The operation log includes, but is not limited to, timestamps, device status, operating parameters, exception information, operation records, fault records, and maintenance records.
[0089] Electronic devices can transmit their current operating logs (i.e., real-time operating logs) to a fault warning device. Since real-time operating logs can characterize the current operating status of electronic devices, it is possible to determine whether an electronic device has malfunctioned through real-time operating logs.
[0090] S302. Extract features from the real-time operation logs according to the fault analysis library to obtain real-time features.
[0091] The fault analysis library includes features related to historical faults of electronic devices.
[0092] Based on the fault characteristics related to electronic equipment faults in the fault analysis database, fault characteristics are extracted from the real-time operation logs to obtain real-time features. In this way, the fault early warning device can filter out the fault characteristics corresponding to faults from the current operation logs, and thus determine whether a fault is about to occur based on these fault characteristics.
[0093] S303. Cluster the real-time features according to the fault characteristics corresponding to different fault types to obtain the feature set to be analyzed.
[0094] Since the extracted real-time features are the fault features corresponding to the faults, these real-time features can be clustered according to the fault features (or the types of fault features) corresponding to each fault type, thus obtaining the feature set to be analyzed. After clustering, the real-time features in the feature set to be analyzed are divided into real-time features corresponding to different fault types, that is, these classified fault types are the faults that electronic devices may be about to experience.
[0095] In one possible implementation, the fault characteristics corresponding to each fault type can be either obtained in advance or found in a fault analysis library.
[0096] S304. Obtain the trend set of the feature set to be analyzed, and compare the similarity of the trend set with the evolution trend set corresponding to different fault types. Based on the comparison results, determine whether the electronic equipment has a predicted fault and determine the fault type of the predicted fault.
[0097] Since faults exhibit corresponding trends before they occur, to determine if an electronic device currently has a fault, a set of trends for the features to be analyzed can be obtained, thus characterizing the current operating state of the electronic device. Then, the similarity of this set of trends can be compared with pre-acquired evolution trend sets corresponding to different fault types. Because different types of faults correspond to different evolution trends (i.e., the changing trends of the electronic device's operating state), it's possible to determine which fault type's evolution trend set is similar to the current operating state's trend (i.e., the trend set). If similar evolution trend sets exist, it can be determined that the electronic device is about to fail (i.e., a predicted fault). The fault type corresponding to the similar evolution trend set can be identified as the predicted fault type. This allows for prediction before a fault occurs and, if a predicted fault is imminent, for identification of the predicted fault type, enabling preventative maintenance and reducing the failure rate of electronic devices.
[0098] In some embodiments, combined with Figure 3,like Figure 4 As shown, the method provided in this application embodiment further includes:
[0099] S401. Obtain the different types of historical faults of electronic devices and the corresponding historical operation logs.
[0100] Since the fault analysis database is obtained in advance, prior to S301, the fault warning device can acquire historical faults of the electronic equipment, as well as the corresponding historical operation logs for each historical fault. These historical faults include different types of faults, and the acquired historical faults can be from a continuous period of time or from multiple time periods. Furthermore, the historical faults of the electronic equipment can be faults from multiple electronic devices of the same type.
[0101] S402. Extract fault features from historical operation logs to obtain a fault analysis library.
[0102] These historical operation logs characterize the fault features of electronic equipment. Therefore, fault features can be extracted from these logs to obtain a fault analysis library, which can also be understood as a set of fault features. This allows the fault analysis library to include the fault features of all faults in electronic equipment as comprehensively as possible. Furthermore, historical faults can include multiple faults, and these faults may contain duplicate types, making the fault features of each type more detailed and comprehensive, thus more clearly characterizing the fault features of each type.
[0103] In some other embodiments, combined with Figure 3 ,like Figure 5 As shown, prior to S301 above, the method provided in this application embodiment further includes:
[0104] S501. Obtain the operation logs of different fault types of electronic devices at the time of occurrence and the operation logs within a preset time period before the fault occurs, as the initial data corresponding to different fault types.
[0105] Since the evolution trend set can be obtained in advance, it is possible to acquire operational logs for different types of historical faults at the time of their occurrence, and also acquire operational logs for a preset time period prior to each historical fault. Because most faults in electronic devices do not occur instantaneously but gradually during operation, these operational logs can characterize the changes in the operating state of the electronic device during the preset time period before the fault occurs. For example, the temperature may be within a predetermined range during normal operation, but gradually increase during the preset time period before the fault occurs. Furthermore, different fault types correspond to different changes in the operating state of the electronic device; therefore, these operational logs can be used to determine the evolution trend set.
[0106] S502. Perform parameter analysis on the initial data and select features with trend change characteristics as screening features.
[0107] The screening features corresponding to different fault types can be the same or different.
[0108] However, the initial data consists of the electronic device's operating logs, which cannot intuitively represent changes in the electronic device's operating status. Therefore, features with trend-changing characteristics can be filtered from the initial data. These features can then be used to characterize changes in the operating status, and the presence of faults can be determined based on these trends. One fault type can correspond to one or more filtering features, and different fault types may have overlapping filtering features.
[0109] S503. For the initial data corresponding to each fault type, sort the data of the screening features within a preset time period according to the time series to obtain the evolution trend of the screening features within the preset time period.
[0110] Among them, one type of fault corresponds to at least one screening feature, and one screening feature corresponds to an evolution trend.
[0111] The initial data includes data for each filtering feature within a preset time period. Therefore, the data for each filtering feature in the initial data can be sorted according to time series order to obtain the evolution trend of that filtering feature's data within the preset time period. In other words, the evolution trend (or change trend) of the filtering feature is represented by its corresponding value. Since the initial data consists of operational logs from when the fault occurred to before the fault occurred, the evolution trend of the filtering features obtained from the initial data can characterize the evolution trend of the filtering features corresponding to the electronic device from normal operation to the point of fault occurrence.
[0112] S504. For each fault type among different fault types, the evolution trend of all screening features corresponding to the fault type is determined as the evolution trend set of the fault type.
[0113] Each fault type corresponds to at least one screening feature. However, determining the evolution trend of a fault type based on the evolution trend of only one screening feature may be inaccurate. Therefore, the evolution trends of all screening features corresponding to each fault type can be determined as the evolution trend set of that fault type. This allows for the characterization of the changing trend of fault types through more features, thereby making fault warnings more accurate.
[0114] Combination Figure 5 ,like Figure 6As shown, in S303 above, the real-time features are clustered according to the fault characteristics corresponding to different fault types to obtain the feature set to be analyzed, which specifically includes:
[0115] S601. Determine the fault type of all faults in the fault analysis library.
[0116] S602. For each fault type among all fault types, the filtering features in the evolution trend set corresponding to the fault type are determined as the filtering feature set corresponding to the fault type.
[0117] S603. Remove duplicates from all the filter features in the filter feature set corresponding to all fault types to obtain the target filter features.
[0118] S604. Select target features from real-time features to complete the clustering of real-time features and obtain the feature set to be analyzed.
[0119] The feature set to be analyzed includes various features to be analyzed. The fault features corresponding to the fault types are the screening features corresponding to the fault types.
[0120] Since the fault analysis library includes faults of all fault types, all fault types can be identified through the library. Then, the filtering features in the evolution trend set corresponding to each fault type can be determined. These filtering features can then be considered as the fault features of the fault type corresponding to that evolution trend set. Therefore, the filtering features in the evolution trend set can be defined as the set of filtering features corresponding to the fault type of that evolution trend set, i.e., the fault feature set, which can be represented as a filtering feature table. In other words, the fault characteristics of each fault type can be determined using the fault analysis library and the filtering features in the evolution trend set.
[0121] Subsequently, since the screening features corresponding to different fault types may overlap, it is necessary to deduplicate all screening features in the screening feature set corresponding to all fault types to obtain the target screening features. In other words, the target screening features include the fault features of each fault type.
[0122] Subsequently, since there may be some features in the current real-time features that do not belong to the fault features, all target screening features in the real-time features can be screened and classified. That is, all fault features of all fault types in the real-time features can be screened and classified according to the fault type, so as to obtain the feature set to be analyzed. The feature set to be analyzed includes the fault features corresponding to each fault in multiple fault types.
[0123] In some other embodiments, combined with Figure 6 ,like Figure 7As shown, in S304 above, obtaining the trend set of the feature set to be analyzed specifically includes:
[0124] S701. For each target screening feature in the feature set to be analyzed, sort the data of the target screening feature within a preset time period according to the time series to obtain the changing trend of the target screening feature.
[0125] S702. Reorganize the changing trends of various target screening features in the feature set to be analyzed according to the screening feature set corresponding to different fault types to obtain the changing trend set of the feature set to be analyzed.
[0126] Since the feature set to be analyzed is obtained from multiple target screening features selected from the current fault features, for each target screening feature in the feature set to be analyzed, the data of the target screening feature within a preset time period (i.e., the data corresponding to the target screening feature in the real-time operation log) can be sorted according to the time series (i.e., time order) to obtain the changing trend of the target screening feature, that is, to obtain the changing trend of the fault corresponding to the target screening feature.
[0127] Since different fault types correspond to different sets of screening features, we can determine which target screening features in the feature set to be analyzed are corresponding to each fault type, and thus determine which fault type the target screening features are. For example, if we determine that the multiple target screening features include all the screening features corresponding to a certain fault type based on the set of screening features corresponding to each fault type, then we can determine the changing trend set of all the screening features corresponding to that fault type in the multiple target screening features as the changing trend set of the feature set to be analyzed.
[0128] In some other embodiments, comparing the similarity between the set of change trends and the set of evolution trends corresponding to different fault types specifically includes: for each set of evolution trends corresponding to each fault type, associating the change trend of each target screening feature in the set of change trends with the evolution trend of each screening feature in the set of evolution trends to calculate the similarity.
[0129] The calculation formula includes:
[0130]
[0131] Where, x i For a target, select the i-th observation of the changing trend of features, y i Let i be the i-th observation representing the evolution trend of a selected feature. The mean of all observations in the trend of change. Let be the mean of all observations of the trend, n be the number of observations, and r be the similarity between the trend and the evolution trend.
[0132] Since each fault type corresponds to an evolution trend set, and this set characterizes the changing trend of the electronic device's operating state before the fault, the similarity between the evolution trend of each selected feature in the evolution trend set and the evolution trend of each target selected feature in the change trend set can be calculated. This results in multiple similarity scores, the number of which is the number of selected features in the evolution trend set multiplied by the number of target selected features in the change trend set. In other words, the correlation between the evolution trend and the change trend is calculated one by one. Furthermore, the observed values are the data of the evolution trend being calculated or the observed data (i.e., the data or values corresponding to the real-time operation log).
[0133] It should be noted that when r > 0, the changing region of a target screening feature is similar to the changing trend of a screening feature. In other words, when the similarity is less than zero, it can be determined that the changing trend of the target screening feature is similar to the evolution trend of the screening feature.
[0134] Subsequently, if the change trend of each target screening feature in the change trend set is similar to the evolution trend of a single screening feature in the target evolution trend set, it is determined that the electronic device has a predicted fault, and the fault type corresponding to the target evolution trend set is the predicted fault type. In other words, if the evolution trends of multiple screening features in the target evolution trend set are similar one-to-one to the change trends of the target screening features in the feature set to be analyzed, then the feature set to be analyzed is similar to the target evolution trend set. Therefore, it can be determined that the electronic device has a predicted fault, and the fault type corresponding to the target evolution trend set is the predicted fault type corresponding to the feature set to be analyzed. The fault warning device can then issue an early warning, allowing staff to address the fault in advance and prevent its occurrence.
[0135] Conversely, if the similarity is less than or equal to 0, it can be determined that the two are not similar. If the feature set to be analyzed is not similar to all fault types, it can be determined that the current electronic device does not have a fault.
[0136] In this way, through the above scheme, the fault early warning device can monitor the status of electronic equipment in real time, and when the electronic equipment is about to fail, it will issue a fault warning and output the fault type. This allows staff to handle the fault before it occurs, thereby reducing the failure rate of the equipment and minimizing the losses caused by the fault.
[0137] Furthermore, if the fault warning device fails to successfully predict a fault before it occurs, it can also determine the type of fault after the fault has occurred. The following is a description of how the fault type is determined.
[0138] In some embodiments, combined with Figure 3 ,like Figure 8 As shown, the method provided in this application embodiment further includes:
[0139] S801. Obtain the fault operation log when the electronic device malfunctions.
[0140] S802. Extract fault features from the fault operation log to obtain the current fault features.
[0141] S803. Compare the current fault characteristics with the fault characteristics in the fault analysis database to determine the current fault type of the electronic equipment.
[0142] After an electronic device malfunctions, the fault operation log at the time of the malfunction can be obtained. Since the fault operation log can characterize the operating status of the electronic device, the fault type of the current malfunction can be determined through the fault operation log.
[0143] Since the fault operation log records the operation at the time of a fault, it can characterize the state of the electronic device when a fault occurs. Therefore, fault features can be extracted from the fault operation log to obtain the current fault features. Then, the current fault features can be compared with fault features in the fault analysis database. Specifically, the similarity between the current fault features and the fault features corresponding to each fault type in the fault analysis database is calculated to determine which fault type the current fault feature is similar to. The fault type corresponding to the similar fault feature is then identified as the current fault type. In this way, by comparing the fault operation log and the fault analysis database, the fault type can be quickly and accurately determined, thus ensuring the timeliness of fault handling.
[0144] It should be understood that between network-connected devices, a fault warning device can confirm the online status and responsiveness of electronic devices by sending "heartbeat" signals. If an electronic device stops responding to the heartbeat signal, it may indicate a fault, after which the fault warning device can request access to the electronic device's fault operation log. Alternatively, the electronic device can detect a fault through its built-in software, or, after a fault occurs, the fault warning device can retrieve the fault operation log based on the user's response.
[0145] In one possible implementation, S1003 specifically includes: calculating the similarity between the fault features of each fault type in the fault analysis library and the current fault features, to obtain multiple similarities;
[0146] The fault type corresponding to the smallest similarity among multiple similarities is determined as the current fault type of the electronic device; wherein the similarity is determined according to the following formula:
[0147]
[0148] In the formula, x j For the j-th current fault feature, y j Let be the j-th fault feature of the target fault type in the fault analysis library, m be the number of all fault features in the target fault type, d be the similarity, and the target fault type be any fault type.
[0149] Since there can be multiple current fault features, we can calculate pixels for each current fault feature and multiple fault features corresponding to the fault type in the fault analysis library, thereby obtaining the similarity values between multiple fault types and the current fault, and determining the fault type corresponding to the smallest similarity value as the fault type of the current fault.
[0150] It should be noted that when the current fault type of an electronic device differs from its actual fault type, the fault operation log, the operation log for a preset time period prior to the fault occurrence, and the actual fault type of the electronic device are stored in the fault analysis database based on the maintenance results. In other words, the fault analysis database can be updated based on the maintenance results of the electronic device, thereby improving the accuracy of fault prediction and fault type identification.
[0151] The foregoing mainly describes the solutions provided by the embodiments of this application from a methodological perspective. To achieve the above functions, it includes corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, based on the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.
[0152] This application embodiment can divide the fault early warning device into functional modules according to the above method example. For example, each function can be divided into a separate functional module, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware or as a software functional module. Optionally, the module division in this application embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.
[0153] like Figure 9 The diagram shown is a structural schematic of a fault warning device provided in an embodiment of this application. Figure 9The fault early warning device shown includes: an acquisition unit 901, an extraction unit 902, a clustering unit 903, a comparison unit 904, and a determination unit 905.
[0154] The acquisition unit 901 is used to acquire the real-time operation log of the electronic device.
[0155] Extraction unit 902 is used to extract features from real-time operation logs according to the fault analysis library to obtain real-time features.
[0156] Clustering unit 903 is used to cluster real-time features according to the fault characteristics corresponding to different fault types to obtain the feature set to be analyzed.
[0157] The acquisition unit 901 is also used to acquire the set of changing trends of the feature set to be analyzed.
[0158] The comparison unit 904 is used to compare the similarity between the set of change trends and the set of evolution trends corresponding to different fault types.
[0159] The determination unit 905 is used to determine whether the electronic device has a predicted fault and to determine the type of the predicted fault based on the comparison results.
[0160] In one possible implementation, the acquisition unit 901 is also used to acquire different types of historical faults of the electronic device and the historical operation logs corresponding to the historical faults.
[0161] Extraction unit 902 is also used to extract fault features from historical operation logs to obtain a fault analysis library.
[0162] In one possible implementation, the device further includes a sorting unit 906.
[0163] The acquisition unit 901 is also used to acquire the operation logs of different fault types of electronic devices when the faults occur and the operation logs within a preset time period before the faults occur, as the initial data corresponding to different fault types.
[0164] Unit 905 is also used to perform parameter analysis on the initial data and filter out features with trend change characteristics as screening features.
[0165] The sorting unit 906 is used to sort the data of the screening features within a preset time period according to the time series for the initial data corresponding to each fault type, so as to obtain the evolution trend of the screening features within the preset time period; one fault type corresponds to at least one screening feature, and one screening feature corresponds to one evolution trend.
[0166] The determining unit 905 is also used to determine the evolution trend of all the screening features corresponding to the fault type as the evolution trend set of the fault type for each fault type in different fault types.
[0167] In one possible implementation, clustering unit 903 is specifically used for:
[0168] Identify the fault type of all faults in the fault analysis library;
[0169] For each fault type among all fault types, the filtering features in the evolution trend set corresponding to the fault type are determined as the filtering feature set corresponding to the fault type.
[0170] Deduplication is performed on all the screening features in the screening feature set corresponding to all fault types to obtain the target screening features;
[0171] Target features are selected from real-time features to complete the clustering of real-time features and obtain the feature set to be analyzed.
[0172] In one possible implementation, the acquisition unit 901 is specifically used for:
[0173] For each target screening feature in the feature set to be analyzed, the data of the target screening feature within a preset time period are sorted according to the time series to obtain the changing trend of the target screening feature;
[0174] The changing trends of various target screening features in the feature set to be analyzed are reorganized according to the screening feature set corresponding to different fault types to obtain the changing trend set of the feature set to be analyzed.
[0175] In one possible implementation, comparison unit 904 is specifically used for:
[0176] For each fault type, the change trend of each target selected feature in the change trend set is correlated with the change trend of each selected feature in the evolution trend set to calculate the similarity. The calculation formula includes:
[0177]
[0178] Where, x i For a target, select the i-th observation of the changing trend of features, y i Let i be the i-th observation representing the evolution trend of a selected feature. The mean of all observations in the trend of change. Let be the mean of all observations of the trend, n be the number of observations, and r be the similarity between the trend and the evolution trend.
[0179] In one possible implementation, the determining unit 905 is specifically used for:
[0180] If the change trend of each target screening feature in the change trend set is similar to the evolution trend of a screening feature in the target evolution trend set, then the electronic device is determined to have a predicted fault and the fault type corresponding to the target evolution trend set is the predicted fault type.
[0181] In one possible implementation, the acquisition unit 901 is also used to acquire the fault operation log when the electronic device malfunctions.
[0182] The extraction unit 902 is also used to extract fault features from the fault operation log to obtain the current fault features;
[0183] The comparison unit 904 is also used to compare the current fault characteristics with the fault characteristics in the fault analysis library to determine the current fault type of the electronic equipment.
[0184] In one possible implementation, the determining unit 905 is specifically used for:
[0185] Calculate the similarity between the fault features of each fault type in the fault analysis library and the current fault features to obtain multiple similarity scores;
[0186] The fault type corresponding to the smallest similarity among multiple similarities is determined as the current fault type of the electronic device; wherein the similarity is determined according to the following formula:
[0187]
[0188] In the formula, x j For the j-th current fault feature, y j Let be the j-th fault feature of the target fault type in the fault analysis library, m be the number of all fault features in the target fault type, d be the similarity, and the target fault type be any fault type.
[0189] In one possible implementation, the device further includes a storage unit 907. The storage unit 907 is used to, when it is determined that the current fault type of the electronic device differs from the actual fault type of the electronic device, store the fault operation log, the operation log for a preset time period prior to the fault occurrence, and the actual fault type of the electronic device into a fault analysis database based on the maintenance results of the electronic device.
[0190] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0191] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0192] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0193] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0194] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the present invention.
Claims
1. A fault early warning method, characterized in that, include: Obtain real-time operating logs of electronic devices; The real-time operation logs are feature-extracted according to the fault analysis library to obtain real-time features; The real-time features are clustered according to the fault characteristics corresponding to different fault types to obtain the feature set to be analyzed. Obtain the trend set of the feature set to be analyzed, and compare the trend set of the feature set with the evolution trend set corresponding to different fault types. Based on the comparison results, determine that the electronic device has a predicted fault and determine the fault type of the predicted fault.
2. The method according to claim 1, characterized in that, The method further includes: Obtain the different types of historical faults of the electronic device and the corresponding historical operation logs of the historical faults; Fault features are extracted from the historical operation logs to obtain the fault analysis library.
3. The method according to claim 1, characterized in that, The method further includes: Obtain the operation logs of the electronic device at the time of the occurrence of different fault types and the operation logs within a preset time period before the occurrence of the fault, as the initial data corresponding to different fault types; The initial data is subjected to parameter analysis to select features with trend change characteristics as screening features; For each fault type, the initial data is sorted according to the time series of the filtering features within the preset time period to obtain the evolution trend of the filtering features within the preset time period; one fault type corresponds to at least one filtering feature, and one filtering feature corresponds to one evolution trend. For each fault type among different fault types, the evolution trend of all screening features corresponding to the fault type is determined as the evolution trend set of the fault type.
4. The method according to claim 3, characterized in that, The fault features corresponding to the fault type are the screening features corresponding to the fault type; The real-time features are clustered according to the fault characteristics corresponding to different fault types to obtain a feature set to be analyzed, including: Determine the fault type of all faults in the fault analysis library; For each fault type among all the fault types, the filtering features in the evolution trend set corresponding to the fault type are determined as the filtering feature set corresponding to the fault type. Deduplication is performed on all the screening features in the screening feature set corresponding to all fault types to obtain the target screening features; The target filtering features are selected from the real-time features to complete the clustering of the real-time features and obtain the feature set to be analyzed.
5. The method according to claim 4, characterized in that, The step of obtaining the set of changing trends of the feature set to be analyzed includes: For each target screening feature in the set of features to be analyzed, the data of the target screening feature within the preset time period are sorted according to the time series to obtain the changing trend of the target screening feature; The changing trends of various target screening features in the feature set to be analyzed are reorganized according to the screening feature set corresponding to different fault types to obtain the changing trend set of the feature set to be analyzed.
6. The method according to claim 5, characterized in that, The step of comparing the set of changing trends with the set of evolution trends corresponding to different fault types includes: For each fault type, the change trend of each target screening feature in the change trend set is correlated with the change trend of each screening feature in the change trend set to calculate the similarity. The calculation formula includes: Where, x i For a target, select the i-th observation of the changing trend of features, y i Let i be the i-th observation representing the evolution trend of a selected feature. The mean of all observations in the stated trend is... Let be the mean of all observations of the evolution trend, n be the number of observations, and r be the similarity between the change trend and the evolution trend.
7. The method according to claim 6, characterized in that, When r > 0, the changing trend of the target screening feature is similar to the evolution trend of the screening feature; The step of determining that the electronic device has a predicted fault based on the comparison results and determining the fault type of the predicted fault includes: If the change trend of each target screening feature in the change trend set is similar to the evolution trend of a screening feature in the target evolution trend set, it is determined that the electronic device has a predicted fault and the fault type corresponding to the target evolution trend set is the fault type of the predicted fault.
8. The method according to claim 2, characterized in that, The method further includes: Obtain the fault operation log when the electronic device malfunctions; The fault features are extracted from the fault operation log to obtain the current fault features; The current fault characteristics are compared with the fault characteristics in the fault analysis database to determine the current fault type of the electronic device.
9. The method according to claim 8, characterized in that, The step of comparing the current fault characteristics with fault characteristics in the fault analysis database to determine the current fault type of the electronic device includes: Calculate the similarity between the fault features of each fault type in the fault analysis library and the current fault features to obtain multiple similarity scores; The fault type corresponding to the smallest similarity among the plurality of similarities is determined as the current fault type of the electronic device; wherein the similarity is determined in a manner that satisfies the following formula: In the formula, x j For the j-th current fault feature, y j Let be the j-th fault feature of the target fault type in the fault analysis library, m be the number of all fault features in the target fault type, d be the similarity, and the target fault type be any fault type.
10. The method according to claim 9, characterized in that, The method further includes: If the current fault type of the electronic device is determined to be different from the actual fault type of the electronic device, the fault operation log, the operation log within a preset time period before the fault occurred, and the actual fault type of the electronic device are stored in the fault analysis database according to the maintenance results of the electronic device.
