Fault detection method, device, computer equipment and storage medium
By receiving smart device data and identifying target faults, the problems of timeliness and high cost of smart device fault detection are solved, efficient fault detection and automated processing are achieved, and the reliability and safety of the equipment are improved.
Patent Information
- Application Number
- CN202011332460.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2020-11-24
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2040-11-24
AI Technical Summary
Smart devices are prone to failures during use, but the existing technology is difficult to detect and resolve failures in a timely manner, resulting in low availability, stability and safety of the equipment, and regular inspections require a lot of manpower and material resources and high cost.
By receiving device data sent by smart devices, determining the target operation characteristics, and the corresponding relationship between the monitoring characteristics and fault and its fault probability constructed based on historical device data, automatically identify target faults, generate fault detection results, and improve the accuracy of fault detection. Sex and timely.
It improves the accuracy and timeliness of the fault detection of smart equipment, ensures the reliability, stability and safety of the equipment, and reduces the cost of manpower and material resources.
Smart Images

Figure CN113537268B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of Internet technology, and in particular to a fault detection method, device, computer equipment and storage medium. Background Art
[0002] With the development of Internet technology and science and technology, smart devices are becoming more and more popular. For example, smart devices that can recognize users' faces can be used to verify their identities. For another example, smart devices that can pay by swiping cards can swipe the amount in the user's payment card to complete the payment, etc. Therefore, smart devices have brought many conveniences to people's lives.
[0003] At present, it is inevitable that some faults will occur in the process of using smart devices. The way to know the faults of smart devices is often that the user discovers the faults during use and notifies the maintenance personnel to repair them, or the maintenance personnel regularly go to the location of the smart device to conduct regular inspections to determine whether the smart device has faults. Since it is necessary to manually discover faults and feedback, there is often a situation where feedback is not timely, which reduces the timeliness and convenience of fault discovery and resolution; and since smart devices are scattered all over the place, regular inspections at the location of the smart devices require a large amount of manpower and material costs, which reduces the timeliness and convenience of fault discovery and resolution, resulting in low availability, stability, reliability and security of smart devices. Summary of the invention
[0004] The embodiments of the present application provide a fault detection method, apparatus, computer equipment and storage medium, which can improve the accuracy and timeliness of fault detection and ensure the reliability, stability and security of the use of smart devices.
[0005] To solve the above technical problems, the present application provides the following technical solutions:
[0006] The present application provides a fault detection method, including:
[0007] Receiving device data sent by the smart device, wherein the device data includes operating characteristics;
[0008] determining a target operating characteristic from the device data according to a preset monitoring characteristic;
[0009] Obtain the corresponding relationship between different monitoring features and faults and their failure probabilities based on historical equipment data;
[0010] Determining a failure probability of at least one candidate failure corresponding to the target operating characteristic according to the corresponding relationship;
[0011] Determining a target fault of the smart device from at least one candidate fault based on the fault probability of the at least one candidate fault;
[0012] A fault detection result is generated according to the target fault.
[0013] According to one aspect of the present application, a fault detection device is also provided, comprising:
[0014] A receiving unit, configured to receive device data sent by a smart device, wherein the device data includes operation characteristics;
[0015] a feature determination unit, configured to determine a target operation feature from the device data according to a preset monitoring feature;
[0016] An acquisition unit, used for acquiring corresponding relationships between different monitoring features constructed based on historical equipment data and faults and their fault probabilities;
[0017] a probability determination unit, configured to determine a failure probability of at least one candidate failure corresponding to the target operation feature according to the corresponding relationship;
[0018] a fault determination unit, configured to determine a target fault of the smart device from at least one candidate fault based on the fault probability of the at least one candidate fault;
[0019] A generating unit is used to generate a fault detection result according to the target fault.
[0020] According to one aspect of the present application, a computer device is also provided, including a processor and a memory, wherein a computer program is stored in the memory, and when the processor calls the computer program in the memory, any fault detection method provided in the embodiments of the present application is executed.
[0021] According to one aspect of the present application, a storage medium is further provided, wherein the storage medium is used to store a computer program, and the computer program is loaded by a processor to execute any one of the fault detection methods provided in the embodiments of the present application.
[0022] The embodiment of the present application can receive device data sent by a smart device, the device data including operating characteristics; determine the target operating characteristics from the device data according to preset monitoring characteristics; obtain the corresponding relationship between different monitoring characteristics constructed based on historical device data and faults and their failure probabilities; determine the failure probability of at least one candidate fault corresponding to the target operating characteristics according to the corresponding relationship; determine the target fault of the smart device from at least one candidate fault based on the failure probability of at least one candidate fault; and generate a fault detection result according to the target fault. This solution can determine the failure probability of the candidate fault corresponding to the target operating characteristics in the device data of the smart device based on the corresponding relationship between the monitoring characteristics and the faults and their failure probabilities, automatically determine the target fault of the smart device based on the failure probability of the candidate faults, and generate a fault detection result, thereby improving the accuracy and timeliness of fault detection for smart devices and ensuring the reliability, stability and security of the use of smart devices. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.
[0024] Figure 1 It is a schematic diagram of a scenario in which the fault detection method provided in an embodiment of the present application is applied;
[0025] Figure 2 is a flowchart of a fault detection method provided in an embodiment of the present application;
[0026] Figure 3 is a schematic diagram of the corresponding relationship between monitoring features and faults and their fault probabilities provided in an embodiment of the present application;
[0027] Figure 4 It is a schematic diagram of the correspondence between network faults and monitoring features provided in an embodiment of the present application;
[0028] Figure 5 is a schematic diagram of a feature-based failure provided in an embodiment of the present application;
[0029] Figure 6 is a schematic diagram of calculating the probability of failure caused by a feature provided in an embodiment of the present application;
[0030] Figure 7 is a schematic diagram of determining a fault detection result based on a fault probability provided by an embodiment of the present application;
[0031] Figure 8is another flow chart of the fault detection method provided in an embodiment of the present application;
[0032] Fig. 9 It is a flowchart of establishing and applying a probability model provided in an embodiment of the present application;
[0033] Fig.10 It is a schematic diagram of a flow chart of updating a probability model provided by an embodiment of the present application;
[0034] Fig.11 is a schematic diagram of the corresponding relationship between features and faults based on human recognition scenarios provided by an embodiment of the present application;
[0035] Fig.12 is a schematic diagram of displaying fault detection results provided by an embodiment of the present application;
[0036] Fig.13 is another schematic diagram of displaying fault detection results provided by an embodiment of the present application;
[0037] Fig.14 is another schematic diagram of displaying fault detection results provided by an embodiment of the present application;
[0038] Fig.15 is another schematic diagram of displaying fault detection results provided by an embodiment of the present application;
[0039] Fig.16 is a schematic diagram of a fault detection device provided in an embodiment of the present application;
[0040] Fig.17 It is a schematic diagram of the structure of the server provided in the embodiment of the present application. DETAILED DESCRIPTION
[0041] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.
[0042] Embodiments of the present application provide a fault detection method, apparatus, computer equipment, and storage medium.
[0043] See also Figure 1 , Figure 1This is a schematic diagram of a scenario for the application of the fault detection method provided in an embodiment of the present application. The application of the fault detection method may include a fault detection device, which may be specifically integrated in a server 10. The server 10 may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers. It may also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms, but is not limited to these.
[0044] The server 10 and the smart device 20 may be directly or indirectly connected via wired or wireless communication, which is not limited in this application. The smart device 20 may be a mobile phone, a tablet computer, a laptop computer, a desktop computer, a wearable device, an access control, a gate machine in a station, or a card reader, etc.
[0045] In the process of operation, the smart device 20 can collect its own device data and send the collected device data to the server 10. The device data may include operation characteristics. The server 10 can be used to receive the device data sent by the smart device 20, determine the target operation characteristics from the device data according to the preset monitoring characteristics, and obtain the corresponding relationship between different monitoring characteristics and faults and their failure probabilities constructed based on historical device data, and determine the failure probability of at least one candidate fault corresponding to the target operation characteristics according to the corresponding relationship; then, based on the failure probability of at least one candidate fault, the target fault of the smart device can be determined from at least one candidate fault, and a fault detection result can be generated according to the target fault. The server 10 can also send the fault detection result to the smart device 20, and can also send the fault detection result to the terminal or user contact account (email, instant messaging account or mobile phone number, etc.) associated with the smart device 20 to prompt the user to handle the target fault of the smart device 20.
[0046] It should be noted that Figure 1 The scenario diagram of the fault detection method application shown is only an example. The fault detection method application and scenario described in the embodiment of the present application are intended to more clearly illustrate the technical solution of the embodiment of the present application, and do not constitute a limitation on the technical solution provided by the embodiment of the present application. Ordinary technicians in this field can know that with the evolution of the application of the fault detection method and the emergence of new business scenarios, the technical solution provided by the embodiment of the present application is also applicable to similar technical problems.
[0047] It should be noted that the description order of the following embodiments is not intended to limit the preferred order of the embodiments.
[0048] In this embodiment, the description will be made from the perspective of a fault detection device, which may be specifically integrated into a computer device such as a server.
[0049] See also Figure 2 , Figure 2 : is a flowchart of a fault detection method provided by an embodiment of the present application. The fault detection method may include:
[0050] S101. Receive device data sent by a smart device, where the device data includes operating characteristics.
[0051] Among them, the smart device can be an Internet of Things (IoT) smart device, that is, a hardware device that can connect to the Internet and provide smart services so as to communicate with the server. For example, the smart device can be a mobile phone, a computer, a wearable device, an access control system, or a card reader, etc. Of course, the type of smart device can also be flexibly set according to actual needs, and the specific type is not limited here.
[0052] The server can receive device data generated during the operation process sent by the smart device. The device data may include operation characteristics and other data. The operation characteristics may include business characteristics and technical characteristics. Among them, business characteristics may include transaction events and user operation events. Transaction events may include transactions completed through smart devices, for example, payment transactions completed through a card swiping machine, etc. User operation events may include events generated by user operations on smart devices, for example, users swiping cards on card swiping machines. Technical characteristics may include network connection status and device operation status. The network connection status may include connected to the network, disconnected from the network, the network bandwidth of the environment where the smart device is located meets the data transmission requirements, the network bandwidth of the environment where the smart device is located does not meet the data transmission requirements, etc. The device operation status may include the power-on status, the status of a certain function being available, and the status of a certain function being unavailable.
[0053] S102: Determine target operation characteristics from device data according to preset monitoring characteristics.
[0054] Among them, the preset monitoring features can be pre-set features that need to be monitored on smart devices, which are used to characterize the health status of smart devices. Since the indicators describing the health status of smart devices are different in different scenarios, one or more indicators (i.e., monitoring features) can be set according to specific business needs. For example, the monitoring features can include business features and technical features. Business features can include transaction events and user operation events, etc., and technical features can include network connection status and device operation status, etc.
[0055] The server can determine the target operating characteristics from the device data based on preset monitoring characteristics. For example, when the monitoring characteristics include the network connection status, characteristics related to the network connection status can be extracted from the device data as the target operating characteristics; for example, when the monitoring characteristics include transaction events, characteristics related to the transaction events can be extracted from the device data as the target operating characteristics; and so on.
[0056] In one embodiment, the monitoring characteristics include business characteristics and technical characteristics, the business characteristics include transaction events and user operation events, and the technical characteristics include network connection status and device operation status. Determining the target operation characteristics from the device data according to the preset monitoring characteristics may include: extracting transaction events and user operation events from the device data according to the business characteristics, and extracting network connection status and device operation status from the device data according to the technical characteristics to obtain the target operation characteristics.
[0057] In order to improve the accuracy and reliability of determining the target operation characteristics, the target operation characteristics corresponding to the smart device can be determined based on the monitoring characteristics. For example, the server can extract relevant characteristics such as transaction events and user operation events from the device data according to the business characteristics as the target operation characteristics, and extract relevant characteristics such as network connection status and device operation status from the device data according to the technical characteristics as the target operation characteristics, so that the smart device can be monitored accordingly based on the target operation characteristics.
[0058] S103: Obtain the corresponding relationship between different monitoring features constructed based on historical equipment data and faults and their fault probabilities.
[0059] When it is necessary to use the correspondence between different monitoring features and faults and their failure probabilities, the server can obtain the correspondence between different monitoring features and faults and their failure probabilities that are pre-stored based on the historical device data of the smart device from the database, or the server can obtain the pre-stored historical device data of the smart device from the database, perform statistical analysis on the historical faults of the smart device based on the historical device data, and obtain the correspondence between different monitoring features and faults and their failure probabilities.
[0060] Among them, a monitoring feature can correspond to one or more faults and the fault probability of the fault occurring, for example Figure 3 As shown, monitoring feature 1 can correspond to fault A, fault B, and fault C, etc., that is, based on monitoring feature 1, fault A, fault B, and fault C, etc. can occur, and the probability of fault A occurring based on monitoring feature 1 is P A1 , the failure probability of failure B based on monitoring feature 1 is P B1 , and the failure probability of failure C based on monitoring feature 1 is P C1 wait.
[0061] It should be noted that the monitoring feature is the data manifestation of the fault, and there is a causal relationship between the fault and the monitoring feature, for example Figure 4 As shown in the figure, a network failure of a smart device will cause the heartbeat quality of the smart device to deteriorate, or even drop to 0, and may also cause the success rate of the smart device executing business process events to be 0, etc., but there is no direct causal relationship in reverse. For example, if the heartbeat of a smart device drops to 0, it may not be a network failure, but may also be due to the user's active shutdown or other reasons. Therefore, there may be a one-to-one relationship or a many-to-many relationship between monitoring features and faults, and it is not possible to directly conclude what fault the smart device sends based on monitoring features.
[0062] In one embodiment, obtaining the correspondence between different monitoring features constructed based on historical equipment data and faults and their failure probabilities may include: obtaining historical equipment data corresponding to the smart device; performing fault statistical analysis on the historical equipment data based on the monitoring features to obtain the correspondence between different monitoring features and faults and their failure probabilities.
[0063] In order to improve the reliability of obtaining the corresponding relationship between different monitoring features and faults and their failure probabilities, the server can obtain the pre-stored historical device data corresponding to the smart device from a database (such as a Clickhouse database). For example, the database can store historical device data corresponding to multiple smart devices in association with the device identification of the smart device. At this time, the server can obtain the device identification of the smart device, obtain the historical device data corresponding to the device identification from the database, and obtain the historical device data of the smart device. Among them, the device identification can be used to uniquely identify the smart device. The device identification can be the physical address (Media Access Control Address, MAC), name or number of the smart device, and the device identification can be composed of numbers, letters, characters or text.
[0064] Then, the server can perform fault statistical analysis on the historical equipment data based on the monitoring features to obtain the corresponding relationship between different monitoring features and faults and their fault probabilities. The specific method of fault statistical analysis can be flexibly set according to actual needs.
[0065] In one embodiment, performing fault statistical analysis on historical equipment data based on monitoring features to obtain the corresponding relationship between different monitoring features and faults and their failure probabilities may include: calculating the number of reports of monitoring features based on historical equipment data, faults that occur based on monitoring features, and the number of occurrences of faults that occur based on monitoring features; and calculating the failure probability of a fault based on the number of reports and the number of occurrences.
[0066] In order to improve the convenience of fault statistical analysis of historical equipment data, the number of reports of monitoring features, the faults caused by monitoring features, and the number of occurrences of faults caused by monitoring features can be counted based on historical equipment data. When there are multiple monitoring features, the number of reports of each monitoring feature, the faults caused by each monitoring feature, and the number of occurrences of each fault caused by each monitoring feature can be counted separately. For example, Figure 5 As shown, for a smart device to execute a complete normal process, it is necessary to execute monitoring feature A (monitoring features can be referred to as features), monitoring feature B, monitoring feature C, monitoring feature D and monitoring feature E in sequence. When the smart device is running, it executes monitoring feature A, monitoring feature B and monitoring feature C in sequence, and then returns to execute monitoring feature B. This indicates that the smart device has retried and a failure has occurred based on monitoring feature C, that is, based on the features (i.e. monitoring features) included in the historical device data reported by the smart device, it is determined that the smart device exits the process based on the failure of monitoring feature C. Since the historical device data can include multiple process events executed by the smart device, if Figure 6 As shown, multiple process events can be superimposed (i.e., event streams can be superimposed), and the number of reports of each monitoring feature, such as monitoring feature A, monitoring feature B, monitoring feature C, monitoring feature D, and monitoring feature E, and the number of occurrences of each fault based on each monitoring feature can be counted. Then, the failure probability of the fault can be calculated based on the number of reports of the monitoring feature and the number of occurrences of the fault based on the monitoring feature. For example, the ratio between the number of occurrences and the number of reports can be used as the failure probability, such as Figure 6 In the example, feature C is reported 3 times, and the number of failures based on feature C is 1 time. Therefore, it can be calculated that the failure probability based on feature C is 1 / 3.
[0067] In one embodiment, obtaining the correspondence between different monitoring features constructed based on historical equipment data and faults and their failure probabilities may include: obtaining historical equipment data corresponding to the smart device, and obtaining multiple pre-set candidate monitoring features; performing fault statistical analysis on the historical equipment data based on each candidate monitoring feature to obtain a candidate correspondence between each candidate monitoring feature and the fault and its failure probability; clustering multiple candidate monitoring features based on the candidate correspondence to obtain monitoring features, and determining the failure probability of the fault corresponding to the monitoring feature based on the candidate correspondence to obtain the correspondence between different monitoring features and the fault and its failure probability.
[0068] Since the occurrence of the same fault may be caused by multiple monitoring features, and the multiple monitoring features may be linearly correlated, in order to improve the accuracy of the construction of the correspondence between the monitoring features and the faults and their fault probabilities, the linearly correlated monitoring features may be classified. For example, the server may obtain the historical device data corresponding to the pre-stored intelligent device from the database, and obtain multiple pre-set candidate monitoring features, and perform fault statistical analysis on the historical device data based on each candidate monitoring feature according to the above-mentioned fault statistical analysis method to obtain the candidate correspondence between each candidate monitoring feature and the fault and its fault probability. Then, based on the candidate correspondence between the candidate monitoring features and the faults and their fault probabilities, multiple candidate monitoring features may be clustered to obtain monitoring features. For example, different candidate monitoring features corresponding to the same or multiple faults may be classified into the same category of features, for example, based on candidate monitoring feature A, fault a, fault b, and fault c may occur, and the fault probability of fault a based on candidate monitoring feature A is x1, the fault probability of fault b based on candidate monitoring feature A is y1, and the fault probability of fault c based on candidate monitoring feature A is z1; based on candidate monitoring feature B, fault a, fault b, and fault c may occur, and based on candidate monitoring feature B, fault The failure probability of candidate monitoring feature a is x1, the failure probability of candidate monitoring feature b is y1, and the failure probability of candidate monitoring feature c is z1; based on candidate monitoring feature C, failure a, failure d and failure e will occur, and the failure probability of candidate monitoring feature a is x2, the failure probability of candidate monitoring feature b is y2, and the failure probability of candidate monitoring feature c is z2. At this time, since candidate monitoring feature A and candidate monitoring feature B have the same failure and the same failure probability, candidate monitoring feature A and candidate monitoring feature B can be classified as the same type of monitoring feature. And the failure probability of the failure corresponding to the monitoring feature can be determined based on the candidate corresponding relationship to obtain the corresponding relationship between different monitoring features and failures and their failure probabilities. For example, the corresponding relationship between candidate monitoring feature A and the failure probability x1 of failure a, the failure probability y1 of failure b and the failure probability z1 of failure c can be obtained.
[0069] It should be noted that the server can also establish a heartbeat event for the communication of the smart device based on historical device data statistics, for example, according to a preset time period, the number of heartbeats of the smart device is counted, and the heartbeat quality score of the smart device is determined based on the number of heartbeats of the smart device, so as to judge whether the heartbeat of the smart device is normal based on the heartbeat quality score. If the heartbeat is insufficient, it means that the network of the smart device is abnormal; if the heartbeat is 0, it means that the smart device is not turned on or the smart device cannot communicate with the server, etc. The server can also count the time consumed by the network connection request of the smart device based on historical device data, and determine the average network speed according to the average time consumed by the network connection request of the smart device in the preset time period, so as to judge whether the network bandwidth of the environment where the smart device is located meets the data transmission needs based on the average network speed. If the average network speed is greater than the preset threshold, it means that the network bandwidth of the environment where the smart device is located meets the data transmission needs. If the average network speed is less than or equal to the preset threshold, it means that the network bandwidth of the environment where the smart device is located cannot meet the data transmission needs. The server can also count the number of normal process events completed by the smart device based on historical device data, and determine the success rate of the execution process event based on the number of normal process events completed, so as to judge whether the human-computer interaction of the smart device is normal based on the success rate. Of course, other information of smart devices based on historical device data statistics can also be collected according to actual needs, and the specific content is not limited here.
[0070] In one embodiment, the fault detection method may further include: detecting whether the pre-stored historical equipment data has changed according to an update cycle; if the historical equipment data has changed, performing fault statistical analysis on the changed historical equipment data based on the monitoring characteristics to update the correspondence between different monitoring characteristics and faults and their failure probabilities.
[0071] In order to improve the timeliness of updating the correspondence between monitoring features and faults and their failure probabilities, the server can detect whether the pre-stored historical device data has changed according to the update cycle. The update cycle can be flexibly set according to actual needs. For example, the update cycle can be every hour or every day. If the historical device data changes, the fault statistical analysis is performed on the changed historical device data based on the monitoring features according to the above-mentioned fault statistical analysis method to update the correspondence between different monitoring features and faults and their failure probabilities. For example, when the smart device reports the device data, the server can store the reported device data in the database. At this time, the stored historical device data is added, and the correspondence between different monitoring features and faults and their failure probabilities can be updated based on the changed historical device data.
[0072] In one embodiment, the fault detection method may further include: detecting whether the monitoring characteristics have changed according to an update cycle; if the monitoring characteristics have changed, performing fault statistical analysis on historical equipment data based on the changed monitoring characteristics to update the correspondence between different monitoring characteristics and faults and their failure probabilities.
[0073] In order to improve the flexibility of updating the correspondence between monitoring features and faults and their fault probabilities, the server can detect whether the monitoring features have changed according to the update cycle. The update cycle can be flexibly set according to actual needs, for example, the update cycle can be every hour or every day. If the monitoring features have changed, the historical equipment data is subjected to fault statistical analysis based on the changed monitoring features according to the above-mentioned fault statistical analysis method to update the correspondence between different monitoring features and faults and their fault probabilities. For example, if a monitoring feature is added, the server can update the correspondence between different monitoring features and faults and their fault probabilities based on the added monitoring features and historical equipment data.
[0074] S104. Determine a failure probability of at least one candidate failure corresponding to the target operating characteristic according to the corresponding relationship.
[0075] For example, after obtaining the correspondence between different monitoring features and faults and their failure probabilities, the correspondence between different monitoring features and faults and their failure probabilities can be queried to find the monitoring features corresponding to the target operating features, and one or more faults corresponding to the monitoring features can be used as one or more candidate faults corresponding to the target operating features, and the failure probabilities of one or more faults corresponding to the monitoring features can be used as the failure probabilities of one or more candidate faults corresponding to the target operating features, so as to determine the occurrence of candidate faults based on the target operating features and the failure probability of the candidate faults.
[0076] For example, Figure 3 As shown in the figure, when the target operation feature is monitoring feature 3, fault A, fault B and fault C corresponding to monitoring feature 3 can be set as candidate faults corresponding to the target operation feature, that is, the candidate faults include fault A, fault B and fault C, and the fault probability P of fault A occurring based on monitoring feature 3 is A3 , the failure probability P of failure B based on monitoring feature 3 B3 and the failure probability P of failure C based on monitoring feature 3 C3 , are respectively set as the failure probability P of failure A occurring based on the target operation characteristics A3 , the failure probability P of failure B based on the target operation characteristics B3 and the failure probability P of failure C based on the target operation characteristics C3 .
[0077] S105 . Determine a target fault of the smart device from the at least one candidate fault based on the fault probability of the at least one candidate fault.
[0078] In one embodiment, the candidate faults include multiple ones, and based on the failure probability of at least one candidate fault, determining the target failure of the smart device from at least one candidate fault may include: from the failure probabilities of the multiple candidate faults, selecting the candidate fault corresponding to the maximum failure probability as the target failure of the smart device.
[0079] Among them, the candidate faults and their corresponding fault probabilities may include one or more. When the candidate fault and its corresponding fault probability are one, the candidate fault can be directly set as the target fault of the smart device, and the fault probability of the candidate fault can be set as the fault probability of the target fault. When the candidate faults and their corresponding fault probabilities are multiple, the candidate fault corresponding to the maximum fault probability can be selected from the fault probabilities of multiple candidate faults as the target fault of the smart device, and the fault probability of the candidate fault can be set as the fault probability of the target fault. In this way, the fault of the smart device can be quickly determined for timely processing.
[0080] S106. Generate a fault detection result according to the target fault.
[0081] Among them, the style and specific content of the fault detection results can be flexibly set according to actual needs. The health status of the smart device can be determined through the fault detection results, and the fault probability speculation method can be used to assist in evaluating the health status of the smart device, timely and proactively discover faulty smart devices and repair them in time to improve the service quality of the smart device.
[0082] In one embodiment, generating a fault detection result based on a target fault may include: when the failure probability of the target fault is a first threshold, the generated fault detection result is to determine that the smart device has not failed; when the failure probability of the target fault is a second threshold, the generated fault detection result is to determine that the failure occurring in the smart device is the target fault, and the second threshold is greater than the first threshold; when the failure probability of the target fault is within the range of the first threshold to the second threshold, the generated fault detection result is to predict that the smart device will have a target fault, and the possible probability of occurrence is the failure probability of the target fault.
[0083] The first threshold and the second threshold can be flexibly set according to actual needs. For example, the first threshold can be set to 0 and the second threshold can be set to 1. Figure 7As shown, when the failure probability P of the target failure is 0, the fault detection result that can be generated is to determine that the smart device has not failed, that is, the smart device is healthy; when the failure probability of the target failure is 1, the fault detection result that can be generated is to determine that the smart device has the target failure. When the failure probability of the target failure is in the range of 0 to 1, the fault detection result that can be generated is to predict that the smart device may have the target failure, and the possible probability of occurrence is the failure probability corresponding to the target failure, that is, it cannot be determined that the smart device will definitely have the target failure. The greater the failure probability, the higher the possibility of the target failure. At this time, the change in the failure probability of the target failure can be continuously monitored to determine whether the smart device will definitely have the target failure, or the failure of the smart device can be determined through manual troubleshooting, etc. For example, the failure of the smart device can be determined through fault logs or on-site detection.
[0084] In one embodiment, after generating a fault detection result according to the target fault, the fault detection method may further include: sending the fault detection result to a terminal or a user contact account associated with the smart device to prompt the user to handle the target fault.
[0085] After obtaining the fault detection results, the server can send the fault detection results to a terminal associated with the smart device. For example, the terminal can be a computer or mobile phone used to display the fault detection results of the smart device. The server can also send the fault detection results to a user contact account associated with the smart device. The user contact account can include an email address, instant messaging account or mobile phone number, etc., to prompt the user to promptly handle the target faults of the smart device and improve the business service quality of the smart device.
[0086] It should be noted that, when the network connection between the smart device and the server is normal, if other faults (non-network faults) occur in the smart device, the server can also send the fault detection results to the smart device, so that the smart device can output the fault detection results through a display screen or voice broadcast, etc., so as to remind the user who is currently using the smart device, so that the user can know what kind of fault has occurred in the smart device, so as to stop using it and avoid using it when a fault occurs, which will affect the service life and safety of use of the smart device.
[0087] The embodiment of the present application can receive device data sent by a smart device, the device data including operating characteristics; determine the target operating characteristics from the device data according to preset monitoring characteristics; obtain the corresponding relationship between different monitoring characteristics constructed based on historical device data and faults and their failure probabilities; determine the failure probability of at least one candidate fault corresponding to the target operating characteristics according to the corresponding relationship; determine the target fault of the smart device from at least one candidate fault based on the failure probability of at least one candidate fault; and generate a fault detection result according to the target fault. This solution can determine the failure probability of the candidate fault corresponding to the target operating characteristics in the device data of the smart device based on the corresponding relationship between the monitoring characteristics and the faults and their failure probabilities, automatically determine the target fault of the smart device based on the failure probability of the candidate faults, and generate a fault detection result, thereby improving the accuracy and timeliness of fault detection for smart devices and ensuring the reliability, stability and security of the use of smart devices.
[0088] The method described in the above embodiment is further described in detail below with examples.
[0089] This embodiment takes the fault detection device integrated in the server as an example, and takes the server's fault detection of the face recognition payment application scenario of the smart device as an example. Figure 8 , Figure 8 A schematic diagram of a flow chart of a fault detection method provided in an embodiment of the present application. The method flow may include:
[0090] S201. Obtain historical device data of facial recognition payment by smart devices.
[0091] During the process of facial recognition payment, the smart device can report the collected device data to the server. The server can associate the received device data with the device identification of the smart device and store it in a database. At this time, the device data stored in the database is used as historical device data.
[0092] S202: Perform fault statistical analysis on historical equipment data based on pre-set monitoring features to establish a probability model of the corresponding relationship between different monitoring features and faults and their failure probabilities.
[0093] For example, Fig. 9 As shown, the server performs fault statistics analysis on historical equipment data based on pre-set monitoring features. For example, the number of reports of monitoring features, faults that occur based on monitoring features, and the number of occurrences of faults that occur based on monitoring features can be counted based on historical equipment data. The fault probability of the fault is calculated based on the number of reports and the number of occurrences, so as to establish a probability model of the corresponding relationship between different monitoring features and faults and their fault probabilities. For example, Figure 3 As shown, the monitoring feature 1 and the fault A and its fault probability P can be established. A1, fault B and its failure probability P B1 and fault C and its failure probability P C1 The corresponding relationship between etc. is the probability model.
[0094] It should be noted that if Fig.10 As shown, the probability model of the corresponding relationship between the monitoring features and the faults and their failure probabilities can be updated in time according to the changes in the historical device data of the smart device for face recognition payment, or according to the changes in the monitoring features.
[0095] For example, Fig.11 As shown, taking the application scenario of face recognition payment as an example, the established probability model can include the correspondence between monitoring features such as not scanning the face for N consecutive days (the value of N can be flexibly set according to actual needs) and not being connected to the Internet (i.e., connected to the Internet) for N consecutive days and failures such as merchant abandonment, merchant network problems, and hardware failures and their failure probabilities.
[0096] S203: Receive device data corresponding to facial recognition payment sent by the smart device, where the device data includes operating characteristics.
[0097] The server can receive device data generated during the face recognition payment process sent by the smart device, which may include operating characteristics and other data.
[0098] S204: Determine target operation characteristics from the equipment data according to the target monitoring characteristics.
[0099] The target monitoring feature may be a pre-set feature that needs to be monitored on the smart device, and the server may determine the target operation feature from the device data according to the target monitoring feature. For example, when the target monitoring feature includes the network connection status, features related to the network connection status may be extracted from the device data as the target operation feature.
[0100] S205. Determine the failure probabilities of multiple candidate failures corresponding to the target operation characteristics according to the probability model.
[0101] After the probability model is obtained, the failure probabilities of multiple candidate failures corresponding to the target operation characteristics can be determined according to the probability model, for example, Figure 3 As shown in the figure, when the target operation feature is monitoring feature 3, fault A, fault B and fault C corresponding to monitoring feature 3 can be set as candidate faults corresponding to the target operation feature, that is, the candidate faults include fault A, fault B and fault C, and the fault probability P of fault A occurring based on monitoring feature 3 is A3 , the failure probability P of failure B based on monitoring feature 3 B3 and the failure probability P of failure C based on monitoring feature 3 C3, are respectively set as the failure probability P of failure A occurring based on the target operation characteristics A3 , the failure probability P of failure B based on the target operation characteristics B3 and the failure probability P of failure C based on the target operation characteristics C3 .
[0102] For example, Fig.11 As shown, when the target operation feature is not scanning the face for N consecutive days, the faults such as merchant abandonment, merchant network problems, and hardware failure corresponding to not scanning the face for N consecutive days can be set as candidate faults corresponding to the target operation feature, and the failure probability of merchant abandonment, merchant network problems, and hardware failure can be set to the failure probability of each fault corresponding to not scanning the face for N consecutive days.
[0103] S206 . Select, from the failure probabilities of the multiple candidate failures, a candidate failure corresponding to the maximum failure probability as a target failure of the smart device.
[0104] When there are multiple candidate faults and their corresponding fault probabilities, the candidate fault corresponding to the maximum fault probability can be selected from the fault probabilities of the multiple candidate faults as the target fault of the smart device, and the failure probability of the candidate fault can be set as the failure probability of the target fault.
[0105] It should be noted that the fault detection of smart devices can be divided into two categories: white-box speculation and black-box speculation. White-box speculation can be to clearly know the fault of the smart device, while black-box speculation can be to speculate the possible fault through data representation when the fault of the smart device cannot be clearly known. White-box speculation is used for the target operating characteristics in the reported device data that have a clear relationship with the fault, so that the occurrence of the corresponding fault can be inferred relatively accurately, for example, Figure 7 As shown, white box speculation can include: clearly knowing the fault correspondence of the smart device, that is, the fault probability P = 1, clearly knowing that the smart device is normal and healthy, that is, the fault probability P = 0, when the fault probability P is 0 and 1, it can be judged whether the smart device is healthy or faulty. If the white box speculation data is not enough to make the fault probability P 0 or 1, the probability model can be used to speculate the possible faults of the smart device. For example, in the face recognition payment (which can be called face payment) scenario: the event that the face is too low relative to the acquisition camera indicates that the recognized face is lower, resulting in the failure of face recognition. This is a clear error and can correspond to a clear fault solution.
[0106] If the target operating characteristics in the reported device data have no direct and clear relationship with the fault, black box speculation can be used. The operating characteristics of the corresponding device data will be sensitive to the fault of the smart device and can play a role in detecting the fault. For example, in the face payment scenario, the success rate of face recognition payment will be affected by faults caused by the network, algorithm, device hardware (camera and connection cables, etc.), and device software (APP logic). Therefore, the fault is speculated by monitoring the success rate of face recognition payment. Based on the comprehensive iteration of white box speculation and black box speculation, the accuracy of judging the health status of smart devices can be improved.
[0107] S207: Generate a fault detection result according to the target fault, and send the fault detection result to a terminal or a user contact account associated with the smart device.
[0108] For example, when the failure probability P of the target failure is 0, the fault detection result that can be generated is to determine that the smart device has not failed; when the failure probability of the target failure is 1, the fault detection result that can be generated is to determine that the smart device has the target failure. When the failure probability of the target failure is in the range of 0 to 1, the fault detection result that can be generated is to predict that the smart device may have the target failure, and the possible probability of occurrence is the failure probability corresponding to the target failure, that is, it cannot be determined that the smart device will definitely have the target failure. The greater the failure probability, the higher the possibility of the target failure. At this time, the change in the failure probability of the target failure can be continuously monitored to determine whether the smart device will definitely have the target failure, or the failure of the smart device can be determined through manual troubleshooting.
[0109] After obtaining the fault detection results, the server can send the fault detection results to a terminal associated with the smart device. For example, the terminal can be a computer or mobile phone used to display the fault detection results of the smart device. The server can also send the fault detection results to a user contact account associated with the smart device. The user contact account can include an email address, instant messaging account or mobile phone number, etc., to prompt the user to promptly handle the target faults of the smart device and improve the business service quality of the smart device.
[0110] For example, Fig.12 As shown, the fault detection results of face recognition payment by smart devices can be displayed through the terminal. Fig.12 The face scanning status of the smart device within the preset time period can be displayed in the form of a change curve, including the number of face scanning starts, the number of face scanning payments, and the number of recognition passes. When the corresponding change curves of the number of face scanning payments and the number of recognition passes return to zero, it means that the smart device has experienced a face scanning payment or face recognition payment related failure. Fig.13As shown, the conversion rate (i.e., success rate) corresponding to each operating feature of the smart device during the face recognition payment process can be displayed in the form of a funnel chart, such as Fig.14 As shown in FIG, when the conversion rate corresponding to each operating feature of the smart device during the face recognition payment process is low, it indicates that a fault related to each operating feature has occurred in the smart device. Fig.15 As shown, the faults corresponding to various operating characteristics of the smart device during the face recognition payment process can be displayed in the form of a pie chart, wherein the camera stream start can refer to turning on the video stream of the camera. After the camera is turned on, the video stream can be turned on first, and then the image with the face can be captured, etc.
[0111] In the above embodiments, the description of each embodiment has its own emphasis. For the part that is not described in detail in a certain embodiment, please refer to the detailed description of the fault detection method above, and will not be repeated here.
[0112] The server in the embodiment of the present application can establish a probability model of the corresponding relationship between different monitoring features and faults and their failure probabilities based on the historical device data of facial recognition payment of the smart device, and receive the device data corresponding to the facial recognition payment sent by the smart device, determine the target operation features from the device data based on the target monitoring features, and determine the failure probabilities of multiple candidate faults corresponding to the target operation features based on the probability model, select the candidate fault corresponding to the maximum failure probability from the failure probabilities of multiple candidate faults as the target fault of the smart device, and generate a fault detection result. At this time, the fault detection result can be sent to the terminal or user contact account associated with the smart device to remind the user to deal with it in time, thereby improving the accuracy and timeliness of fault detection for smart devices and ensuring the reliability, stability and security of the use of smart devices.
[0113] In order to better implement the fault detection method provided in the embodiment of the present application, the embodiment of the present application also provides a device based on the above fault detection method. The meanings of the terms are the same as those in the above fault detection method, and the specific implementation details can refer to the description in the method embodiment.
[0114] See also Fig.16 , Fig.16 A schematic diagram of the structure of a fault detection device provided in an embodiment of the present application, wherein the fault detection device may include a receiving unit 301, a feature determination unit 302, an acquisition unit 303, a probability determination unit 304, a fault determination unit 305, and a generation unit 306, etc.
[0115] The receiving unit 301 is used to receive device data sent by the smart device, and the device data includes operation characteristics.
[0116] The feature determination unit 302 is used to determine the target operation feature from the device data according to the preset monitoring feature.
[0117] The acquisition unit 303 is used to acquire the corresponding relationship between different monitoring features constructed based on historical equipment data and faults and their fault probabilities.
[0118] The probability determination unit 304 is used to determine the failure probability of at least one candidate failure corresponding to the target operation feature according to the corresponding relationship.
[0119] The fault determination unit 305 is configured to determine a target fault of the smart device from at least one candidate fault based on the fault probability of at least one candidate fault.
[0120] The generating unit 306 is configured to generate a fault detection result according to the target fault.
[0121] In one implementation, the acquiring unit 303 may include:
[0122] An acquisition subunit is used to acquire historical device data corresponding to the smart device;
[0123] The analysis subunit is used to perform fault statistical analysis on historical equipment data based on monitoring features, and obtain the corresponding relationship between different monitoring features and faults and their fault probabilities.
[0124] In one embodiment, the analysis subunit can be specifically used to: count the number of reports of monitoring features, faults occurring based on monitoring features, and the number of occurrences of faults occurring based on monitoring features based on historical equipment data; and calculate the failure probability of the fault based on the number of reports and the number of occurrences.
[0125] In one embodiment, the acquisition unit 303 can be specifically used to: obtain historical device data corresponding to the smart device, and obtain multiple pre-set candidate monitoring features; perform fault statistical analysis on the historical device data based on each candidate monitoring feature to obtain a candidate correspondence between each candidate monitoring feature and the fault and its failure probability; cluster multiple candidate monitoring features based on the candidate correspondence to obtain monitoring features, and determine the failure probability of the fault corresponding to the monitoring feature based on the candidate correspondence to obtain the correspondence between different monitoring features and the fault and its failure probability.
[0126] In one embodiment, the fault detection device may further include:
[0127] The first updating unit is used to detect whether the pre-stored historical equipment data has changed according to the updating cycle; if the historical equipment data has changed, a fault statistical analysis is performed on the changed historical equipment data based on the monitoring characteristics to update the corresponding relationship between different monitoring characteristics and faults and their fault probabilities.
[0128] In one embodiment, the fault detection device may further include:
[0129] The second updating unit is used to detect whether the monitoring characteristics have changed according to the updating cycle; if the monitoring characteristics have changed, the historical equipment data is subjected to fault statistical analysis based on the changed monitoring characteristics to update the corresponding relationship between different monitoring characteristics and faults and their fault probabilities.
[0130] In one implementation, the candidate faults include multiple ones, and the fault determination unit 305 may be specifically configured to: select a candidate fault corresponding to the maximum fault probability from the fault probabilities of the multiple candidate faults as the target fault of the smart device.
[0131] In one embodiment, the monitoring characteristics include business characteristics and technical characteristics. The business characteristics include transaction events and user operation events. The technical characteristics include network connection status and device operation status. The characteristic determination unit 302 can be specifically used to: extract transaction events and user operation events from device data according to business characteristics, and extract network connection status and device operation status from device data according to technical characteristics to obtain target operation characteristics.
[0132] In one embodiment, the generation unit 306 can be specifically used for: when the failure probability of the target fault is a first threshold, the generated fault detection result is to determine that the smart device has not failed; when the failure probability of the target fault is a second threshold, the generated fault detection result is to determine that the fault occurring in the smart device is the target fault, and the second threshold is greater than the first threshold; when the failure probability of the target fault is within the range from the first threshold to the second threshold, the generated fault detection result is to predict that the target fault occurs in the smart device, and the possible probability of occurrence is the failure probability of the target fault.
[0133] In one embodiment, the fault detection device may further include:
[0134] The sending unit is used to send the fault detection result to the terminal or user contact account associated with the smart device to prompt the user to handle the target fault.
[0135] In the embodiment of the present application, the receiving unit 301 receives the device data sent by the smart device, and the device data includes operation characteristics; the characteristic determination unit 302 determines the target operation characteristics from the device data according to the preset monitoring characteristics; the acquisition unit 303 obtains the corresponding relationship between different monitoring characteristics constructed based on historical device data and faults and their failure probabilities; the probability determination unit 304 determines the failure probability of at least one candidate fault corresponding to the target operation characteristics according to the corresponding relationship; the fault determination unit 305 determines the target fault of the smart device from at least one candidate fault based on the failure probability of at least one candidate fault; and the generation unit 306 generates a fault detection result according to the target fault. This solution can determine the failure probability of the candidate fault corresponding to the target operation characteristics in the device data of the smart device based on the corresponding relationship between the monitoring characteristics and the faults and their failure probabilities, automatically determine the target fault of the smart device based on the failure probability of the candidate faults, and generate a fault detection result, thereby improving the accuracy and timeliness of fault detection for smart devices and ensuring the reliability, stability and security of the use of smart devices.
[0136] The embodiment of the present application also provides a computer device, the server may be a server such as Fig.17 As shown, it shows a schematic diagram of the structure of the server involved in the embodiment of the present application, specifically:
[0137] The server may include one or more processing core processors 401, one or more computer-readable storage media memories 402, a power supply 403, an input unit 404 and other components. Those skilled in the art will appreciate that Fig.17 The server structure shown in the figure does not constitute a limitation on the server, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.
[0138] in:
[0139] The processor 401 is the control center of the server, and uses various interfaces and lines to connect various parts of the entire server. By running or executing software programs and / or modules stored in the memory 402, and calling data stored in the memory 402, the processor 401 performs various functions of the server and processes data, thereby monitoring the server as a whole. Optionally, the processor 401 may include one or more processing cores; preferably, the processor 401 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application programs, and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 401.
[0140] The memory 402 can be used to store software programs and modules. The processor 401 executes various functional applications and data processing by running the software programs and modules stored in the memory 402. The memory 402 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created according to the use of the server, etc. In addition, the memory 402 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. Accordingly, the memory 402 may also include a memory controller to provide the processor 401 with access to the memory 402.
[0141] The server also includes a power supply 403 for supplying power to each component. Preferably, the power supply 403 can be logically connected to the processor 401 through a power management system, so that the power management system can manage charging, discharging, power consumption and other functions. The power supply 403 can also include one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, power status indicators and other arbitrary components.
[0142] The server may further include an input unit 404, which may be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control.
[0143] Although not shown, the server may further include a display unit, etc., which will not be described in detail herein. Specifically in this embodiment, the processor 401 in the server will load the executable files corresponding to the processes of one or more application programs into the memory 402 according to the following instructions, and the processor 401 will run the application programs stored in the memory 402, thereby realizing various functions, as follows:
[0144] Receive device data sent by an intelligent device, the device data including operating characteristics; determine target operating characteristics from the device data according to preset monitoring characteristics; obtain the corresponding relationship between different monitoring characteristics constructed based on historical device data and faults and their failure probabilities; determine the failure probability of at least one candidate fault corresponding to the target operating characteristics according to the corresponding relationship; determine a target fault of the intelligent device from at least one candidate fault based on the failure probability of at least one candidate fault; and generate a fault detection result according to the target fault.
[0145] In the above embodiments, the description of each embodiment has its own emphasis. For the part that is not described in detail in a certain embodiment, please refer to the detailed description of the fault detection method above, and will not be repeated here.
[0146] According to one aspect of the present application, a computer program product or a computer program is provided, the computer program product or the computer program including computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the methods provided in various optional implementations in the above-mentioned embodiments.
[0147] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments can be completed by computer instructions, or by controlling related hardware through computer instructions, and the computer instructions can be stored in a computer-readable storage medium and loaded and executed by a processor. To this end, an embodiment of the present application provides a storage medium, in which a computer program is stored, and the computer program may include computer instructions, and the computer program can be loaded by a processor to execute any of the fault detection methods provided in the embodiments of the present application.
[0148] The specific implementation of the above operations can be found in the previous embodiments, which will not be described in detail here.
[0149] The storage medium may include: a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.
[0150] Since the instructions stored in the storage medium can execute the steps in any fault detection method provided in the embodiments of the present application, the beneficial effects that can be achieved by any fault detection method provided in the embodiments of the present application can be achieved. Please refer to the previous embodiments for details and will not be repeated here.
[0151] The above is a detailed introduction to a fault detection method, device, computer equipment and storage medium provided in the embodiments of the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for technical personnel in this field, according to the ideas of the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.
Claims
1. A fault detection method, It is characterized in that include: Receiving device data sent by the smart device, wherein the device data includes operating characteristics; determining a target operating characteristic from the device data according to a preset monitoring characteristic; Obtaining the corresponding relationship between different monitoring features constructed based on historical device data and faults and their fault probabilities, including: obtaining the historical device data corresponding to the smart device; counting the number of reports of the monitoring features, faults occurring based on the monitoring features, and the number of occurrences of the faults occurring based on the monitoring features based on the historical device data; calculating the fault probability of the fault according to the number of reports and the number of occurrences, and obtaining the corresponding relationship between different monitoring features and faults and their fault probabilities; Determining a failure probability of at least one candidate failure corresponding to the target operating characteristic according to the corresponding relationship; Determining a target fault of the smart device from at least one candidate fault based on the fault probability of the at least one candidate fault; A fault detection result is generated according to the target fault.
2. The fault detection method according to claim 1, It is characterized in that The obtaining of the corresponding relationship between different monitoring features constructed based on historical equipment data and faults and their fault probabilities also includes: Acquire historical device data corresponding to the smart device, and acquire a plurality of preset candidate monitoring features; Performing a fault statistical analysis on the historical equipment data based on each candidate monitoring feature to obtain a candidate corresponding relationship between each candidate monitoring feature and the fault and its fault probability; The plurality of candidate monitoring features are clustered based on the candidate corresponding relationships to obtain the monitoring features, and the failure probabilities of the failures corresponding to the monitoring features are determined based on the candidate corresponding relationships to obtain the corresponding relationships between different monitoring features and the failures and their failure probabilities.
3. The fault detection method according to claim 1, It is characterized in that The fault detection method further comprises: Detecting whether the pre-stored historical device data has changed according to an update cycle; If the historical equipment data changes, a fault statistical analysis is performed on the changed historical equipment data based on the monitoring characteristics to update the corresponding relationship between different monitoring characteristics and faults and their fault probabilities.
4. The fault detection method according to claim 1, It is characterized in that The fault detection method further comprises: Detecting whether the monitoring characteristics have changed according to the update cycle; If the monitoring feature changes, a fault statistical analysis is performed on the historical equipment data based on the changed monitoring feature to update the corresponding relationship between different monitoring features and faults and their fault probabilities.
5. The fault detection method according to claim 1, It is characterized in that The candidate faults include a plurality of faults, and determining the target fault of the smart device from the at least one candidate fault based on the fault probability of the at least one candidate fault includes: From the failure probabilities of multiple candidate failures, a candidate failure corresponding to a maximum failure probability is selected as a target failure of the smart device.
6. The fault detection method according to claim 1, It is characterized in that The monitoring features include business features and technical features, the business features include transaction events and user operation events, the technical features include network connection status and device operation status, and the determining of target operation features from the device data according to the preset monitoring features includes: Transaction events and user operation events are extracted from the device data according to the business characteristics, and network connection status and device operation status are extracted from the device data according to the technical characteristics to obtain target operation characteristics.
7. The fault detection method according to any one of claims 1 to 6, It is characterized in that Generating a fault detection result according to the target fault includes: When the failure probability of the target failure is a first threshold, the generated failure detection result is to determine that the smart device has not failed; When the failure probability of the target failure is a second threshold, the generated failure detection result is to determine that the failure occurring in the smart device is the target failure, and the second threshold is greater than the first threshold; When the failure probability of the target failure is within the range from the first threshold to the second threshold, the generated failure detection result is a prediction that the target failure occurs in the smart device, and the possible probability of occurrence is the failure probability of the target failure.
8. The fault detection method according to any one of claims 1 to 6, It is characterized in that After generating the fault detection result according to the target fault, the fault detection method further includes: The fault detection result is sent to a terminal or a user contact account associated with the smart device to prompt the user to handle the target fault.
9. A fault detection device, It is characterized in that include: A receiving unit, configured to receive device data sent by a smart device, wherein the device data includes operation characteristics; a feature determination unit, configured to determine a target operation feature from the device data according to a preset monitoring feature; An acquisition unit is used to acquire the corresponding relationship between different monitoring features constructed based on historical device data and faults and their fault probabilities, including: acquiring the historical device data corresponding to the smart device; counting the number of reports of the monitoring features, faults occurring based on the monitoring features, and the number of occurrences of the faults occurring based on the monitoring features based on the historical device data; calculating the fault probability of the fault according to the number of reports and the number of occurrences, and obtaining the corresponding relationship between different monitoring features and faults and their fault probabilities; a probability determination unit, configured to determine a failure probability of at least one candidate failure corresponding to the target operation feature according to the corresponding relationship; a fault determination unit, configured to determine a target fault of the smart device from at least one candidate fault based on the fault probability of the at least one candidate fault; A generating unit is used to generate a fault detection result according to the target fault.
10. A computer device, It is characterized in that The method comprises a processor and a memory, wherein the memory stores a computer program, and the processor executes the fault detection method according to any one of claims 1 to 8 when calling the computer program in the memory.
11. A storage medium, It is characterized in that The storage medium is used to store a computer program, and the computer program is loaded by a processor to execute the fault detection method according to any one of claims 1 to 8.
