A device fault detection and analysis method based on the Internet of Things platform

Abstract

The present invention provides a device fault detection and analysis method based on an Internet of Things platform, comprising: obtaining Internet of Things devices that have a communication connection relationship with the Internet of Things platform, and capturing the device work log of the Internet of Things device; extracting the fault information in the device work log; performing cluster analysis on the fault information of all Internet of Things devices that have a communication connection relationship, and searching for the fault cause that matches the fault feature from the preset fault cause set; constructing a fault cause set of the corresponding Internet of Things device based on the fault cluster set to which the fault information contained in each Internet of Things device belongs, and analyzing the deviation degree of the fault cluster set to which each fault cause belongs and the device's own device attributes, calibrating the fault level of the matched corresponding fault cause, and obtaining the fault level of the corresponding Internet of Things device. The method can effectively reduce the amount of processed data, improve the platform's response speed to faults, and improve the efficiency of fault detection and analysis.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things fault detection, and in particular to an equipment fault detection and analysis method based on an Internet of Things platform. Background Art

[0002] Equipment fault diagnosis is a technology that "diagnoses" equipment. It is a technology that understands and grasps the status of the equipment during use, determines whether it is normal or abnormal as a whole or in part, discovers faults and their causes at an early stage, and predicts the development trend of faults.

[0003] With the further development of intelligence, more and more IoT devices are emerging. However, due to the huge number of physical network devices, a large amount of data transmitted to the platform will cause the platform to respond slower to device failures, which will reduce the platform's response speed to a certain extent.

[0004] Therefore, the present invention proposes a device fault detection and analysis method based on the Internet of Things platform. Summary of the invention

[0005] The present invention provides an equipment fault detection and analysis method based on the Internet of Things platform, which is used to cluster the fault information existing in a number of devices, obtain the fault characteristics of each set and then match the fault cause, thereby effectively reducing the amount of processed data, improving the platform's response speed to faults, and improving the efficiency of fault detection and analysis.

[0006] The present invention provides a device fault detection and analysis method based on an Internet of Things platform, comprising:

[0007] Step 1: Obtain an IoT device that has a communication connection relationship with the IoT platform, and capture the device work log of the IoT device;

[0008] Step 2: Extracting the fault information in the equipment work log;

[0009] Step 3: clustering analysis is performed on the acquired fault information of all IoT devices with communication connection relationships, and according to the fault characteristics of the fault cluster set, a fault cause matching the fault characteristics is searched from a preset fault cause set;

[0010] Step 4: Based on the fault cluster set to which the fault information contained in each IoT device belongs, a set of fault causes of the corresponding IoT device is constructed;

[0011] Step 5: Analyze the deviation degree of each fault cause in the fault cause set based on the fault cluster set to which it belongs and the device attributes of the corresponding IoT device, and calibrate the fault level of each fault cause in the matching fault cause set;

[0012] Step 6: According to the fault level calibration result, obtain the fault level of the corresponding IoT device.

[0013] Preferably, obtaining an IoT device that has a communication connection relationship with the IoT platform and capturing a device work log of the IoT device includes:

[0014] Determine the first moment when the IoT device establishes a communication connection relationship with the physical network platform;

[0015] After a preset period based on the first moment, the log generated by the Internet of Things device is captured to obtain a device work log.

[0016] Preferably, extracting the fault information in the equipment work log includes:

[0017] Based on the log parsing model, the device work log is parsed, and the parsed result is traversed for faults n times, and each time the fault is traversed, a fault mark is set for the first position where the fault exists in the result;

[0018] The information of the setting mark position is extracted as the fault information.

[0019] Preferably, cluster analysis is performed on the acquired fault information of all IoT devices with communication connection relationships, including:

[0020] According to the pre-set cluster center points, establish the distance value between each cluster center point and each fault information corresponding to each IoT device;

[0021] According to the distance value, the corresponding fault information is classified into the cluster range of the corresponding cluster center point.

[0022] Preferably, searching for a fault cause matching the fault feature from a preset fault cause set according to the fault feature of the fault cluster set includes:

[0023] Obtaining a cluster distribution of each fault cluster set, and determining a distribution density of a distribution point of the cluster distribution at a current position;

[0024] According to the determination result, a concentrated distribution area and a discrete distribution area are determined, and the fault information corresponding to the concentrated distribution area is used as the primary information for the first analysis, and the fault information corresponding to the discrete distribution area is used as the secondary information for the second analysis;

[0025] Determining a degree of match between the second analysis result and the first analysis result;

[0026] If the matching degree is higher than a preset degree, searching for a fault cause matching the fault feature corresponding to the first analysis result from a preset set of fault causes;

[0027] Otherwise, the fault characteristics corresponding to different analysis results are searched from the preset fault cause set respectively, and the fault cause corresponding to the second analysis result is stored in the auxiliary storage unit.

[0028] Preferably, based on the fault cluster set to which the fault information contained in each IoT device belongs, a set of fault causes of the corresponding IoT device is constructed, including:

[0029] Determine whether each fault information in the corresponding IoT device is successfully attributed. If so, obtain the first code of the attribute set corresponding to each fault information, and obtain the fault cause set;

[0030] Otherwise, the second distance between the unattributed fault information and the cluster center point of each fault cluster set is determined, and the set corresponding to the smallest second cluster is selected as the attribution set to obtain the fault cause set.

[0031] Preferably, the fault level of each fault cause in the matched fault cause set is calibrated, including:

[0032] Obtain n setting results according to n traversals, and extract intersection fault identifiers, non-intersection overlapping fault identifiers, and individual fault identifiers;

[0033] According to the identity consistency, the intersection fault identity, the non-intersection overlapping fault identity and the single fault identity are divided into the same identity, and the reference degree of the sub-information of each position is determined by combining the overlapping times corresponding to the intersection, the overlapping times corresponding to the non-intersection and the single times;

[0034] Obtaining a clustering graph of each fault cause in the fault cause set based on the fault cluster set to which it belongs, and obtaining a first distance between the corresponding fault cause and a center point of the corresponding fault cluster set;

[0035] Determining a degree of deviation corresponding to a fault cause according to a first ratio of the first distance to a cluster concentration distance;

[0036] According to the historical evaluation fault table of the corresponding IoT device, determining the cause of the fault is based on the regularity of the occurrence of the historical evaluation fault table;

[0037] The fault value is calculated based on the reference degree Y1 of the sub-information at the corresponding position of each fault cause, the corresponding deviation degree Y2, and the occurrence regularity Y3;

[0038] Y=∝1Y1+∝2Y2+∝3Y3

[0039] Among them, ∝1 represents the first weight corresponding to the reference degree; ∝2 represents the second weight corresponding to the deviation degree; ∝3 represents the third weight corresponding to the regularity;

[0040] Based on the value-level list, the fault level consistent with the fault value is matched, and the corresponding fault cause is calibrated for significance.

[0041] Preferably, according to the fault level calibration result, the fault level of the corresponding IoT device is obtained, including:

[0042]

[0043] Among them, X1 represents the fault level of the corresponding IoT device; s i1 represents the fault level corresponding to the i-th calibration result in the corresponding IoT device; (s i1 ) max Indicates the maximum fault level corresponding to the corresponding IoT device; (s i1 ) min It represents the minimum fault level corresponding to the corresponding IoT device; ln represents the sign of the logarithmic function.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] By clustering the fault information of several devices, obtaining the fault characteristics of each set and then matching the fault causes, the amount of data processed can be effectively reduced, the platform's response speed to faults can be increased, and the efficiency of fault detection and analysis can be improved.

[0046] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0047] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0049] Figure 1 The present invention is a flowchart of a method for detecting and analyzing device faults based on an Internet of Things platform in an embodiment of the present invention. DETAILED DESCRIPTION

[0050] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0051] The present invention provides a device fault detection and analysis method based on the Internet of Things platform, such as Figure 1 As shown, including:

[0052] Step 1: Obtain an IoT device that has a communication connection relationship with the IoT platform, and capture the device work log of the IoT device;

[0053] Step 2: Extracting the fault information in the equipment work log;

[0054] Step 3: clustering analysis is performed on the acquired fault information of all IoT devices with communication connection relationships, and according to the fault characteristics of the fault cluster set, a fault cause matching the fault characteristics is searched from a preset fault cause set;

[0055] Step 4: Based on the fault cluster set to which the fault information contained in each IoT device belongs, a set of fault causes of the corresponding IoT device is constructed;

[0056] Step 5: Analyze the deviation degree of each fault cause in the fault cause set based on the fault cluster set to which it belongs and the device attributes of the corresponding IoT device, and calibrate the fault level of each fault cause in the matching fault cause set;

[0057] Step 6: According to the fault level calibration result, obtain the fault level of the corresponding IoT device.

[0058] In this embodiment, the physical network device refers to a device that has a communication connection relationship with the Internet of Things platform, such as a mobile phone, a computer, a notebook, a smart watch, etc.

[0059] In this embodiment, the fault information may refer to a fault generated during operation, such as a power failure, a communication failure, and the like.

[0060] In this embodiment, cluster analysis refers to grouping similar faults together to determine the characteristics of the set. For example, if there are fault information 1, 2, 3, 4, 5, 6, and 7, then fault information 1, 2, and 6 are one class, and fault information 3, 4, 5, and 7 are another class. At this time, fault feature 1 corresponding to fault information 1, 2, and 6, and fault feature 2 corresponding to fault information 3, 4, 5, and 7 are obtained.

[0061] In this embodiment, the preset fault cause set includes different fault features and fault causes corresponding to the fault features, and fault feature 1 is retained to correspond to fault cause 11. At this time, the corresponding fault cause 11 can be retrieved from the database.

[0062] In this embodiment, although fault information 1, 2, and 6 correspond to fault feature 1, fault information 2 and 6 basically completely match fault feature 1. Although fault information 1 also matches fault feature 1, there are some inconsistencies. Fault information 3, 4, 5, and 7 correspond to fault feature 2.

[0063] In this embodiment, device 1 contains fault information 1 and 3. At this time, the deviation degree of fault cluster set 1 to which fault information 1 belongs is a1, and the deviation degree of fault cluster set 2 to which fault information 3 belongs is a2. At this time, combined with the properties of the device itself, the levels of fault information 1 and fault information 3 are calibrated.

[0064] In this embodiment, a corresponding calibration level is obtained according to the calibration result.

[0065] The beneficial effect of the above technical solution is: by clustering the fault information of several devices, obtaining the fault characteristics of each set and then matching the fault causes, the amount of data processed can be effectively reduced, the platform's response speed to faults can be improved, and the efficiency of fault detection and analysis can be improved.

[0066] The present invention provides a device fault detection and analysis method based on an Internet of Things platform, which obtains an Internet of Things device that has a communication connection relationship with the Internet of Things platform and captures the device work log of the Internet of Things device, including:

[0067] Determine the first moment when the IoT device establishes a communication connection relationship with the physical network platform;

[0068] After a preset period based on the first moment, the log generated by the Internet of Things device is captured to obtain a device work log.

[0069] In this embodiment, the first moment refers to the moment when the IoT device and the platform just establish a connection relationship, and the preset period refers to the time after the connection, such as 5 seconds, when log capture is performed.

[0070] The beneficial effect of the above technical solution is: by determining the connection time and capturing the log after a preset period of time, the log is obtained, which is convenient for ensuring the rationality of log capture on the basis of stable connection.

[0071] The present invention provides a device fault detection and analysis method based on an Internet of Things platform, which extracts fault information existing in a working log of the device, including:

[0072] Based on the log parsing model, the device work log is parsed, and the parsed result is traversed for faults n times, and each time the fault is traversed, a fault mark is set for the first position where the fault exists in the result;

[0073] The information of the setting mark position is extracted as the fault information.

[0074] In this embodiment, the log parsing model is pre-trained and is obtained by sample training based on different logs and calibration of faults in the logs, and the training samples may be 1000 or more.

[0075] In this embodiment, n traversals are performed to ensure that all faults in the log are marked out and the integrity of the extracted fault information is guaranteed.

[0076] The beneficial effects of the above technical solution are: by parsing the log, it is convenient to effectively identify all existing faults, and by performing n traversals, it is convenient to ensure the integrity of fault extraction, providing an effective basis for subsequent fault analysis.

[0077] The present invention provides a device fault detection and analysis method based on an Internet of Things platform, which performs cluster analysis on the acquired fault information of all Internet of Things devices with communication connection relationships, including:

[0078] According to the pre-set cluster center points, establish the distance value between each cluster center point and each fault information corresponding to each IoT device;

[0079] According to the distance value, the corresponding fault information is classified into the cluster range of the corresponding cluster center point.

[0080] In this embodiment, the cluster center point is mainly used to classify different faults, and can effectively cluster the fault information within a preset distance.

[0081] The beneficial effect of the above technical solution is that by effectively dividing the existing fault information according to the cluster center, the reasonable clustering of the fault information can be ensured, providing a reasonable basis for matching the fault cause.

[0082] The present invention provides a device fault detection and analysis method based on an Internet of Things platform, which searches for a fault cause matching the fault feature from a preset fault cause set according to the fault feature of the fault cluster set, including:

[0083] Obtaining a cluster distribution of each fault cluster set, and determining a distribution density of a distribution point of the cluster distribution at a current position;

[0084] According to the determination result, a concentrated distribution area and a discrete distribution area are determined, and the fault information corresponding to the concentrated distribution area is used as the primary information for the first analysis, and the fault information corresponding to the discrete distribution area is used as the secondary information for the second analysis;

[0085] Determining a degree of match between the second analysis result and the first analysis result;

[0086] If the matching degree is higher than a preset degree, searching for a fault cause matching the fault feature corresponding to the first analysis result from a preset set of fault causes;

[0087] Otherwise, the fault characteristics corresponding to different analysis results are searched from the preset fault cause set respectively, and the fault cause corresponding to the second analysis result is stored in the auxiliary storage unit.

[0088] In this embodiment, after clustering the fault information, a cluster distribution centered on different cluster centers is obtained. Each point in the cluster distribution represents a fault information, and due to the different locations of the fault information, the final distribution density is also different. Therefore, the concentrated distribution area and the discrete distribution area in the cluster distribution can be determined.

[0089] In this embodiment, after determining the relevant information of the collective distribution area and the discrete distribution area, the information in the area is analyzed to determine the matching degree of different analysis results, mainly to judge whether they are consistent, and then effectively find and match the existing faults.

[0090] In this embodiment, the preset fault cause set includes different fault causes and matching fault features, so the fault cause can be obtained.

[0091] In this embodiment, the auxiliary storage unit is used to store the corresponding second analysis result.

[0092] The beneficial effect of the above technical solution is: by determining the density of the cluster distribution, the concentrated distribution of the cluster distribution and the area of ​​the discrete distribution are determined, and then the regional matching degree of the two areas is calculated, the fault is effectively and reasonably determined, and an effective basis is provided for matching the cause of the fault.

[0093] The present invention provides a device fault detection and analysis method based on an Internet of Things platform, which constructs a set of fault causes of the corresponding Internet of Things device based on the fault cluster set to which the fault information contained in each Internet of Things device belongs, including:

[0094] Determine whether each fault information in the corresponding IoT device is successfully attributed. If so, obtain the first code of the attribute set corresponding to each fault information, and obtain the fault cause set;

[0095] Otherwise, the second distance between the unattributed fault information and the cluster center point of each fault cluster set is determined, and the set corresponding to the smallest second cluster is selected as the attribution set to obtain the fault cause set.

[0096] In this embodiment, during the process of clustering fault information, the attribution may fail. Therefore, the unsuccessful attribution of fault information is assigned according to the distance from the cluster center to ensure the effective attribution of all fault information and avoid omissions.

[0097] The beneficial effect of the above technical solution is: by judging whether the fault information is successfully attributed, the corresponding fault cause set is obtained, and the fault information that is not successfully attributed is attributed according to the distance, thereby ensuring the validity of the fault cause set and providing an effective basis for determining the equipment fault level.

[0098] The present invention provides an equipment fault detection and analysis method based on an Internet of Things platform, which performs fault level calibration on each fault cause in a matching fault cause set, including:

[0099] Obtain n setting results according to n traversals, and extract intersection fault identifiers, non-intersection overlapping fault identifiers, and individual fault identifiers;

[0100] According to the identity consistency, the intersection fault identity, the non-intersection overlapping fault identity and the single fault identity are divided into the same identity, and the reference degree of the sub-information of each position is determined by combining the overlapping times corresponding to the intersection, the overlapping times corresponding to the non-intersection and the single times;

[0101] Obtaining a clustering graph of each fault cause in the fault cause set based on the fault cluster set to which it belongs, and obtaining a first distance between the corresponding fault cause and a center point of the corresponding fault cluster set;

[0102] Determining a degree of deviation corresponding to a fault cause according to a first ratio of the first distance to a cluster concentration distance;

[0103] According to the historical evaluation fault table of the corresponding IoT device, determining the cause of the fault is based on the regularity of the occurrence of the historical evaluation fault table;

[0104] The fault value is calculated based on the reference degree Y1 of the sub-information at the corresponding position of each fault cause, the corresponding deviation degree Y2, and the occurrence regularity Y3;

[0105] Y=∝1Y1+∝2Y2+∝3Y3

[0106] Among them, ∝1 represents the first weight corresponding to the reference degree; ∝2 represents the second weight corresponding to the deviation degree; ∝3 represents the third weight corresponding to the regularity;

[0107] Based on the value-level list, the fault level consistent with the fault value is matched, and the corresponding fault cause is calibrated for significance.

[0108] In this embodiment, the intersection fault identifier refers to the existence of the same fault identifier in each of the n setting results, and the overlapping fault identifier refers to the existence of the same fault identifier in some of the n setting results, and the range is 2 to n-1.

[0109] In this embodiment, the higher the number of overlaps, the greater the corresponding reference degree.

[0110] In this embodiment, the larger the first ratio is, the larger the corresponding deviation is.

[0111] In this embodiment, the historical evaluation fault table includes different fault causes and regularities, because different fault causes correspond to different regular causes.

[0112] In this embodiment, the value-level list includes different fault values ​​and corresponding levels, and the significance calibration results corresponding to different levels are different.

[0113] The beneficial effect of the above technical solution is: by determining the number of overlaps, the reference degree of different fault causes can be obtained, and the degree of deviation can be determined through the ratio, and then the fault value can be calculated to provide a basis for the significance calibration of the level.

[0114] The present invention provides a device fault detection and analysis method based on an Internet of Things platform, which obtains the fault level of the corresponding Internet of Things device according to the fault level calibration result, including:

[0115]

[0116] Among them, X1 represents the fault level of the corresponding IoT device; s i1 represents the fault level corresponding to the i-th calibration result in the corresponding IoT device; (s i1 ) max Indicates the maximum fault level corresponding to the corresponding IoT device; (s i1 ) min It represents the minimum fault level corresponding to the corresponding IoT device; ln represents the sign of the logarithmic function.

[0117] The beneficial effect of the above technical solution is: the fault level is calculated through different fault levels, providing a basis for the follow-up.

[0118] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A device fault detection and analysis method based on the Internet of Things platform, characterized in that: include: Step 1: Obtain an IoT device that has a communication connection relationship with the IoT platform, and capture the device work log of the IoT device; Step 2: Extracting the fault information in the equipment work log; Step 3: clustering analysis is performed on the acquired fault information of all IoT devices with communication connection relationships, and according to the fault characteristics of the fault cluster set, a fault cause matching the fault characteristics is searched from a preset fault cause set; Step 4: Based on the fault cluster set to which the fault information contained in each IoT device belongs, a set of fault causes of the corresponding IoT device is constructed; Step 5: Analyze each fault cause in the fault cause set, and calibrate the fault level of each fault cause in the matching fault cause set based on the degree of deviation of the fault cluster set to which it belongs and the device attributes of the corresponding IoT device; Step 6: According to the fault level calibration result, the fault level of the corresponding IoT device is obtained; Among them, according to the fault level calibration result, the fault level of the corresponding IoT device is obtained, including: Among them, X1 represents the fault level of the corresponding IoT device; s i1 represents the fault level corresponding to the i-th calibration result in the corresponding IoT device; (s i1 ) max Indicates the maximum fault level corresponding to the corresponding IoT device; (s i1 ) min Indicates the minimum fault level corresponding to the corresponding IoT device; ln represents the sign of the logarithmic function; The extraction of fault information in the equipment work log includes: Based on the log parsing model, the device work log is parsed, and the parsed result is traversed for faults n times, and each time the fault is traversed, a fault mark is set for the first position where the fault exists in the result; Extracting information of the setting mark position as fault information; The fault level of each fault cause in the matched fault cause set is calibrated, including: Obtain n setting results according to n traversals, and extract intersection fault identifiers, non-intersection overlapping fault identifiers, and individual fault identifiers; According to the identity consistency, the intersection fault identity, the non-intersection overlapping fault identity and the single fault identity are divided into the same identity, and the reference degree of the sub-information of each position is determined by combining the overlapping times corresponding to the intersection, the overlapping times corresponding to the non-intersection and the single times; Obtaining a clustering graph of each fault cause in the fault cause set based on the fault cluster set to which it belongs, and obtaining a first distance between the corresponding fault cause and a center point of the corresponding fault cluster set; Determining a degree of deviation corresponding to a fault cause according to a first ratio of the first distance to a cluster concentration distance; Determine the cause of the fault according to the historical evaluation fault table of the corresponding IoT device, based on the regularity of the occurrence of the historical evaluation fault table; The fault value is calculated based on the reference degree Y1 of the sub-information at the corresponding position of each fault cause, the corresponding deviation degree Y2, and the occurrence regularity Y3; Y=∝1Y1+∝2Y2+∝3Y3 Among them, ∝1 represents the first weight corresponding to the reference degree; ∝2 represents the second weight corresponding to the deviation degree; ∝3 represents the third weight corresponding to the regularity; Based on the value-level list, the fault level consistent with the fault value is matched, and the corresponding fault cause is calibrated for significance.

2. The device fault detection and analysis method based on the Internet of Things platform according to claim 1, characterized in that: Obtain an IoT device that has a communication connection relationship with the IoT platform, and capture the device work log of the IoT device, including: Determine the first moment when the IoT device establishes a communication connection relationship with the IoT platform; After a preset period based on the first moment, the log generated by the Internet of Things device is captured to obtain a device work log.

3. The device fault detection and analysis method based on the Internet of Things platform according to claim 1, characterized in that: Perform cluster analysis on the fault information of all IoT devices with communication connections, including: According to the pre-set cluster center points, establish the distance value between each cluster center point and each fault information corresponding to each IoT device; According to the distance value, the corresponding fault information is classified into the cluster range of the corresponding cluster center point.

4. The device fault detection and analysis method based on the Internet of Things platform as claimed in claim 1, characterized in that: According to the fault characteristics of the fault cluster set, searching for a fault cause matching the fault characteristics from a preset fault cause set includes: Obtaining a cluster distribution of each fault cluster set, and determining a distribution density of a distribution point of the cluster distribution at a current position; According to the determination result, a concentrated distribution area and a discrete distribution area are determined, and the fault information corresponding to the concentrated distribution area is used as the primary information for the first analysis, and the fault information corresponding to the discrete distribution area is used as the secondary information for the second analysis; Determining a degree of match between the second analysis result and the first analysis result; If the matching degree is higher than a preset degree, searching for a fault cause matching the fault feature corresponding to the first analysis result from a preset set of fault causes; Otherwise, the fault characteristics corresponding to different analysis results are searched from the preset fault cause set respectively, and the fault cause corresponding to the second analysis result is stored in the auxiliary storage unit.

5. The device fault detection and analysis method based on the Internet of Things platform as claimed in claim 1, characterized in that: Based on the fault cluster set to which the fault information contained in each IoT device belongs, a set of fault causes of the corresponding IoT device is constructed, including: Determine whether each fault information in the corresponding IoT device is successfully attributed. If so, obtain the first code of the attribute set corresponding to each fault information, and obtain the fault cause set; Otherwise, the second distance between the unattributed fault information and the cluster center point of each fault cluster set is determined, and the set corresponding to the smallest second cluster is selected as the attribution set to obtain the fault cause set.

Images