Internet of Things fault diagnosis method and system based on intelligent optimization algorithm

Through the combination of intelligent optimization algorithms and state evaluation models, the high cost and time-consuming problems of IoT fault diagnosis in complex fault conditions are solved, and the rapid and accurate fault identification and diagnosis are achieved, which improves the reliability and safety of the equipment.

CN118473901BActive Publication Date: 2024-11-01深圳市双银科技有限公司
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410624493.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-05-20
Publication Date
2024-11-01
Estimated Expiration
2044-05-20

AI Technical Summary

Technical Problem

The existing IoT fault diagnosis methods are cost-effective and time-consuming when facing complex and changing fault conditions, making it difficult to ensure the credibility and decision-making basis for fault diagnosis, especially on large-scale data sets.

Method used

The IoT fault diagnosis method based on intelligent optimization algorithm is adopted to eliminate noise and outliers through data preprocessing, and quickly identify potential faults using the status evaluation model, and conduct comprehensive analysis with historical data to reduce manual intervention and improve diagnostic accuracy.

Benefits of technology

It provides a reliable data foundation, quickly and accurately diagnoses faulty equipment, improves the efficiency and accuracy of fault identification, reduces diagnostic errors, and improves the reliability and safety of equipment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118473901B_ABST
    Figure CN118473901B_ABST
Patent Text Reader

Abstract

The present invention discloses an Internet of Things fault diagnosis method and system based on an intelligent optimization algorithm, and relates to the technical field of Internet of Things fault diagnosis. In order to solve the problem that the training and optimization process of a feedforward neural network may be very time-consuming, especially on a large-scale data set, it is difficult to cope with complex and changeable fault conditions, which affects the credibility of fault diagnosis and the basis for decision-making; the Internet of Things fault diagnosis method based on an intelligent optimization algorithm includes the following steps: Step 1: data collection; Step 2: data processing; Step 3: fault diagnosis; Step 4: fault processing; by pre-processing the data at the front end, a reliable data basis is provided for subsequent Internet of Things fault diagnosis, the location of the faulty device and the faulty device is effectively determined, and the fault mode related to the abnormal data is identified by combining the historical operation data and fault records of the device, and the abnormal condition of the device is discovered in time, and the occurrence of equipment failure is prevented, and the reliability and safety of the equipment are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things fault diagnosis, and in particular to an Internet of Things fault diagnosis method and system based on an intelligent optimization algorithm. Background Art

[0002] Due to the large number and wide distribution of IoT devices, fault diagnosis has become an important and challenging issue. Regarding IoT fault diagnosis, a Chinese patent with publication number: CN113541985B discloses a training method for an IoT fault diagnosis model based on a feedforward neural network, the method comprising: collecting IoT fault data to obtain a sample set, and dividing it into a training sample set and a test sample set; inputting the training sample set into a preset feedforward neural network for pre-training to obtain a trained feedforward neural network; calculating the prediction error; determining the current accuracy rate based on the prediction error, if the accuracy rate threshold is not reached, using the gradient descent method to optimize the trained feedforward neural network to obtain an initial IoT fault diagnosis model; the test sample set verifies and corrects the initial IoT fault diagnosis model to obtain a target IoT fault diagnosis model. The above invention achieves the beneficial effect of improving the accuracy of fault identification by the IoT fault pre-diagnosis model.

[0003] Due to the diversity and complexity of IoT devices, failure modes may change over time, but existing fault diagnosis methods have high computational costs, and the training and optimization process of feedforward neural networks can be very time-consuming, especially on large-scale data sets. It is difficult to cope with complex and changeable fault conditions, affecting the credibility of fault diagnosis and decision-making basis. Summary of the invention

[0004] The purpose of the present invention is to provide an Internet of Things fault diagnosis method and system based on an intelligent optimization algorithm, which provides a reliable data basis for subsequent Internet of Things fault diagnosis by preprocessing data at the front end, reduces manual intervention and diagnostic errors, and improves the efficiency and accuracy of equipment maintenance, so as to solve the problems raised in the above-mentioned background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] The IoT fault diagnosis method based on intelligent optimization algorithm includes the following steps:

[0007] Step 1: Data collection: Collect the operating data of devices in the Internet of Things and pre-process the operating data to eliminate the influence of noise and outliers;

[0008] Step 2: Data processing: Process the pre-processed operating data, identify abnormal data and potential faulty equipment, determine whether the equipment's operating status is normal, and locate the faulty equipment;

[0009] Step 3: Fault diagnosis: Analyze the faulty equipment in detail based on the output results of the intelligent optimization algorithm, determine the cause of the fault, explore the potential causes of the fault, and conduct a comprehensive analysis of the fault in combination with the actual operation status and historical data;

[0010] Step 4: Fault handling: The fault information of the faulty equipment is sent to the control center. The maintenance personnel make corresponding processing and response based on the fault information, repair the faulty equipment, and restore the normal operation of the equipment.

[0011] Furthermore, the data in step 1 is preprocessed to eliminate the influence of noise and outliers, including:

[0012] Read the collected operation data, determine the data type of the operation data, and obtain corresponding parameter verification rules according to the data type;

[0013] Extracting data feature information of the operating data, verifying the data feature information based on the parameter verification rule, obtaining a verification result, and determining whether there is outlier data in the operating data, and if so, deleting the outlier data;

[0014] Data integrity verification is performed based on the timestamp of the operating data and the parameter verification rules. If the data is incomplete, it is filled using basic statistical filling technology based on the data distribution. If the data is complete, a complete operating data transmission file is generated.

[0015] Furthermore, for the abnormal data and potential faulty equipment identified in step 2, determine whether the operating status of the equipment is normal, specifically:

[0016] Filter the operation data according to the effective operation status of the IoT device to obtain target operation data, wherein the operation data corresponds to the operation status of the IoT device one by one;

[0017] Extracting data features of the target operating data, and performing clustering processing on the target data based on the data features to obtain an operating data group of an IoT device;

[0018] Determine a target value of the operation data group, draw an operation change curve based on the target value, and determine an operation change weight value of the operation data group based on the operation change curve;

[0019] Based on the weight value, an analysis is performed through a constructed Internet of Things state evaluation model to obtain a performance evaluation value of the Internet of Things device;

[0020] Compare the performance evaluation value with a preset threshold value. If the performance evaluation value is less than the preset threshold value, determine that the device corresponding to the target data has abnormal data and has a potential fault problem, and further analyze the device;

[0021] Otherwise, it is determined that the performance of the device corresponding to the target data is normal and the operating state of the device is normal.

[0022] Furthermore, the fault diagnosis in step 3 specifically includes:

[0023] Reading the operation data of the device with abnormal data, and determining the possible failure mode of the device based on the reading result, describing the operation state of the device based on the possible failure mode, and determining the failure state of the device based on the description result;

[0024] Determine the fault phenomenon and target based on the fault state of the device, determine the starting point and path of the search according to the fault phenomenon and target, and find all fault models that may meet the conditions in the knowledge base. When multiple fault models match the current fault phenomenon or target, use a specific conflict resolution strategy to select the most likely model;

[0025] According to the selected fault model, the corresponding knowledge is extracted, questions are asked to the user based on the corresponding knowledge, the user is guided to perform fault diagnosis, and the user's answer is fuzzy matched with the standard answer in the knowledge base;

[0026] According to the confidence transfer algorithm, the confidence of the target is calculated. If the confidence is lower than the preset threshold, it goes back to the previous step and repeats this process until the cause of the fault is found. When the confidence is higher than the preset threshold, it continues to search the next layer until the search ends and the cause of the fault is output.

[0027] Furthermore, the comprehensive analysis of the fault in step 3 also includes:

[0028] Obtain the device type based on the output fault cause, analyze the device structure, working principle and common fault modes, and further explore potential fault causes;

[0029] Extracting historical operation data of the device, obtaining a historical operation change curve of the device based on the historical operation data, and analyzing performance change trends and common failure modes of the device;

[0030] Based on the structural characteristics of the equipment and the analysis results of historical operation data, a comprehensive assessment of the fault is conducted, the failure mode and cause of the equipment are summarized, and a maintenance plan is generated based on the failure mode and cause.

[0031] The present invention provides another technical solution, an Internet of Things fault diagnosis system based on an intelligent optimization algorithm, comprising:

[0032] Data collection unit, which is used to collect the operating data of devices in the Internet of Things through sensors and communication devices, and transmit it to the data center for storage and processing;

[0033] A data processing unit, used to obtain the operation data collected by the data acquisition unit, and identify faulty devices in the Internet of Things based on the correlation of the operation data of the Internet of Things devices, to obtain the devices with faults;

[0034] The fault diagnosis unit is used to diagnose the fault type of the faulty equipment based on the faulty equipment identification result, determine the fault mode and cause, and issue an alarm based on the diagnosis result and fault type.

[0035] Furthermore, the data processing unit comprises:

[0036] The device self-check module is used to traverse the operation data of the device according to its own characteristics and extract abnormal data that does not meet its own requirements;

[0037] A data classification module, used to classify the operation data of the device to obtain the device hardware operation data and software operation data;

[0038] The data association module is used to establish a first data association between the device hardware operation data and the software operation data based on the performance characteristics of the device operation status, establish a second data association between the device hardware operation data and the software operation data based on the rule characteristics of the device operation status, and establish a third data association between the device hardware operation data and the software operation data based on the common failure modes of the device operation status.

[0039] Furthermore, the device self-check module includes:

[0040] A time information data extraction module, used to extract the usage time information data of the device;

[0041] A usage time coefficient acquisition module, used to acquire a usage time coefficient using the usage time information data of the device;

[0042] A threshold adjustment module is used to adjust the threshold of the operation data of the device using the usage time coefficient, wherein the threshold of the operation data is used to determine whether there is abnormal data in the operation data; wherein the threshold of the adjusted operation data is obtained by the following formula:

[0043]

[0044] Where X represents the threshold of the adjusted operating data; X y represents the initial threshold value before adjustment of the operating data; t represents the usage time coefficient.

[0045] Further, a time coefficient acquisition module is used, including:

[0046] A first time information extraction module is used to extract the total usage time of the device from the startup revelation moment to the present;

[0047] A second time information extraction module is used to extract the continuous operation time corresponding to each start and stop use phase of the device;

[0048] The first usage time coefficient acquisition module is used to obtain the first usage time coefficient of the device by using the total usage time of the device from the start time to the current time; wherein the first usage time coefficient is obtained by the following formula:

[0049]

[0050] Among them, t 1 represents the first usage time coefficient; M represents the total number of starts and stops of the current equipment; m represents the number of starts and stops of the equipment in the current year; T k Indicates the number of years the equipment has been in operation; T e Indicates the number of years of operation corresponding to the rated service life of the equipment; T y Indicates the operating time of the equipment in the current operating years; T r Indicates the proportion of time that the equipment has been stopped in the current years of operation.

[0051] The second usage time coefficient acquisition module is used to obtain the second usage time coefficient of the device by using the continuous operation time corresponding to each start-stop usage phase of the device; wherein the second usage time coefficient is obtained by the following formula:

[0052]

[0053] Among them, t 2 Represents the second usage time coefficient; T i Indicates the operating time corresponding to the i-th start and stop of the equipment in the current year of operation; T max Indicates the maximum running time of a single start and stop in the previous year of the equipment's operation. When the equipment is currently in its first year of operation, the value of Tmax is 1.23T 0 , T 0 Indicates the maximum running time of a single start and stop of the equipment in the current year;

[0054] The usage time coefficient calculation module is used to obtain the usage time coefficient of the device by using the first usage time coefficient and the second usage time coefficient; wherein the usage time coefficient is obtained by the following formula:

[0055]

[0056] λ 1 =t 1 -t 2

[0057]

[0058] Wherein, t represents the usage time coefficient; e represents a constant.

[0059] Furthermore, the fault diagnosis unit comprises:

[0060] A first judgment module is used to judge whether the performance characteristics of the device meet the first data association, and if so, judge that the performance indicators of the device itself are normal; otherwise, judge that the performance indicators of the device itself are abnormal and issue a second alarm reminder;

[0061] A second judgment module is used to judge whether the rule characteristics of the device operation status meet the second data association after determining that the performance index of the device itself is normal, and if so, judge that the device driver is normal; otherwise, judge that the device driver is abnormal and issue a third alarm reminder;

[0062] A third judgment module is used to judge whether the common failure mode of the running state of the device satisfies the third data association after determining that the performance index of the device itself is normal and the device driver is normal;

[0063] If so, determine that the current operating state of the device is a common failure mode, and output the corresponding common failure mode as the fault diagnosis result; otherwise, determine that the current failure mode of the device is not an uncommon failure mode, and issue a fourth alarm reminder.

[0064] Compared with the prior art, the present invention has the following beneficial effects:

[0065] By preprocessing the data at the front end, a reliable data foundation is provided for subsequent IoT fault diagnosis. The IoT status assessment model analyzes the performance evaluation value, quickly and accurately diagnoses equipment with potential fault problems, and effectively determines the faulty equipment and the location of the faulty equipment. It can analyze the characteristics and patterns of abnormal data, combine the historical operation data and fault records of the equipment, identify the fault mode related to the abnormal data, promptly discover the abnormal situation of the equipment, prevent the occurrence of equipment failure, improve the reliability and safety of the equipment, reduce manual intervention and diagnostic errors, and improve the efficiency and accuracy of equipment maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 This is a flow chart of the Internet of Things fault diagnosis method based on the intelligent optimization algorithm of the present invention;

[0067] Figure 2 This is a module diagram of the Internet of Things fault diagnosis system based on the intelligent optimization algorithm of the present invention. DETAILED DESCRIPTION

[0068] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0069] To address the technical issues that the training and optimization process of feedforward neural networks may be very time-consuming, especially on large-scale data sets, and that it is difficult to cope with complex and changing fault conditions, which affects the credibility of fault diagnosis and the basis for decision making, please refer to Figure 1 , this embodiment provides the following technical solutions:

[0070] The IoT fault diagnosis method based on intelligent optimization algorithm includes the following steps:

[0071] Step 1: Data collection: Collect the operating data of the devices in the Internet of Things, including the status information, performance indicators, and environmental parameters of the devices, and pre-process the operating data to eliminate the influence of noise and outliers;

[0072] In this embodiment, the data is preprocessed in step 1 to eliminate the influence of noise and outliers, specifically including:

[0073] Read the collected operation data, determine the data type of the operation data, and obtain the corresponding parameter verification rules according to the data type;

[0074] Extract data feature information of the running data, verify the data feature information based on parameter verification rules, obtain verification results, determine whether there are outlier data in the running data, and if so, delete the outlier data;

[0075] Data integrity verification is performed based on the timestamp and parameter verification rules of the running data. If the data is incomplete, basic statistical filling technology is used to fill it according to the data distribution. If the data is complete, a complete running data transmission file is generated;

[0076] In this embodiment, the parameter verification rules can be guaranteed to fully verify the operation data, avoid data omission, improve precision, accuracy and stability, help identify and delete outlier data, further improve the quality of data, help reduce misjudgment caused by abnormal data, and improve the accuracy of fault diagnosis. The data integrity verification based on timestamp can ensure the continuity and accuracy of data in time series. For incomplete data, basic statistical filling technology is used to fill it, which ensures the integrity of the data. The data is pre-processed at the front end to provide a reliable data basis for subsequent IoT fault diagnosis.

[0077] Step 2: Data processing: Process the pre-processed operation data, including classification, clustering and association analysis of the operation data, identification of abnormal data and potential faulty equipment, judgment of whether the equipment's operating status is normal, and location of the faulty equipment;

[0078] In this embodiment, in order to identify abnormal data and potential faulty equipment in step 2, it is determined whether the operating status of the equipment is normal, specifically:

[0079] The operation data is filtered according to the effective operation status of the IoT device to obtain the target operation data, wherein the operation data corresponds to the operation status of the IoT device one by one. By filtering the operation data according to the effective operation status of the IoT device, accurate data corresponding to the operation status of the device can be obtained, thereby improving the pertinence of the diagnosis;

[0080] Extract the data features of the target operation data, and cluster the target data based on the data features to obtain the operation data group of the IoT device, which helps to understand the operation status and performance of the device in a more systematic way;

[0081] Determine the target value of the operation data group, draw the operation change curve based on the target value, and determine the operation change weight value of the operation data group based on the operation change curve, which provides a quantitative basis for the performance evaluation of the equipment and makes the evaluation result more objective and accurate;

[0082] Based on the weight value, the constructed IoT state evaluation model is used for analysis to obtain the performance evaluation value of the IoT device;

[0083] Compare the performance evaluation value with the preset threshold. If the performance evaluation value is less than the preset threshold, it is determined that the device corresponding to the target data has abnormal data and potential fault problems, and further analyze the device.

[0084] Otherwise, it is determined that the performance of the device corresponding to the target data is normal and the operating status of the device is normal;

[0085] In this embodiment, the performance evaluation value is analyzed through the IoT status evaluation model, which can quickly and accurately diagnose devices with potential fault problems. If the traffic of a certain device suddenly increases significantly, and this increase does not match the normal usage scenario, then it is possible that it has been attacked. The network connection of the device should be checked immediately to confirm whether there is a security threat and observe the network connection of the device. If it is found that the device is connected to an unknown or suspicious IP address, then the device may have been attacked. If the resource utilization of the device's CPU, memory, and storage suddenly increases or decreases significantly, then the device may have been attacked or a fault has occurred, which improves the efficiency and accuracy of fault diagnosis, and the IoT status evaluation model can be adjusted and optimized according to actual needs to effectively determine the faulty device and the location of the faulty device;

[0086] Step 3: Fault diagnosis: Analyze the faulty equipment in detail based on the output results of the intelligent optimization algorithm, determine the cause of the fault, explore the potential causes of the fault, and conduct a comprehensive analysis of the fault in combination with the actual operation status and historical data;

[0087] In this embodiment, the fault diagnosis in step 3 specifically includes:

[0088] Read the operation data of the equipment with abnormal data, and determine the possible failure mode of the equipment based on the reading results. Describe the operation status of the equipment based on the possible failure mode, including the description of the working environment, operation parameters, performance indicators, etc. of the equipment, so as to have a more comprehensive understanding of the equipment status and possible problems. At the same time, determine the failure status of the equipment based on the description results. According to the abnormal data, failure mode and operation status mentioned in the description results, it can be judged whether the equipment has a failure, the type and severity of the failure, etc.;

[0089] Determine the fault phenomenon and target based on the fault status of the equipment, determine the starting point and path of the search based on the fault phenomenon and target, and find all possible fault models that meet the conditions in the knowledge base. When multiple fault models match the current fault phenomenon or target, use a specific conflict resolution strategy to select the most likely model;

[0090] In this embodiment, the knowledge base may contain information such as various failure modes, failure causes, solutions, etc. of the equipment. Through algorithms such as fuzzy matching and association rule mining, a comprehensive evaluation can be performed based on factors such as the probability of failure, the severity of the failure, and the time of failure, and the most likely failure model can be selected, so that the failure model that matches the current failure phenomenon or target can be quickly found;

[0091] According to the selected fault model, the corresponding knowledge is extracted, and questions are asked to the user based on the corresponding knowledge to guide the user to perform fault diagnosis. The user's answer is fuzzy matched with the standard answer in the knowledge base, and the user is asked about the equipment's operating parameters, maintenance history, use environment, etc., so as to more accurately determine the cause of the fault;

[0092] According to the confidence transfer algorithm, the confidence of the target is calculated. If the confidence is lower than the preset threshold, it goes back to the previous step and repeats this process until the cause of the fault is found. When the confidence is higher than the preset threshold, it continues to search the next layer until the search ends and the cause of the fault is output.

[0093] In this embodiment, the characteristics and rules of abnormal data can be analyzed, and the fault mode related to the abnormal data can be identified by combining the historical operation data and fault records of the equipment. Based on the possible fault modes, when the equipment has abnormal sound, vibration or temperature abnormality, these phenomena can be regarded as fault phenomena, and fault diagnosis can be carried out with this as the goal. Starting from the structure, working principle and common fault modes of the equipment, the starting point and path of the search are determined to quickly locate the fault location and cause, realize intelligent analysis and fault diagnosis of equipment operation data, timely discover abnormal conditions of equipment, prevent the occurrence of equipment failures, improve the reliability and safety of equipment, reduce manual intervention and diagnostic errors, and improve the efficiency and accuracy of equipment maintenance;

[0094] In this embodiment, the comprehensive analysis of the fault in step 3 further includes:

[0095] Obtain the device type based on the output fault cause, analyze the device structure, working principle and common fault modes, and further explore potential fault causes;

[0096] Extract the historical operation data of the equipment, obtain the historical operation change curve of the equipment based on the historical operation data, and analyze the performance change trend and common failure modes of the equipment;

[0097] Comprehensively evaluate the faults based on the structural characteristics of the equipment and the analysis results of historical operation data, summarize the failure modes and causes of the equipment, and generate maintenance plans based on the failure modes and causes;

[0098] Step 4: Fault handling: The fault information of the faulty equipment is sent to the control center. The maintenance personnel make corresponding processing and response based on the fault information, repair the faulty equipment, and restore the normal operation of the equipment.

[0099] In this embodiment, the various steps are interrelated and coordinated with each other to achieve the best diagnostic effect. Through data collection and processing, the operating data of the equipment is obtained and abnormal data and faulty equipment are identified. The cause of the faulty equipment is determined through fault diagnosis. Based on fault handling, the equipment failure is discovered and handled in time, thereby improving the reliability and stability of the equipment and bringing more economic and social benefits to the enterprise.

[0100] See also Figure 2 , this embodiment provides an Internet of Things fault diagnosis system based on an intelligent optimization algorithm, including:

[0101] Data collection unit, which is used to collect the operating data of devices in the Internet of Things through sensors and communication devices, and transmit it to the data center for storage and processing;

[0102] The data processing unit is used to obtain the operation data collected by the data acquisition unit, and identify the faulty devices of the Internet of Things based on the correlation of the operation data of the Internet of Things devices, and obtain the devices with faults;

[0103] In this embodiment, the data processing unit includes:

[0104] The equipment self-check module is used to traverse the equipment's operating data according to the equipment's own characteristics, extract abnormal data that does not meet its own requirements, determine the equipment's operating abnormalities, and issue an alarm reminder;

[0105] The data classification module is used to classify the operation data of the device to obtain the device hardware operation data and software operation data. The device hardware operation data includes voltage, current, temperature, hard disk usage, and memory usage. The software operation data includes program operation logs, system logs, and network traffic.

[0106] A data association module, used to establish a first data association between the device hardware operation data and the software operation data based on the performance characteristics of the device operation state, establish a second data association between the device hardware operation data and the software operation data based on the rule characteristics of the device operation state, and establish a third data association between the device hardware operation data and the software operation data based on the common failure mode of the device operation state;

[0107] The fault diagnosis unit is used to diagnose the fault type of the faulty equipment based on the faulty equipment identification result, determine the fault mode and cause, and issue an alarm based on the diagnosis result and fault type.

[0108] Specifically, the device self-check module includes:

[0109] A time information data extraction module, used to extract the usage time information data of the device;

[0110] A usage time coefficient acquisition module, used to acquire a usage time coefficient using the usage time information data of the device;

[0111] A threshold adjustment module is used to adjust the threshold of the operation data of the device using the usage time coefficient, wherein the threshold of the operation data is used to determine whether there is abnormal data in the operation data; wherein the threshold of the adjusted operation data is obtained by the following formula:

[0112]

[0113] Where X represents the threshold of the adjusted operating data; X y represents the initial threshold value before adjustment of the operating data; t represents the usage time coefficient.

[0114] The technical effect of the above technical solution is: the equipment self-check module automatically extracts the equipment usage time information data, and further uses this data to automatically adjust the threshold of the operation data. This simplifies the equipment self-check process, reduces the need for manual intervention, and improves the efficiency and accuracy of equipment self-check. The usage time coefficient acquisition module can dynamically adjust the threshold according to the equipment usage time information data. This means that the threshold is not fixed, but is dynamically adjusted according to the actual usage and wear of the equipment. This dynamic adjustment mechanism makes the threshold closer to the actual situation of the equipment and improves the accuracy of anomaly detection.

[0115] The threshold adjustment module can adjust the threshold of the operating data according to the usage time coefficient, which helps to detect abnormal data more accurately. As the equipment is used for a longer time, its performance and parameters may change. By adjusting the threshold, these changes can be better adapted to ensure the effectiveness and accuracy of abnormal detection. Then, by extracting and analyzing the equipment's usage time information data, the wear and aging of the equipment can be better understood. This helps to detect potential problems and failures in a timely manner and take corresponding maintenance measures, thereby extending the service life of the equipment.

[0116] At the same time, by monitoring the equipment's operating data in real time and detecting abnormal data, potential problems can be discovered and resolved in a timely manner. This helps improve the reliability and stability of the equipment and reduce unexpected downtime and maintenance costs. Since the equipment self-check module can automatically adjust thresholds and detect anomalies, the need for manual intervention can be reduced. This reduces labor costs and error rates and improves the automation and intelligence level of the equipment.

[0117] In summary, the above technical solution realizes the automation and intelligent detection of equipment through the equipment self-check module, improves the accuracy of abnormality detection and the reliability of equipment, helps to extend the service life of equipment and reduce maintenance costs.

[0118] Specifically, the time coefficient acquisition module is used, including:

[0119] A first time information extraction module is used to extract the total usage time of the device from the startup revelation moment to the present;

[0120] A second time information extraction module is used to extract the continuous operation time corresponding to each start and stop use phase of the device;

[0121] The first usage time coefficient acquisition module is used to obtain the first usage time coefficient of the device by using the total usage time of the device from the start time to the current time; wherein the first usage time coefficient is obtained by the following formula:

[0122]

[0123] Among them, t 1 represents the first usage time coefficient; M represents the total number of starts and stops of the current equipment; m represents the number of starts and stops of the equipment in the current year; T k Indicates the number of years the equipment has been in operation; T e Indicates the number of years of operation corresponding to the rated service life of the equipment; T y Indicates the operating time of the equipment in the current operating years; T r Indicates the proportion of time that the equipment has been stopped in the current years of operation.

[0124] The second usage time coefficient acquisition module is used to obtain the second usage time coefficient of the device by using the continuous operation time corresponding to each start-stop usage phase of the device; wherein the second usage time coefficient is obtained by the following formula:

[0125]

[0126] Among them, t 2 Represents the second usage time coefficient; T i Indicates the operating time corresponding to the i-th start and stop of the equipment in the current year of operation; T max Indicates the maximum running time of a single start and stop in the previous year of the equipment's operation. When the equipment is currently in its first year of operation, the value of Tmax is 1.23T 0 , T 0 Indicates the maximum running time of a single start and stop of the equipment in the current year;

[0127] The usage time coefficient calculation module is used to obtain the usage time coefficient of the device by using the first usage time coefficient and the second usage time coefficient; wherein the usage time coefficient is obtained by the following formula:

[0128]

[0129] λ 1 =t 1 -t 2

[0130]

[0131] Wherein, t represents the usage time coefficient; e represents a constant.

[0132] The technical effect of the above technical solution is that through the first time information extraction module and the second time information extraction module, key time information such as the total usage time of the device and the continuous operation time of each start-stop usage phase can be fully extracted. This information provides basic data for the subsequent calculation of the usage time coefficient.

[0133] The first usage time coefficient is calculated by a specific formula using the total usage time of the equipment from the start-up time to the current time, combined with the number of starts and stops, the number of years of operation and other information. This coefficient reflects the overall usage and wear of the equipment. For each start-stop use phase of the equipment, the second usage time coefficient is calculated by extracting the corresponding continuous operation time and comparing these time periods with the maximum operation time in the previous year. This coefficient reflects the performance changes and stability of the equipment in different operation phases. The usage time coefficient calculation module comprehensively considers the first usage time coefficient and the second usage time coefficient, combines them through a specific formula, and obtains the final usage time coefficient. This coefficient comprehensively reflects the overall usage and performance of the equipment.

[0134] By adjusting the threshold using the time factor, you can ensure that the threshold always matches the actual status of the device. As the device is used for a longer time, its performance and parameters may change. Adjusting the threshold can better adapt to these changes and improve the accuracy of anomaly detection. By monitoring the operating data and usage time factor of the device, you can better understand the wear and aging of the device. This helps to detect potential problems in time and take corresponding maintenance measures, thereby extending the service life of the equipment.

[0135] Real-time monitoring of equipment operating data and detection of abnormal data helps to promptly identify and resolve potential problems and improve equipment reliability and stability. This can reduce unexpected downtime and maintenance costs and improve the overall performance of the equipment. Since the equipment self-check module can automatically adjust thresholds and detect anomalies, it can reduce the need for manual intervention. This reduces labor costs and error rates and improves the automation and intelligence level of the equipment.

[0136] In summary, the above technical solution realizes the automation and intelligent detection of equipment through the equipment self-check module, improves the accuracy of abnormality detection and the reliability of equipment, helps to extend the service life of equipment and reduce maintenance costs.

[0137] In this embodiment, the fault diagnosis unit includes:

[0138] A first judgment module is used to judge whether the performance characteristics of the device meet the first data association. If so, it is judged that the performance indicators of the device itself are normal; otherwise, it is judged that the performance indicators of the device itself are abnormal and a second alarm reminder is issued;

[0139] The second judgment module is used to judge whether the rule characteristics of the device operation status meet the second data association after determining that the performance indicators of the device itself are normal. If so, it is judged that the device driver is normal; otherwise, it is judged that the device driver is abnormal and a third alarm reminder is issued;

[0140] A third judgment module is used to judge whether the common failure mode of the device operation state satisfies the third data association after determining that the performance index of the device itself is normal and the device driver is normal;

[0141] If so, determine that the current operating state of the device is a common failure mode, and output the corresponding common failure mode as the fault diagnosis result; otherwise, determine that the current failure mode of the device is not an uncommon failure mode, and issue a fourth alarm reminder.

[0142] In this embodiment, the operation data of the Internet of Things devices are comprehensively collected through various sensors and communication devices, the devices with abnormal operation are quickly identified, the hardware and software operation data are classified, and multi-level data associations are established, which provides richer clues and basis for fault diagnosis. The first, second and third judgment modules realize comprehensive fault diagnosis from performance indicators to drivers to common fault modes, improve the accuracy and completeness of diagnosis, and provide corresponding alarm reminders according to abnormal conditions at different stages during the diagnosis process, so that maintenance personnel can locate and handle faults in a timely and accurate manner. It can provide comprehensive, accurate and efficient support for fault diagnosis of Internet of Things devices, help to timely discover and solve potential fault problems, and ensure the normal operation of Internet of Things devices.

[0143] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. The IoT fault diagnosis method based on intelligent optimization algorithm is characterized by: The following steps are involved: Step 1: Data collection: Collect the operating data of devices in the Internet of Things and pre-process the operating data to eliminate the influence of noise and outliers; Step 2: Data processing: Process the pre-processed operating data, identify abnormal data and potential faulty equipment, determine whether the equipment's operating status is normal, and locate the faulty equipment; In view of the abnormal data and potential faulty equipment identified in step 2, determine whether the equipment is operating normally, specifically: Filter the operation data according to the effective operation status of the IoT device to obtain target operation data, wherein the operation data corresponds to the operation status of the IoT device one by one; Extracting data features of the target operation data, and performing clustering processing on the target operation data based on the data features to obtain an operation data group of the Internet of Things device; Determine a target value of the operation data group, draw an operation change curve based on the target value, and determine an operation change weight value of the operation data group based on the operation change curve; Based on the weight value, an analysis is performed through a constructed Internet of Things state evaluation model to obtain a performance evaluation value of the Internet of Things device; Compare the performance evaluation value with a preset threshold value. If the performance evaluation value is less than the preset threshold value, determine that the device corresponding to the target operation data has abnormal data and has a potential fault problem, and further analyze the device; Otherwise, it is determined that the equipment performance corresponding to the target operation data is normal and the operation status of the equipment is normal; Step 3: Fault diagnosis: Analyze the faulty equipment in detail based on the output results of the intelligent optimization algorithm, determine the cause of the faulty equipment, explore the potential causes of the fault, and conduct a comprehensive analysis of the fault in combination with the actual operation status and historical data; The fault diagnosis in step 3 specifically includes: Reading the operation data of the device with abnormal data, and determining the possible failure mode of the device based on the reading result, describing the operation state of the device based on the possible failure mode, and determining the failure state of the device based on the description result; Determine the fault phenomenon and target based on the fault state of the device, determine the starting point and path of the search according to the fault phenomenon and target, and find all fault models that may meet the conditions in the knowledge base. When multiple fault models match the current fault phenomenon or target, use a specific conflict resolution strategy to select the most likely model; According to the selected fault model, the corresponding knowledge is extracted, questions are asked to the user based on the corresponding knowledge, the user is guided to perform fault diagnosis, and the user's answer is fuzzy matched with the standard answer in the knowledge base; According to the confidence transfer algorithm, the confidence of the target is calculated. If the confidence is lower than the preset threshold, it goes back to the previous step and repeats this process until the cause of the fault is found. When the confidence is higher than the preset threshold, it continues to search the next layer until the search ends and the cause of the fault is output. Comprehensive analysis of the fault in step 3 also includes: Obtain the device type based on the output fault cause, analyze the device structure, working principle and common fault modes, and further explore potential fault causes; Extracting historical operation data of the device, obtaining a historical operation change curve of the device based on the historical operation data, and analyzing performance change trends and common failure modes of the device; Comprehensively evaluate the faults based on the structural characteristics of the equipment and the analysis results of historical operation data, summarize the failure modes and causes of the equipment, and generate maintenance plans based on the failure modes and causes; Step 4: Fault handling: The fault information of the faulty equipment is sent to the control center, and the maintenance personnel make corresponding processing and response based on the fault information, repair the faulty equipment, and restore the normal operation of the equipment; The IoT fault diagnosis method based on intelligent optimization algorithm uses the following system for diagnosis, which includes: Data collection unit, which is used to collect the operating data of devices in the Internet of Things through sensors and communication devices, and transmit it to the data center for storage and processing; The data processing unit is used to obtain the operation data collected by the data acquisition unit, and identify the faulty devices of the Internet of Things based on the correlation of the operation data of the Internet of Things devices to obtain the faulty devices, including: The device self-check module is used to traverse the operation data of the device according to its own characteristics and extract abnormal data that does not meet its own requirements; A data classification module, used to classify the operation data of the device to obtain the device hardware operation data and software operation data; A data association module, used to establish a first data association between the device hardware operation data and the software operation data based on the performance characteristics of the device operation state, establish a second data association between the device hardware operation data and the software operation data based on the rule characteristics of the device operation state, and establish a third data association between the device hardware operation data and the software operation data based on the common failure mode of the device operation state; A fault diagnosis unit is used to diagnose the fault type of the faulty equipment based on the faulty equipment identification result, determine the fault mode and cause, and issue an alarm based on the diagnosis result and the fault type; Equipment self-check module, including: A time information data extraction module, used to extract the usage time information data of the device; A usage time coefficient acquisition module, used to acquire a usage time coefficient using the usage time information data of the device; A threshold adjustment module is used to adjust the threshold of the operation data of the device using the usage time coefficient, wherein the threshold of the operation data is used to determine whether there is abnormal data in the operation data; wherein the threshold of the adjusted operation data is obtained by the following formula: Where X represents the threshold of the adjusted operating data; X y represents the initial threshold value before adjustment of the operating data; t represents the usage time coefficient; Use the time coefficient acquisition module, including: A first time information extraction module is used to extract the total usage time of the device from the startup revelation moment to the present; A second time information extraction module is used to extract the continuous operation time corresponding to each start and stop use phase of the device; The first usage time coefficient acquisition module is used to obtain the first usage time coefficient of the device by using the total usage time of the device from the start time to the current time; wherein the first usage time coefficient is obtained by the following formula: Wherein, t1 represents the first usage time coefficient; M represents the total number of starts and stops of the current equipment; m represents the number of starts and stops of the equipment in the current year; T k Indicates the number of years the equipment has been in operation; T e Indicates the number of years of operation corresponding to the rated service life of the equipment; T y Indicates the operating time of the equipment in the current operating years; T r Indicates the proportion of time that the equipment has been stopped in the current years of operation. The second usage time coefficient acquisition module is used to obtain the second usage time coefficient of the device by using the continuous operation time corresponding to each start-stop usage phase of the device; wherein the second usage time coefficient is obtained by the following formula: Wherein, t2 represents the second usage time coefficient; T i Indicates the operating time corresponding to the i-th start and stop of the equipment in the current year of operation; T max Indicates the maximum running time of a single start and stop of the equipment in the previous year of operation. When the equipment is currently in its first year of operation, the value of Tmax is 1.23T0, where T0 indicates the maximum running time of a single start and stop of the equipment in the current year of operation. The usage time coefficient calculation module is used to obtain the usage time coefficient of the device by using the first usage time coefficient and the second usage time coefficient; wherein the usage time coefficient is obtained by the following formula: Wherein, t represents the usage time coefficient; e represents a constant.

2. The method for fault diagnosis of the Internet of Things based on the intelligent optimization algorithm according to claim 1, characterized in that: Preprocess the data in step 1 to eliminate the influence of noise and outliers, including: Read the collected operation data, determine the data type of the operation data, and obtain corresponding parameter verification rules according to the data type; Extracting data feature information of the operating data, verifying the data feature information based on the parameter verification rule, obtaining a verification result, and determining whether there is outlier data in the operating data, and if so, deleting the outlier data; Data integrity verification is performed based on the timestamp of the operating data and the parameter verification rules. If the data is incomplete, it is filled using basic statistical filling technology based on the data distribution. If the data is complete, a complete operating data transmission file is generated.

3. The method for fault diagnosis of the Internet of Things based on intelligent optimization algorithm according to claim 1, characterized in that: Fault diagnosis unit, including: A first judgment module is used to judge whether the performance characteristics of the device meet the first data association, and if so, judge that the performance indicators of the device itself are normal; otherwise, judge that the performance indicators of the device itself are abnormal and issue a second alarm reminder; A second judgment module is used to judge whether the rule characteristics of the device operation status meet the second data association after determining that the performance index of the device itself is normal, and if so, judge that the device driver is normal; otherwise, judge that the device driver is abnormal and issue a third alarm reminder; A third judgment module is used to judge whether the common failure mode of the running state of the device satisfies the third data association after determining that the performance index of the device itself is normal and the device driver is normal; If so, determine that the current operating state of the device is a common failure mode, and output the corresponding common failure mode as the fault diagnosis result; otherwise, determine that the current failure mode of the device is not an uncommon failure mode, and issue a fourth alarm reminder.