Internet of Things equipment fault diagnosis method and system based on artificial intelligence

Abstract

The invention relates to the field of equipment fault diagnosis, in particular to an Internet of Things equipment fault diagnosis method and system based on artificial intelligence, and the method comprises the steps: collecting operation state data and alarm signals of Internet of Things equipment, and carrying out the feature extraction to obtain a feature parameter set of the alarm signals; acquiring historical false alarm signals, extracting features of the false alarm signals and generating a false alarm signal feature library; performing feature matching on the feature parameter set of the alarm signal and a false alarm signal feature library, identifying and marking a false alarm signal label, and extracting a corresponding fault feature parameter from the false alarm signal feature library; judging the alarm signal with the false alarm signal label through a time delay verification method, and if the fault characteristic parameter of the alarm signal disappears, judging the alarm signal as a false alarm signal and removing the false alarm signal; if the fault characteristic parameter of the alarm signal does not disappear, determining that the alarm signal is a real alarm signal; real fault signals of equipment can be diagnosed more accurately, so that the accuracy and efficiency of fault diagnosis are improved.

Description

Technical Field

[0001] The present invention relates to the field of equipment fault diagnosis, and in particular to an Internet of Things equipment fault diagnosis method and system based on artificial intelligence. Background Art

[0002] The Intensive Care Unit (ICU) is the place in the hospital with the most stringent requirements for patient life support and monitoring. Its core relies on various Internet of Things (IoT) devices to achieve comprehensive monitoring of patient status. These devices include vital sign monitors (such as electrocardiograms and blood oxygen saturation monitors), ventilators, injection pumps, infusion monitoring systems, etc. The stability of their operating status is directly related to the safety of patients. However, in actual applications, due to the complex environment in the ICU, such as electromagnetic interference from multiple devices, changes in temperature and humidity, and increased network transmission load, IoT devices are prone to operating abnormalities.

[0003] Existing fault diagnosis technologies for ICU IoT devices mainly rely on traditional monitoring and maintenance methods, such as alarm systems based on single-point anomaly detection or regular manual maintenance. Traditional technologies have difficulty in achieving effective integration and dynamic analysis of multi-source information, and cannot quickly determine the authenticity of alarm signals and the operating status of equipment, which may lead to delayed diagnosis or false alarm interference. In addition, since there are many ICU IoT devices and many alarm signals are generated, many false alarm signals are often generated, which will greatly interfere with the accurate monitoring of patient status.

[0004] Therefore, there is an urgent need for a fault diagnosis method for ICU IoT devices that can accurately diagnose the real fault signals of the equipment, thereby improving the accuracy and efficiency of fault diagnosis. Summary of the invention

[0005] (1) Technical issues to be solved

[0006] The purpose of the present invention is to provide an IoT device fault diagnosis method and system based on artificial intelligence to solve the problem that many false alarm signals are often generated in the alarm signals generated by IoT devices, and the false alarm signals cannot be effectively identified, which will greatly interfere with the safety monitoring of the status of the monitored object.

[0007] (2) Technical solution

[0008] To achieve the above objectives, on the one hand, the present invention provides an IoT device fault diagnosis method based on artificial intelligence, the method comprising:

[0009] S1. Collect the operation status data and alarm signals of IoT devices in real time and record them as first data, perform data preprocessing on the first data to obtain second data, perform feature extraction on the second data to obtain a feature parameter set of the alarm signal, wherein the feature parameter set includes frequency change, signal duration, and amplitude abnormality; the data preprocessing includes noise filtering, missing value filling, and standardization.

[0010] S2. Obtain historical false alarm signals of IoT devices and perform feature analysis through a deep learning model to extract features of false alarm signals and generate a false alarm signal feature library; perform feature matching on the feature parameter set of the alarm signal and the false alarm signal feature library and calculate cosine similarity. When the cosine similarity is greater than a preset similarity threshold, mark the alarm signal with a false alarm signal label and extract corresponding fault feature parameters from the false alarm signal feature library.

[0011] S3. The alarm signal with a false alarm signal label is judged by a time delay verification method. The time delay verification method is to continuously collect the operating status data of the Internet of Things device within the delay time threshold and record it as the third data, and perform dynamic trend analysis on the fault characteristic parameters in combination with the third data to determine whether the fault characteristic parameters of the alarm signal disappear. If the fault characteristic parameters of the alarm signal disappear, it is judged as a false alarm signal, and the alarm signal is removed; if the fault characteristic parameters of the alarm signal do not disappear, it is judged as a real alarm signal, and fault alarm information is generated and the Internet of Things device maintenance process is triggered.

[0012] Furthermore, the delay verification method is to continuously collect the operating status data of the IoT device within the delay time threshold and record it as the third data, and perform dynamic trend analysis on the fault characteristic parameters in combination with the third data. The method for determining whether the fault characteristic parameters of the alarm signal disappear includes:

[0013] Obtain historical fault data of IoT devices, obtain the fault response time after each fault occurs based on the historical fault data, and calculate the average fault response time and standard deviation σ T , and calculate the delay time threshold T d ; The delay time threshold T d The calculation formula is:

[0014]

[0015] Where β is a confidence factor, which is used to determine the sensitivity of the delay time threshold.

[0016] At the delay time threshold T d The operating status data of IoT devices are collected at a preset sampling frequency and recorded as the delay data set Dt , D t ={x1,…,x i ,…,x n}, t∈[0,T d ]; where x i represents the operating status data of the IoT device at the i-th sampling time; n is in [0,T d ] The number of sampling times within the time period.

[0017] The delay data set D t Combine the fault feature parameters of IoT devices to build the fault feature matrix X:

[0018]

[0019] Among them, x i,j is the characteristic value of the jth fault at the i-th sampling moment; i∈[1,n], j∈[1,m]; m represents the number of fault characteristic parameters.

[0020] The alarm signal continuity index C of each fault characteristic parameter within the delay time threshold is calculated through a sliding window j ; The continuity index C j The calculation formula is:

[0021]

[0022] Among them, μ j is the historical average value of each fault characteristic parameter; ∈ j Tolerance interval, indicating the characteristic value x of the current sample i,j The allowable error range; I(·) is the indicator function used to determine the characteristic value x of the current sampling i,j Is it within the tolerance range?

[0023] The significance index S of each fault characteristic parameter is obtained through the significance index calculation formula j , the significance index calculation formula is:

[0024]

[0025] Among them, max(x i,j ) indicates that the delay time threshold T d The maximum sampling characteristic value of the jth fault characteristic parameter, min(x i,j ) indicates that the delay time threshold T d The minimum sampling characteristic value of the jth fault characteristic parameter within σ j is the historical standard deviation of the j-th fault characteristic parameter.

[0026] The continuity indicator C of the alarm signalj and the significance index S j A comprehensive evaluation value is obtained through a linear weighted algorithm, and the comprehensive evaluation value is compared with the preset disappearance threshold to determine whether the fault characteristic parameters of the alarm signal have disappeared; if the comprehensive evaluation value is less than the preset disappearance threshold, it is determined that the fault characteristic parameters of the alarm signal have disappeared; if the comprehensive evaluation value is greater than the preset disappearance threshold, it is determined that the fault characteristic parameters of the alarm signal have not disappeared.

[0027] Furthermore, the method further comprises:

[0028] Obtain historical alarm signals of IoT devices, and analyze the false alarm signal probability and true alarm signal probability of the historical alarm signals through data statistics; combine the fault characteristic parameters of the alarm signal and use the probability density estimation method to construct the conditional probability distribution.

[0029] The false alarm suppression strategy of the Internet of Things device is determined according to the conditional probability distribution; when the false alarm distribution probability in the conditional probability distribution is greater than the preset suppression threshold, the false alarm suppression strategy is triggered and the alarm signal is marked with a false alarm label, otherwise, the delay time threshold is adjusted.

[0030] The delay time threshold is adjusted by a dynamic correction factor, where the dynamic correction factor is the ratio of the false alarm distribution probability to the true alarm distribution probability.

[0031] Furthermore, the method further comprises:

[0032] The network signal transmission parameters are collected in real time through the network monitoring tool and recorded in time series. The network signal transmission parameters include transmission delay, packet loss rate and jitter value.

[0033] The network signal transmission parameters are applied through a weighted algorithm to obtain the transmission characteristics of the alarm signal under different network conditions. The transmission characteristics are:

[0034] d t =d0+α·q+β·p;

[0035] Where: d t is the comprehensive transmission delay of the signal in the network; α represents the weight factor of the jitter value q on the transmission delay, β represents the weight factor of the packet loss rate p on the transmission delay, α and β are obtained by fitting and analyzing historical data; d0 represents the transmission time required for the signal to be transmitted from the IoT device to the monitoring system.

[0036] Get the standard transmission delay d under ideal network conditions l , calculate the comprehensive transmission delay d t and standard transmission delay d l The delay error E d .

[0037] E d =d t ―d l ; when E d If the error is greater than the preset error threshold, it is determined that the current network status affects the transmission of the alarm signal, and the delay error E d Modify the delay time threshold; when E d If the error is less than the preset threshold, it is determined that the current network status has no effect on the alarm signal transmission.

[0038] A security monitoring time threshold of the Internet of Things device is obtained, and the corrected delay time threshold is less than the security monitoring time threshold.

[0039] Furthermore, the method further comprises:

[0040] Get the historical monitoring data D of the IoT device monitoring object l ; Obtain the physiological parameters P of the monitored object collected in real time s And form the time series:

[0041] P s ={p t |t=1,2,...,T};

[0042] p t is the real-time physiological parameter at the acquisition time t; the historical monitoring data D l and physiological parameters P s Joint composition analysis dataset:

[0043] D f ={D l , P s};

[0044] Through historical monitoring data D l Construct the abnormal state distribution model of the monitored object to obtain the abnormal state P a (x):

[0045]

[0046] Among them, μ a is the mean value of physiological parameters under abnormal conditions in historical monitoring data; is the variance of physiological parameters under abnormal conditions in historical monitoring data.

[0047] The physiological parameter P s With abnormal status P a (x) The deviation degree D is obtained by the deviation degree calculation method p :

[0048]

[0049] Deviation degree D p Indicates the degree of abnormal state of the monitored object; σ a is the standard deviation of physiological parameters in abnormal conditions in historical monitoring data; according to the degree of deviation D p Dynamically adjust the security monitoring time threshold T of IoT devices s :

[0050]

[0051] Among them, T s0 is the initial safety monitoring time threshold; λ is the correction factor used to control the sensitivity of the deviation degree to the adjustment of the safety monitoring time threshold; It depends on the degree of deviation D p An increasing and decreasing exponential correction term.

[0052] Based on the same inventive concept, the present invention also provides an IoT device fault diagnosis system based on artificial intelligence, the system comprising: a characteristic parameter set acquisition module, an alarm signal analysis module, an alarm signal verification and determination module, and each module is sequentially connected in communication;

[0053] The characteristic parameter set acquisition module is used to collect the operating status data and alarm signals of the Internet of Things devices in real time and record them as first data, perform data preprocessing on the first data to obtain second data, and perform feature extraction on the second data to obtain the characteristic parameter set of the alarm signal, wherein the characteristic parameter set includes frequency change, signal duration, and amplitude abnormal values; the data preprocessing includes noise filtering, missing value filling and standardization processing.

[0054] The alarm signal analysis module is used to obtain historical false alarm signals of IoT devices and perform feature analysis through a deep learning model, extract the features of the false alarm signals and generate a false alarm signal feature library; feature match the feature parameter set of the alarm signal with the false alarm signal feature library and calculate the cosine similarity. When the cosine similarity is greater than a preset similarity threshold, the alarm signal is marked with a false alarm signal label, and the corresponding fault feature parameters are extracted from the false alarm signal feature library.

[0055] The alarm signal verification and judgment module is used to judge the alarm signal with a false alarm signal label through a delay verification method. The delay verification method is to continuously collect the operating status data of the Internet of Things device within the delay time threshold and record it as the third data, and perform dynamic trend analysis on the fault characteristic parameters in combination with the third data to determine whether the fault characteristic parameters of the alarm signal disappear. If the fault characteristic parameters of the alarm signal disappear, it is determined to be a false alarm signal and the alarm signal is removed; if the fault characteristic parameters of the alarm signal do not disappear, it is determined to be a real alarm signal, and fault alarm information is generated and the Internet of Things device maintenance process is triggered.

[0056] Furthermore, the alarm signal verification and determination module comprises the following steps:

[0057] Obtain historical fault data of IoT devices, obtain the fault response time after each fault occurs based on the historical fault data, and calculate the average fault response time and standard deviation σ T , and calculate the delay time threshold T d ; The delay time threshold T d The calculation formula is:

[0058]

[0059] Where β is a confidence factor, which is used to determine the sensitivity of the delay time threshold.

[0060] At the delay time threshold T d The operating status data of IoT devices are collected at a preset sampling frequency and recorded as the delay data set D t , D t ={x1,…,x i ,…,x n}, t∈[0,T d ]; where x i represents the operating status data of the IoT device at the i-th sampling time; n is in [0,T d ] The number of sampling times within the time period.

[0061] The delay data set D t Combine the fault feature parameters of IoT devices to build the fault feature matrix X:

[0062]

[0063] Among them, x i,j is the characteristic value of the jth fault at the i-th sampling moment; i∈[1,n], j∈[1,m]; m represents the number of fault characteristic parameters.

[0064] The alarm signal continuity index C of each fault characteristic parameter within the delay time threshold is calculated through a sliding window j ; The continuity index C j The calculation formula is:

[0065]

[0066] Among them, μ j is the historical average value of each fault characteristic parameter; ∈ j Tolerance interval, indicating the characteristic value x of the current sample i,j The allowable error range; I(·) is the indicator function used to determine the characteristic value x of the current sampling i,j Is it within the tolerance range?

[0067] The significance index S of each fault characteristic parameter is obtained through the significance index calculation formula j , the significance index calculation formula is:

[0068]

[0069] Among them, max(x i,j ) indicates that the delay time threshold T d The maximum sampling characteristic value of the jth fault characteristic parameter, min(x i,j ) indicates that the delay time threshold T d The minimum sampling characteristic value of the j-th fault characteristic parameter within ; j is the historical standard deviation of the j-th fault characteristic parameter.

[0070] The continuity indicator C of the alarm signal j and the significance index S j A comprehensive evaluation value is obtained through a linear weighted algorithm, and the comprehensive evaluation value is compared with the preset disappearance threshold to determine whether the fault characteristic parameters of the alarm signal have disappeared; if the comprehensive evaluation value is less than the preset disappearance threshold, it is determined that the fault characteristic parameters of the alarm signal have disappeared; if the comprehensive evaluation value is greater than the preset disappearance threshold, it is determined that the fault characteristic parameters of the alarm signal have not disappeared.

[0071] Furthermore, the system further comprises the following steps:

[0072] Obtain historical alarm signals of IoT devices, and analyze the false alarm signal probability and true alarm signal probability of the historical alarm signals through data statistics; combine the fault characteristic parameters of the alarm signal and use the probability density estimation method to construct the conditional probability distribution.

[0073] The false alarm suppression strategy of the Internet of Things device is determined according to the conditional probability distribution; when the false alarm distribution probability in the conditional probability distribution is greater than the preset suppression threshold, the false alarm suppression strategy is triggered and the alarm signal is marked with a false alarm label, otherwise, the delay time threshold is adjusted.

[0074] The delay time threshold is adjusted by a dynamic correction factor, where the dynamic correction factor is the ratio of the false alarm distribution probability to the true alarm distribution probability.

[0075] Furthermore, the system further comprises the following steps:

[0076] The network signal transmission parameters are collected in real time through the network monitoring tool and recorded in time series. The network signal transmission parameters include transmission delay, packet loss rate and jitter value.

[0077] The network signal transmission parameters are applied through a weighted algorithm to obtain the transmission characteristics of the alarm signal under different network conditions. The transmission characteristics are:

[0078] d t =d0+α·q+β·p.

[0079] Where: d t is the comprehensive transmission delay of the signal in the network; α represents the weight factor of the jitter value q on the transmission delay, β represents the weight factor of the packet loss rate p on the transmission delay, α and β are obtained by fitting and analyzing historical data; d0 represents the transmission time required for the signal to be transmitted from the IoT device to the monitoring system.

[0080] Get the standard transmission delay d under ideal network conditions l , calculate the comprehensive transmission delay d t and standard transmission delay d l The delay error E d .

[0081] E d =d t ―d l ; when E d If the error is greater than the preset error threshold, it is determined that the current network status affects the transmission of the alarm signal, and the delay error E d Modify the delay time threshold; when E d If the error is less than the preset threshold, it is determined that the current network status has no effect on the alarm signal transmission.

[0082] A security monitoring time threshold of the Internet of Things device is obtained, and the corrected delay time threshold is less than the security monitoring time threshold.

[0083] Furthermore, the system further comprises the following steps:

[0084] Get the historical monitoring data D of the IoT device monitoring object l ; Obtain the physiological parameters P of the monitored object collected in real time s And form the time series:

[0085] P s ={p t |t=1,2,…,T}.

[0086] p t is the real-time physiological parameter at the acquisition time t; the historical monitoring data D l and physiological parameters P s Joint composition analysis dataset:

[0087] D f ={D l ,P s};

[0088] Through historical monitoring data D l Construct the abnormal state distribution model of the monitored object to obtain the abnormal state P a (x):

[0089]

[0090] Among them, μ a is the mean value of physiological parameters under abnormal conditions in historical monitoring data; is the variance of physiological parameters under abnormal conditions in historical monitoring data.

[0091] The physiological parameter P s With abnormal status P a (x) The deviation degree D is obtained by the deviation degree calculation method p :

[0092]

[0093] Deviation degree D p Indicates the degree of abnormal state of the monitored object; σ a is the standard deviation of physiological parameters in abnormal conditions in historical monitoring data; according to the degree of deviation D p Dynamically adjust the security monitoring time threshold T of IoT devices s :

[0094]

[0095] Among them, T s0 is the initial safety monitoring time threshold; λ is the correction factor used to control the sensitivity of the deviation degree to the adjustment of the safety monitoring time threshold; It depends on the degree of deviation D pAn increasing and decreasing exponential correction term.

[0096] (3) Beneficial effects

[0097] Compared with the prior art, the present invention has the following beneficial effects:

[0098] 1. Analyze the characteristics of false alarm signals through deep learning models and build a feature library, effectively identify and mark false alarm signals through feature matching and cosine similarity calculation. Further combine the time delay verification method to perform dynamic analysis of fault feature parameters, thereby improving the recognition accuracy of false alarm signals and reducing the waste of resources and equipment maintenance costs caused by false alarms.

[0099] 2. Dynamically adjust the safety monitoring time threshold by comprehensively considering the degree of deviation, network transmission delay and equipment operating status to ensure the reliability of alarm signals under complex network and operating conditions. At the same time, flexibly adjust the delay time threshold in combination with the dynamic correction factor to avoid missed reports or false alarms caused by fixed delay time thresholds, thereby improving the real-time and accuracy of equipment fault diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0100] Figure 1 This is a flowchart of an artificial intelligence-based IoT device fault diagnosis method according to Embodiment 1 of the present invention.

[0101] Figure 2 This is a schematic diagram of the module composition of the artificial intelligence-based Internet of Things device fault diagnosis system of Example 2 of the present invention. DETAILED DESCRIPTION

[0102] 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.

[0103] Before giving examples, it is necessary to explain the application scenarios of the present invention. The present invention is an artificial intelligence-based IoT device fault diagnosis method and system. In the intensive care unit (ICU), IoT devices such as vital sign monitors, ventilators, and injection pumps are essential for patient health monitoring. However, the operation of the device in a complex environment may be affected by factors such as sensor failure, network delays, and false alarms, which can easily lead to inaccurate alarm information or delayed response, thereby threatening patient safety.

[0104] Example 1: Figure 1 As shown, this embodiment provides an IoT device fault diagnosis method based on artificial intelligence, and the method includes:

[0105] S1. The operation status data and alarm signals of IoT devices are collected in real time and recorded as the first data. The first data is preprocessed to obtain the second data, and the second data is feature extracted to obtain the characteristic parameter set of the alarm signal, wherein the characteristic parameter set includes frequency change, signal duration, and amplitude abnormal value; the data preprocessing includes noise filtering, missing value filling and standardization processing; the IoT devices in the ICU record real-time operation status data (such as the ECG signal of the ECG monitor, the pressure waveform of the ventilator, the infusion rate of the infusion pump, etc.), and generate alarm signals under abnormal conditions (such as heart rate exceeding the standard, blood oxygen saturation below the safe range). These operation status data and alarm signals are the basis of fault diagnosis. The goal of data preprocessing is to improve data quality and ensure the accuracy and reliability of subsequent analysis. Noise filtering is used to remove random interference in the operation status data, such as signal spikes caused by poor sensor contact or high-frequency noise generated by electromagnetic interference. Missing value filling is due to the possibility of data missing when the device is running, such as intermittent data loss due to network transmission problems. Common methods include mean filling, interpolation or prediction filling based on machine learning. The purpose of standardization is to make data of different dimensions comparable, and to scale the data values ​​to a uniform range at a certain ratio, such as normalizing the values ​​to the interval [0,1]. Perform feature extraction on the second data and analyze the characteristic parameters in the alarm signal, including: Frequency change refers to the abnormal change of the vibration frequency in the signal, such as irregular changes in the heart rate waveform of the ECG signal, which may indicate sensor problems or patient arrhythmia. Signal duration refers to the abnormally short or long duration of some alarm signals, which can help distinguish false alarms from real alarms. For example, a short-term fluctuation in blood oxygen saturation may be a false alarm caused by patient movement. Amplitude abnormality refers to the out-of-range change of the amplitude of the monitoring signal, such as a sudden fluctuation in the infusion speed of an infusion pump that exceeds the set value, which may indicate abnormal operation of the equipment. The parameter set after feature extraction is called the characteristic parameter set of the alarm signal, which is used for subsequent feature matching and fault diagnosis.

[0106] S2. Obtain the historical false alarm signals of IoT devices and perform feature analysis through a deep learning model to extract the features of false alarm signals and generate a false alarm signal feature library; feature match the feature parameter set of the alarm signal with the false alarm signal feature library and calculate the cosine similarity. When the cosine similarity is greater than the preset similarity threshold, the alarm signal is marked with a false alarm signal label, and the corresponding fault feature parameters are extracted from the false alarm signal feature library; IoT devices (such as ECG monitors, ventilators) may generate false alarm signals during operation, such as false alarms caused by patient activity, sensor position movement, or environmental interference. By collecting and storing the data of these false alarm signals and performing feature analysis on the signals through a deep learning model (such as a convolutional neural network, a recurrent neural network), the features of the false alarm signals can be extracted and a false alarm signal feature library can be generated. The false alarm signal feature library is a structured database containing the feature parameters of each false alarm signal, such as frequency features, amplitude features, time features, etc. When the device generates a new alarm signal, its feature parameter set is extracted and matched with the features in the false alarm signal feature library. During the matching process, the cosine similarity between the alarm signal feature parameter set and the features in the library is calculated using the cosine similarity formula to quantify the similarity between the two. If the similarity exceeds the preset similarity threshold (such as 0.9), the alarm signal is determined to be a false alarm signal. When the alarm signal is determined to be a false alarm signal, the system will mark it with a "false alarm signal" label and extract the corresponding fault feature parameters from the feature library (such as amplitude abnormalities caused by sensor movement, frequency mutations caused by environmental interference, etc.), providing a basis for subsequent fault location and equipment maintenance.

[0107] For example, an infusion pump device sends an alarm signal, indicating that the liquid flow rate is abnormal, but preliminary investigation finds that the liquid delivery is normal, which may be a false alarm. The system collects historical false alarm signals and trains a deep learning model to establish a feature library, which contains the following characteristics of false alarm signals: False alarm type 1 (environmental vibration): small frequency change, short signal duration, and amplitude slightly higher than the normal range. False alarm type 2 (sensor position change): sudden change in frequency, long duration, and significantly abnormal amplitude. False alarm type 3 (network interference): no obvious frequency change, but irregular signal duration. By collecting the feature parameter set of the alarm signal and comparing it with the features of the historical false alarm feature library, the cosine similarity is calculated to determine whether it is a false alarm signal, which reduces the interference of false alarm signals to medical staff and provides a detailed fault analysis basis for the operating status of the equipment.

[0108] S3. The alarm signal with the false alarm signal label is judged by the time delay verification method. The time delay verification method is to continuously collect the operating status data of the Internet of Things device within the time delay time threshold and record it as the third data, and combine the fault characteristic parameters with the third data for dynamic trend analysis to determine whether the fault characteristic parameters of the alarm signal disappear. If the fault characteristic parameters of the alarm signal disappear, it is determined to be a false alarm signal, and the alarm signal is removed; if the fault characteristic parameters of the alarm signal do not disappear, it is determined to be a real alarm signal, and fault alarm information is generated and the Internet of Things device maintenance process is triggered. The time delay verification method is a method for dynamically verifying the authenticity of the alarm signal using a time window, i.e., a time delay threshold. By continuously monitoring the operating status data of the device within a period of time (such as 10 seconds, 30 seconds) after the alarm signal is generated, combined with the alarm signal characteristic parameters, it is determined whether the alarm signal disappears, thereby distinguishing between false alarm signals and real alarm signals. Within the time delay threshold, the operating status data of the device is collected, including the core parameters of the device operation, such as sensor voltage, frequency change, operating temperature, etc. Dynamic trend analysis combines fault characteristic parameters (such as frequency changes and amplitude anomalies extracted from the false alarm feature library) with operating status data to analyze whether these characteristics persist or gradually disappear during the delay time. If parameters such as frequency and amplitude gradually return to the normal range (for example, frequency fluctuations decrease and amplitude tends to be stable), it indicates that the alarm signal may be caused by short-term external factors (such as environmental interference or operational errors) and can be determined as a false alarm. If the fault characteristic parameters persist or the trend worsens during the delay time (such as frequency fluctuations intensify and amplitude anomalies expand), it indicates that the alarm signal may be a true reflection of equipment failure. If the fault characteristic parameters disappear, the alarm signal is marked as a false alarm signal and removed to avoid affecting system performance or misleading medical staff. If the fault characteristic parameters do not disappear, it is determined to be a real alarm signal, and the system generates a fault alarm message to prompt medical staff and trigger the equipment maintenance process. Delay verification effectively avoids false alarms caused by short-term interference, reduces the workload of medical staff, and reduces unnecessary equipment maintenance or misoperation.

[0109] For example, in the intensive care unit, a ventilator sends an alarm signal indicating that the airway pressure is too high, but the staff on duty initially found that there was no obvious abnormality in the patient's breathing state and airway. At this time, the system verifies the alarm signal with a delay. The system sets the delay time threshold to 30 seconds, and continuously collects the operating status data of the ventilator during this period, mainly including: airway pressure change curve, airflow rate, equipment operating status (such as compressor operating frequency, valve switch status). The alarm signal fault characteristic parameters are: sudden change in airway pressure frequency: fluctuation range ≥5Hz, rate abnormality: 20% lower than the set value. The dynamic trend analysis is: the airway pressure fluctuates significantly (8Hz) in the first 15 seconds, but gradually returns to the normal range (fluctuation <2Hz) in the next 15 seconds, and the airflow rate has no significant abnormality within 30 seconds, and remains within 5% of the normal value. Combined with the dynamic trend analysis, it is found that the abnormality of airway pressure is a temporary fluctuation, which may be caused by the patient's short cough or environmental interference, rather than equipment failure. The fault characteristic parameters disappear within the delay time threshold, and the alarm signal is removed by the system to avoid misleading medical staff. If the airway pressure continues to fluctuate and the airflow rate is low after 30 seconds, the system will generate a fault alarm message to prompt medical staff and trigger equipment maintenance procedures, such as checking the airway or replacing the ventilator's sensor components.

[0110] The delay verification method is to continuously collect the operating status data of the IoT device within the delay time threshold and record it as the third data, and combine the fault characteristic parameters with the third data for dynamic trend analysis. The method for judging whether the fault characteristic parameters of the alarm signal disappears includes:

[0111] Obtain historical fault data of IoT devices, obtain the fault response time after each fault occurs based on the historical fault data, and calculate the average fault response time and standard deviation σ T , and calculate the delay time threshold T d ; The delay time threshold T d The calculation formula is:

[0112]

[0113] Among them, β is the confidence factor, which is used to determine the sensitivity of the delay time threshold, and the value range of β is 1.5-2.0;

[0114] At the delay time threshold T d The operating status data of IoT devices are collected at a preset sampling frequency and recorded as the delay data set D t , D t ={x1,…,x i ,…,x n}, t∈[0,T d ]; where xi represents the operating status data of the IoT device at the i-th sampling time; n is in [0,T d ]The number of sampling times in the time period;

[0115] The delay data set D t Combine the fault feature parameters of IoT devices to build the fault feature matrix X:

[0116]

[0117] Among them, x i,j is the characteristic value of the jth fault at the i-th sampling moment; i∈[1,n], j∈[1,m]; m represents the number of fault characteristic parameters;

[0118] The alarm signal continuity index C of each fault characteristic parameter within the delay time threshold is calculated through a sliding window j ; The continuity index C j The calculation formula is:

[0119]

[0120] Among them, μ j is the historical average value of each fault characteristic parameter; ∈ j Tolerance interval, indicating the characteristic value x of the current sample i,j The allowable error range; I(·) is the indicator function used to determine the characteristic value x of the current sampling i,j Is it within the tolerance range?

[0121] The significance index S of each fault characteristic parameter is obtained through the significance index calculation formula j , the significance index calculation formula is:

[0122]

[0123] Among them, max(x i,j ) indicates that the delay time threshold T d The maximum sampling characteristic value of the jth fault characteristic parameter, min(x i,j ) indicates that the delay time threshold T d The minimum sampling characteristic value of the jth fault characteristic parameter within σ j is the historical standard deviation of the j-th fault characteristic parameter;

[0124] The continuity indicator C of the alarm signal j and the significance index S jA comprehensive evaluation value is obtained through a linear weighted algorithm, and the comprehensive evaluation value is compared with the preset disappearance threshold to determine whether the fault characteristic parameters of the alarm signal have disappeared; if the comprehensive evaluation value is less than the preset disappearance threshold, it is determined that the fault characteristic parameters of the alarm signal have disappeared; if the comprehensive evaluation value is greater than the preset disappearance threshold, it is determined that the fault characteristic parameters of the alarm signal have not disappeared.

[0125] The continuity index of the alarm signal is used to measure the continuity of the fault characteristic parameters within the delay time threshold. The closer the value is to 1, the more stable the fault characteristic parameters are, the smaller the change is, and the higher the continuity is. On the contrary, if the value is close to 0, it means that the characteristic parameters have a large change and poor continuity. The sample size of the delay data set indicates the number of samples collected within the delay time threshold. For example, if 100 data points are collected, then n = 100. For example, in an IoT device, j may represent current, voltage, or temperature, and i is each specific sampling moment within the delay time threshold. μ j is the historical mean of the fault feature parameter, which is the average value of the fault feature in the historical data. It is usually used to judge the degree of deviation between the current sampled feature value and the historical mean. j The tolerance interval defines an allowable error range, indicating the current sampling feature value x i,j It can be found at the distance from the historical mean μ j The tolerance interval is chosen based on the actual situation. For example, the current may fluctuate within a range of 0.1A and still be considered continuous, so ∈ j It can be taken as ±0.1A. The I(·) indicator function is used to determine whether the current sampling point is within the set tolerance interval. It is defined as follows: If |x i,j ―μ j |<∈ k , then I=1, that is, the current sampled fault characteristic value is within the tolerance interval, indicating that the fault characteristic parameter changes stably and has good continuity. i,j ―μ j |≥∈ k, then I = 0, that is, the current sampling point is outside the tolerance interval, indicating that the characteristic parameter changes greatly and has poor continuity. All sampling moments are judged, and the samples marked as 1 are summed. The sum result indicates how many proportions of fault characteristic values ​​remain within the tolerance interval of the historical mean within a given delay time threshold. Divide the accumulated result (that is, the number of consecutive samples) by the total number of samples to obtain the continuity index of the alarm signal, which reflects the stability of the fault characteristic value. A high significance index indicates that the fault characteristic changes very dramatically in the delay data set, which may indicate that the device is experiencing a more serious fault or abnormal state, so the alarm signal has a higher warning value. A low significance index indicates that the fault characteristic changes relatively smoothly, indicating that the device status is relatively normal and the urgency of the alarm signal is low. According to the continuity index C j and the significance index S j Linear weighted calculation comprehensive evaluation value: C 综合 =α·C j +(1―α)·S j ; C 综合 is a comprehensive evaluation value that takes into account the continuity and significance of the alarm signal. α is a weighting coefficient, ranging from 0 to 1, which is used to adjust the continuity index C j and the significance index S j Impact on the comprehensive evaluation value. If α is large, it means that the continuity index has a higher weight in the comprehensive evaluation value. If α is large, it means that the significance index has a higher weight in the comprehensive evaluation value. The preset disappearance threshold is used to determine whether the fault characteristic parameters of the alarm signal disappear. If the comprehensive evaluation value is less than the disappearance threshold, it is considered that the fault characteristic parameters of the alarm signal have disappeared, which may be a false alarm signal; if the comprehensive evaluation value is greater than the disappearance threshold, it is considered that the fault characteristic parameters of the alarm signal have not disappeared, and the alarm signal is still valid, indicating a real fault signal.

[0126] The method further comprises:

[0127] Obtain historical alarm signals of IoT devices, and analyze the false alarm signal probability and true alarm signal probability of the historical alarm signals through data statistics; combine the fault characteristic parameters of the alarm signals and use the probability density estimation method to construct the conditional probability distribution;

[0128] Determine the false alarm suppression strategy of the IoT device according to the conditional probability distribution; when the false alarm distribution probability in the conditional probability distribution is greater than a preset suppression threshold, trigger the false alarm suppression strategy and mark the alarm signal with a false alarm tag, otherwise, adjust the delay time threshold;

[0129] The delay time threshold is adjusted by a dynamic correction factor, where the dynamic correction factor is the ratio of the false alarm distribution probability to the true alarm distribution probability.

[0130] When the probability of false alarm distribution is greater than the probability of true alarm distribution, the delay threshold is extended according to the dynamic correction factor; when the probability of false alarm distribution is less than the probability of true alarm distribution, the delay threshold is shortened according to the dynamic correction factor. Combined with the fault characteristic parameters of the alarm signal (such as temperature, voltage, current, etc.) and the historical data of false alarm signal / true alarm signal, the conditional probability distribution of the alarm signal is calculated using probability density estimation methods (such as kernel density estimation, histogram estimation, etc.). For example, if the probability of false alarm in the conditional probability distribution is greater than the preset suppression threshold, it is judged that the alarm signal has a high probability of being a false alarm, thereby triggering the suppression strategy and marking the alarm signal as a false alarm signal. After the suppression strategy is triggered, the system can do the following operations: Mark the alarm signal as a false alarm. Send a "false alarm signal" prompt to the monitoring personnel. In subsequent monitoring, the system can take additional measures, such as reducing the priority of the alarm signal, or reducing the response time of the alarm signal. In order to adjust the response mechanism of the alarm signal, the system adjusts the delay threshold according to the ratio of historical false alarm signals to true alarm signals to further optimize the sensitivity and accuracy of the alarm. For example: if the probability of false alarm is high and the dynamic correction factor is large, the delay time threshold can be appropriately increased to extend the observation time of the alarm signal and avoid premature misjudgment of some normal alarm signals. If the probability of false alarm is low and the dynamic correction factor is small, the delay time threshold can be appropriately reduced to improve the system's ability to respond quickly to fault signals.

[0131] The method comprises:

[0132] Using a network monitoring tool to collect network signal transmission parameters in real time and record them in time series, the network signal transmission parameters including transmission delay, packet loss rate and jitter value;

[0133] The network signal transmission parameters are applied through a weighted algorithm to obtain the transmission characteristics of the alarm signal under different network conditions. The transmission characteristics are:

[0134] d t =d0+α·q+β·p;

[0135] Where: d t is the comprehensive transmission delay of the signal in the network; α represents the weight factor of the jitter value q on the transmission delay, β represents the weight factor of the packet loss rate p on the transmission delay, α and β are obtained through historical data fitting analysis; d0 represents the transmission time required for the signal to be transmitted from the IoT device to the monitoring system;

[0136] Get the standard transmission delay d under ideal network conditions l , calculate the comprehensive transmission delay d t and standard transmission delay d l The delay error Ed ;

[0137] E d =d t ―d l ; when E d If the error is greater than the preset error threshold, it is determined that the current network status affects the transmission of the alarm signal, and the delay error E d Modify the delay time threshold; when E d If the error is less than the preset error threshold, it is determined that the current network status has no effect on the transmission of the alarm signal;

[0138] Obtain the security monitoring time threshold of the IoT device, and the corrected delay time threshold is less than the security monitoring time threshold. Use a network monitoring tool (such as Wireshark or a dedicated network monitoring module) to collect the transmission parameters of the IoT alarm signal in the network in real time. The packet loss rate is the proportion of data packets lost during the transmission process, usually expressed as a percentage. The jitter value is the degree of fluctuation of the network delay and reflects the stability of the transmission. The weight factor is obtained by fitting historical data. For example, the impact of packet loss rate and jitter value on delay under different network conditions in the past is analyzed to obtain the best fitting parameter value. In practice, the jitter value has a greater impact on signals with high real-time requirements. Obtaining the standard transmission delay under ideal network conditions is the transmission delay in the absence of jitter and packet loss. According to the delay error E d The delay time threshold is modified by adopting a segmented dynamic adjustment method, for example, the correction factor function f(E d ):

[0139]

[0140] That is, the correction is triggered only when the error exceeds the preset threshold. k is the correction coefficient, which is used to control the correction amplitude and is determined by experimental data analysis. Then the corrected delay time threshold is T d +f(E d ), and the corrected delay time threshold shall not exceed the security monitoring time threshold of the IoT device, so as to ensure that the alarm signal will not exceed the response time limit of the device security monitoring even when the network conditions are poor.

[0141] The method comprises:

[0142] Get the historical monitoring parameters D of the IoT device monitoring object l ; Obtain the physiological parameters P of the monitored object collected in real time s And form the time series:

[0143] P s ={p t |t=1,2,…,T};

[0144] p t is the real-time physiological parameter at the acquisition time t; the historical monitoring data D l and physiological parameters P s Joint composition analysis dataset:

[0145] D f ={D l ,P s};

[0146] Through historical monitoring data D l Construct the abnormal state distribution model of the monitored object to obtain the abnormal state P a (x):

[0147]

[0148] Among them, μ a is the mean value of physiological parameters under abnormal conditions in historical monitoring data; is the variance of physiological parameters under abnormal conditions in historical monitoring data;

[0149] The physiological parameter P s With abnormal status P a (x) The deviation degree D is obtained by the deviation degree calculation method p :

[0150]

[0151] Deviation degree D p Indicates the degree of abnormal state of the monitored object; σ a is the standard deviation of physiological parameters in abnormal conditions in historical monitoring data; according to the degree of deviation D p Dynamically adjust the security monitoring time threshold T of IoT devices s :

[0152]

[0153] Among them, T s0 is the initial safety monitoring time threshold; λ is the correction factor used to control the sensitivity of the deviation degree to the adjustment of the safety monitoring time threshold; It depends on the degree of deviation D pAn exponential correction term that increases and decreases. The historical monitoring data of the monitored object. In this embodiment, the historical monitoring data refers to historical physiological parameters, such as heart rate, blood pressure, etc. The abnormal risk of the current state of the monitored object is determined by calculating the degree of deviation between the real-time physiological parameters and the abnormal state distribution. When the degree of deviation is large, it means that the monitored object has an abnormal risk, and the safety monitoring time threshold is dynamically shortened to ensure the safety of the monitored object. When the degree of deviation is small, it means that the state of the monitored object is relatively normal, and an appropriately longer safety monitoring time threshold is allowed.

[0154] Embodiment 2: Based on the same inventive concept, this embodiment also provides an IoT device fault diagnosis system based on artificial intelligence, the system comprising: a characteristic parameter set acquisition module, an alarm signal analysis module, an alarm signal verification and determination module, and each module is sequentially connected in communication;

[0155] The characteristic parameter set acquisition module is used to collect the operating status data and alarm signals of the Internet of Things devices in real time and record them as first data, perform data preprocessing on the first data to obtain second data, and perform feature extraction on the second data to obtain the characteristic parameter set of the alarm signal, wherein the characteristic parameter set includes frequency change, signal duration, and amplitude abnormal values; the data preprocessing includes noise filtering, missing value filling and standardization processing.

[0156] The alarm signal analysis module is used to obtain historical false alarm signals of IoT devices and perform feature analysis through a deep learning model, extract the features of the false alarm signals and generate a false alarm signal feature library; feature match the feature parameter set of the alarm signal with the false alarm signal feature library and calculate the cosine similarity. When the cosine similarity is greater than a preset similarity threshold, the alarm signal is marked with a false alarm signal label, and the corresponding fault feature parameters are extracted from the false alarm signal feature library.

[0157] The alarm signal verification and judgment module is used to judge the alarm signal with a false alarm signal label through a delay verification method. The delay verification method is to continuously collect the operating status data of the Internet of Things device within the delay time threshold and record it as the third data, and perform dynamic trend analysis on the fault characteristic parameters in combination with the third data to determine whether the fault characteristic parameters of the alarm signal disappear. If the fault characteristic parameters of the alarm signal disappear, it is determined to be a false alarm signal and the alarm signal is removed; if the fault characteristic parameters of the alarm signal do not disappear, it is determined to be a real alarm signal, and fault alarm information is generated and the Internet of Things device maintenance process is triggered.

[0158] The alarm signal verification and determination module comprises the following steps:

[0159] Obtain historical fault data of IoT devices, obtain the fault response time after each fault occurs based on the historical fault data, and calculate the average fault response time and standard deviation σ T , and calculate the delay time threshold T d ; The delay time threshold T d The calculation formula is:

[0160]

[0161] Wherein, β is a confidence factor, which is used to determine the sensitivity of the delay time threshold, and the value range of β is 1.5-2.0.

[0162] At the delay time threshold T d The operating status data of IoT devices are collected at a preset sampling frequency and recorded as the delay data set D t , D t ={x1,…,x i ,…,x n}, t∈[0,T d ]; where x i represents the operating status data of the IoT device at the i-th sampling time; n is in [0,T d ] The number of sampling times within the time period.

[0163] The delay data set D t Combine the fault feature parameters of IoT devices to build the fault feature matrix X:

[0164]

[0165] Among them, x i,j is the characteristic value of the jth fault at the i-th sampling moment; i∈[1,n], j∈[1,m]; m represents the number of fault characteristic parameters.

[0166] The alarm signal continuity index C of each fault characteristic parameter within the delay time threshold is calculated through a sliding window j ; The continuity index C j The calculation formula is:

[0167]

[0168] Among them, μ j is the historical average value of each fault characteristic parameter; ∈ j Tolerance interval, indicating the characteristic value x of the current sample i,j The allowable error range; I(·) is the indicator function used to determine the characteristic value x of the current sampling i,j Is it within the tolerance range?

[0169] The significance index S of each fault characteristic parameter is obtained through the significance index calculation formula j , the significance index calculation formula is:

[0170]

[0171] Among them, max(x i,j ) indicates that the delay time threshold T d The maximum sampling characteristic value of the jth fault characteristic parameter, min(x i,j ) indicates that the delay time threshold T d The minimum sampling characteristic value of the jth fault characteristic parameter within σ j is the historical standard deviation of the j-th fault characteristic parameter.

[0172] The continuity indicator C of the alarm signal j and the significance index S j A comprehensive evaluation value is obtained through a linear weighted algorithm, and the comprehensive evaluation value is compared with the preset disappearance threshold to determine whether the fault characteristic parameters of the alarm signal have disappeared; if the comprehensive evaluation value is less than the preset disappearance threshold, it is determined that the fault characteristic parameters of the alarm signal have disappeared; if the comprehensive evaluation value is greater than the preset disappearance threshold, it is determined that the fault characteristic parameters of the alarm signal have not disappeared.

[0173] The system further comprises the following steps:

[0174] Obtain historical alarm signals of IoT devices, and analyze the false alarm signal probability and true alarm signal probability of the historical alarm signals through data statistics; combine the fault characteristic parameters of the alarm signal and use the probability density estimation method to construct the conditional probability distribution.

[0175] The false alarm suppression strategy of the Internet of Things device is determined according to the conditional probability distribution; when the false alarm distribution probability in the conditional probability distribution is greater than the preset suppression threshold, the false alarm suppression strategy is triggered and the alarm signal is marked with a false alarm label, otherwise, the delay time threshold is adjusted.

[0176] The delay time threshold is adjusted by a dynamic correction factor, where the dynamic correction factor is the ratio of the false alarm distribution probability to the true alarm distribution probability.

[0177] The system further comprises the following steps:

[0178] The network signal transmission parameters are collected in real time through the network monitoring tool and recorded in time series. The network signal transmission parameters include transmission delay, packet loss rate and jitter value.

[0179] The network signal transmission parameters are applied through a weighted algorithm to obtain the transmission characteristics of the alarm signal under different network conditions. The transmission characteristics are:

[0180] d t =d0+α·q+β·p.

[0181] Where: d t is the comprehensive transmission delay of the signal in the network; α represents the weight factor of the jitter value q on the transmission delay, β represents the weight factor of the packet loss rate p on the transmission delay, α and β are obtained by fitting and analyzing historical data; d0 represents the transmission time required for the signal to be transmitted from the IoT device to the monitoring system.

[0182] Get the standard transmission delay d under ideal network conditions l , calculate the comprehensive transmission delay d t and standard transmission delay d l The delay error E d ;

[0183] E d =d t ―d l ; when E d If the error is greater than the preset error threshold, it is determined that the current network status affects the transmission of the alarm signal, and the delay error E d Modify the delay time threshold; when E d If the error is less than the preset threshold, it is determined that the current network status has no effect on the alarm signal transmission.

[0184] A security monitoring time threshold of the Internet of Things device is obtained, and the corrected delay time threshold is less than the security monitoring time threshold.

[0185] The system further comprises the following steps:

[0186] Get the historical monitoring data D of the IoT device monitoring object l ; Obtain the physiological parameters P of the monitored object collected in real time s And form the time series:

[0187] P s ={p t |t=1,2,…,T}.

[0188] p t is the real-time physiological parameter at the acquisition time t; the historical monitoring data D l and physiological parameters P s Joint composition analysis dataset:

[0189] D f ={D l ,P s}.

[0190] Through historical monitoring data D l Construct the abnormal state distribution model of the monitored object to obtain the abnormal state P a (x):

[0191]

[0192] Among them, μ a is the mean value of physiological parameters under abnormal conditions in historical monitoring data; is the variance of physiological parameters under abnormal conditions in historical monitoring data.

[0193] The physiological parameter P s With abnormal status P a (x) The deviation degree D is obtained by the deviation degree calculation method p :

[0194]

[0195] Deviation degree D p Indicates the degree of abnormal state of the monitored object; σ a is the standard deviation of physiological parameters in abnormal conditions in historical monitoring data; according to the degree of deviation D p Dynamically adjust the security monitoring time threshold T of IoT devices s :

[0196]

[0197] Among them, T s0 is the initial safety monitoring time threshold; λ is the correction factor used to control the sensitivity of the deviation degree to the adjustment of the safety monitoring time threshold; It depends on the degree of deviation D p An increasing and decreasing exponential correction term.

[0198] It should be noted that, regarding the system in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0199] Finally, it should be noted that: Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An artificial intelligence-based method for diagnosing faults in IoT devices, characterized in that: The method comprises: The operation status data and alarm signals of the IoT devices are collected in real time and recorded as first data, the first data are preprocessed to obtain second data, and the second data are feature extracted to obtain a feature parameter set of the alarm signal, wherein the feature parameter set includes frequency change, signal duration, and amplitude abnormal value; the data preprocessing includes noise filtering, missing value filling, and standardization processing; Obtain historical false alarm signals of IoT devices and perform feature analysis through a deep learning model, extract features of false alarm signals and generate a false alarm signal feature library; perform feature matching on the feature parameter set of the alarm signal and the false alarm signal feature library and calculate cosine similarity; when the cosine similarity is greater than a preset similarity threshold, mark the alarm signal with a false alarm signal label, and extract corresponding fault feature parameters from the false alarm signal feature library; The alarm signal with a false alarm signal label is judged by a time delay verification method, wherein the time delay verification method continuously collects the operating status data of the Internet of Things device within a time delay time threshold and records it as the third data, and performs dynamic trend analysis on the fault characteristic parameters in combination with the third data to determine whether the fault characteristic parameters of the alarm signal disappear. If the fault characteristic parameters of the alarm signal disappear, it is determined to be a false alarm signal, and the alarm signal is removed; if the fault characteristic parameters of the alarm signal do not disappear, it is determined to be a real alarm signal, and fault alarm information is generated and the Internet of Things device maintenance process is triggered.

2. The method for diagnosing faults of IoT devices based on artificial intelligence according to claim 1, characterized in that: The delay verification method is to continuously collect the operating status data of the IoT device within the delay time threshold and record it as the third data, and combine the fault characteristic parameters with the third data for dynamic trend analysis. The method for judging whether the fault characteristic parameters of the alarm signal disappears includes: Obtain historical fault data of IoT devices, obtain the fault response time after each fault occurs based on the historical fault data, and calculate the average fault response time and standard deviation σ T , and calculate the delay time threshold T d ; The delay time threshold T d The calculation formula is: Among them, β is the confidence factor, which is used to determine the sensitivity of the delay time threshold; At the delay time threshold T d The operating status data of IoT devices are collected at a preset sampling frequency and recorded as the delay data set D t , D t ={x1,…,x i ,…,x n }, t∈[0,T d ]; where x i represents the operating status data of the IoT device at the i-th sampling time; n is in [0,T d ]The number of sampling times in the time period; The delay data set D t Combine the fault feature parameters of IoT devices to build the fault feature matrix X: Among them, x i,j is the characteristic value of the jth fault at the i-th sampling moment; i∈[1,n], j∈[1,m]; m represents the number of fault characteristic parameters; The alarm signal continuity index C of each fault characteristic parameter within the delay time threshold is calculated through a sliding window j ; The continuity index C j The calculation formula is: Among them, μ j is the historical average value of each fault characteristic parameter; ∈ j Tolerance interval, indicating the characteristic value x of the current sample i,j The allowable error range; I(·) is the indicator function used to determine the characteristic value x of the current sampling i,j Is it within the tolerance range? The significance index S of each fault characteristic parameter is obtained through the significance index calculation formula j , the significance index calculation formula is: Among them, max(x i,j ) indicates that the delay time threshold T d The maximum sampling characteristic value of the jth fault characteristic parameter, min(x i,j ) indicates that the delay time threshold T d The minimum sampling characteristic value of the jth fault characteristic parameter within σ j is the historical standard deviation of the j-th fault characteristic parameter; The continuity indicator C of the alarm signal j and the significance index S j A comprehensive evaluation value is obtained through a linear weighted algorithm, and the comprehensive evaluation value is compared with the preset disappearance threshold to determine whether the fault characteristic parameters of the alarm signal have disappeared; if the comprehensive evaluation value is less than the preset disappearance threshold, it is determined that the fault characteristic parameters of the alarm signal have disappeared; if the comprehensive evaluation value is greater than the preset disappearance threshold, it is determined that the fault characteristic parameters of the alarm signal have not disappeared.

3. The method for diagnosing faults of IoT devices based on artificial intelligence according to claim 2, characterized in that: The method further comprises: Obtain historical alarm signals of IoT devices, and analyze the false alarm signal probability and true alarm signal probability of the historical alarm signals through data statistics; combine the fault characteristic parameters of the alarm signals and use the probability density estimation method to construct the conditional probability distribution; Determine the false alarm suppression strategy of the IoT device according to the conditional probability distribution; when the false alarm distribution probability in the conditional probability distribution is greater than a preset suppression threshold, trigger the false alarm suppression strategy and mark the alarm signal with a false alarm tag, otherwise, adjust the delay time threshold; The delay time threshold is adjusted by a dynamic correction factor, where the dynamic correction factor is the ratio of the false alarm distribution probability to the true alarm distribution probability.

4. The method for diagnosing faults of IoT devices based on artificial intelligence according to claim 3 is characterized in that: The method further comprises: Using a network monitoring tool to collect network signal transmission parameters in real time and record them in time series, the network signal transmission parameters including transmission delay, packet loss rate and jitter value; The network signal transmission parameters are used through a weighted algorithm to obtain the transmission characteristics of the alarm signal under different network conditions. The transmission characteristics are: d t =d0+α·q+β·p; Where: d t is the comprehensive transmission delay of the signal in the network; α represents the weight factor of the jitter value q on the transmission delay, β represents the weight factor of the packet loss rate p on the transmission delay, α and β are obtained through historical data fitting analysis; d0 represents the transmission time required for the signal to be transmitted from the IoT device to the monitoring system; Get the standard transmission delay d under ideal network conditions l , calculate the comprehensive transmission delay d t and standard transmission delay d l The delay error E d ; E d =d t ―d l ; when E d If the error is greater than the preset error threshold, it is determined that the current network status affects the transmission of the alarm signal, and the delay error E d Modify the delay time threshold; when E d If the error is less than the preset error threshold, it is determined that the current network status has no effect on the transmission of the alarm signal; A security monitoring time threshold of the Internet of Things device is obtained, and the corrected delay time threshold is less than the security monitoring time threshold.

5. The method for diagnosing faults of IoT devices based on artificial intelligence according to claim 4 is characterized in that: The method further comprises: Get the historical monitoring data D of the IoT device monitoring object l ; Obtain the physiological parameters P of the monitored object collected in real time s And form the time series: P s ={p t ∣t=1,2,…,T}; p t is the real-time physiological parameter at the acquisition time t; the historical monitoring data D l and physiological parameters P s Joint composition analysis dataset: D f ={D l ,P s }; Through historical monitoring data D l Construct the abnormal state distribution model of the monitored object to obtain the abnormal state P a (x): Among them, μ a is the mean value of physiological parameters under abnormal conditions in historical monitoring data; is the variance of physiological parameters under abnormal conditions in historical monitoring data; The physiological parameter P s With abnormal status P a (x) The deviation degree D is obtained by the deviation degree calculation method p : Degree of deviation D p Indicates the degree of abnormal state of the monitored object; σ a is the standard deviation of physiological parameters in abnormal conditions in historical monitoring data; according to the degree of deviation D p Dynamically adjust the security monitoring time threshold T of IoT devices s : Among them, T s0 is the initial safety monitoring time threshold; λ is the correction factor used to control the sensitivity of the deviation degree to the adjustment of the safety monitoring time threshold; It depends on the degree of deviation D p An increasing and decreasing exponential correction term.

6. The IoT equipment fault diagnosis system based on artificial intelligence is characterized by: The system comprises: a characteristic parameter set acquisition module, an alarm signal analysis module, an alarm signal verification and determination module, and each module is sequentially connected in communication; A feature parameter set acquisition module is used to collect the operation status data and alarm signals of the IoT device in real time and record them as first data, perform data preprocessing on the first data to obtain second data, and perform feature extraction on the second data to obtain a feature parameter set of the alarm signal, wherein the feature parameter set includes frequency change, signal duration, and amplitude abnormal value; the data preprocessing includes noise filtering, missing value filling, and standardization processing; The alarm signal analysis module is used to obtain the historical false alarm signals of the IoT device and perform feature analysis through a deep learning model, extract the features of the false alarm signals and generate a false alarm signal feature library; perform feature matching on the feature parameter set of the alarm signal and the false alarm signal feature library and calculate the cosine similarity. When the cosine similarity is greater than a preset similarity threshold, the alarm signal is marked with a false alarm signal label, and the corresponding fault feature parameters are extracted from the false alarm signal feature library; The alarm signal verification and judgment module is used to judge the alarm signal with a false alarm signal label through a delay verification method. The delay verification method is to continuously collect the operating status data of the Internet of Things device within the delay time threshold and record it as the third data, and perform dynamic trend analysis on the fault characteristic parameters in combination with the third data to determine whether the fault characteristic parameters of the alarm signal disappear. If the fault characteristic parameters of the alarm signal disappear, it is determined to be a false alarm signal and the alarm signal is removed; if the fault characteristic parameters of the alarm signal do not disappear, it is determined to be a real alarm signal, and fault alarm information is generated and the Internet of Things device maintenance process is triggered.

7. The artificial intelligence-based IoT device fault diagnosis system according to claim 6 is characterized in that: The alarm signal verification and determination module comprises the following steps: Obtain historical fault data of IoT devices, obtain the fault response time after each fault occurs based on the historical fault data, and calculate the average fault response time and standard deviation σ T , and calculate the delay time threshold T d ; The delay time threshold T d The calculation formula is: Among them, β is the confidence factor, which is used to determine the sensitivity of the delay time threshold; At the delay time threshold T d The operating status data of IoT devices are collected at a preset sampling frequency and recorded as the delay data set D t , D t ={x1,…,x i ,…,x n }, t∈[0,T d ]; where x i represents the operating status data of the IoT device at the i-th sampling time; n is in [0,T d ]The number of sampling times in the time period; The delay data set D t Combine the fault feature parameters of IoT devices to build the fault feature matrix X: Among them, x i,j is the characteristic value of the jth fault at the i-th sampling moment; i∈[1,n], j∈[1,m]; m represents the number of fault characteristic parameters; The alarm signal continuity index C of each fault characteristic parameter within the delay time threshold is calculated through a sliding window j ; The continuity index C j The calculation formula is: Among them, μ j is the historical average value of each fault characteristic parameter; ∈ j Tolerance interval, indicating the characteristic value x of the current sample i,j The allowable error range; I(·) is the indicator function used to determine the characteristic value x of the current sampling i,j Is it within the tolerance range? The significance index S of each fault characteristic parameter is obtained through the significance index calculation formula j , the significance index calculation formula is: Among them, max(x i,j ) indicates that the delay time threshold T d The maximum sampling characteristic value of the jth fault characteristic parameter, min(x i,j ) indicates that the delay time threshold T d The minimum sampling characteristic value of the jth fault characteristic parameter within σ j is the historical standard deviation of the j-th fault characteristic parameter; The continuity indicator C of the alarm signal j and the significance index S j A comprehensive evaluation value is obtained through a linear weighted algorithm, and the comprehensive evaluation value is compared with the preset disappearance threshold to determine whether the fault characteristic parameters of the alarm signal have disappeared; if the comprehensive evaluation value is less than the preset disappearance threshold, it is determined that the fault characteristic parameters of the alarm signal have disappeared; if the comprehensive evaluation value is greater than the preset disappearance threshold, it is determined that the fault characteristic parameters of the alarm signal have not disappeared.

8. The artificial intelligence-based IoT device fault diagnosis system according to claim 7 is characterized in that: The system further comprises the following steps: Obtain historical alarm signals of IoT devices, and analyze the false alarm signal probability and true alarm signal probability of the historical alarm signals through data statistics; combine the fault characteristic parameters of the alarm signals and use the probability density estimation method to construct the conditional probability distribution; Determine the false alarm suppression strategy of the IoT device according to the conditional probability distribution; when the false alarm distribution probability in the conditional probability distribution is greater than a preset suppression threshold, trigger the false alarm suppression strategy and mark the alarm signal with a false alarm tag, otherwise, adjust the delay time threshold; The delay time threshold is adjusted by a dynamic correction factor, where the dynamic correction factor is the ratio of the false alarm distribution probability to the true alarm distribution probability.

9. The artificial intelligence-based IoT device fault diagnosis system according to claim 8, characterized in that: The system further comprises the following steps: Using a network monitoring tool to collect network signal transmission parameters in real time and record them in time series, the network signal transmission parameters including transmission delay, packet loss rate and jitter value; The network signal transmission parameters are used through a weighted algorithm to obtain the transmission characteristics of the alarm signal under different network conditions. The transmission characteristics are: d t =d0+α·q+β·p; Where: d t is the comprehensive transmission delay of the signal in the network; α represents the weight factor of the jitter value q on the transmission delay, β represents the weight factor of the packet loss rate p on the transmission delay, α and β are obtained through historical data fitting analysis; d0 represents the transmission time required for the signal to be transmitted from the IoT device to the monitoring system; Get the standard transmission delay d under ideal network conditions l , calculate the comprehensive transmission delay d t and standard transmission delay d l The delay error E d ; E d =d t -d l ; when E d If the error is greater than the preset error threshold, it is determined that the current network status affects the transmission of the alarm signal, and the delay error E d Modify the delay time threshold; when E d If the error is less than the preset error threshold, it is determined that the current network status has no effect on the transmission of the alarm signal; A security monitoring time threshold of the Internet of Things device is obtained, and the corrected delay time threshold is less than the security monitoring time threshold.

10. The artificial intelligence-based IoT device fault diagnosis system according to claim 9, characterized in that: The system further comprises the following steps: Get the historical monitoring data D of the IoT device monitoring object l ; Obtain the physiological parameters P of the monitored object collected in real time s And form the time series: P s ={p t |t=1,2,...,T}; p t is the real-time physiological parameter at the acquisition time t; the historical monitoring data D l and physiological parameters P s Joint composition analysis dataset: D f ={D l ,P s }; Through historical monitoring data D l Construct the abnormal state distribution model of the monitored object to obtain the abnormal state P a (x): Among them, μ a is the mean value of physiological parameters under abnormal conditions in historical monitoring data; is the variance of physiological parameters under abnormal conditions in historical monitoring data; The physiological parameter P s With abnormal status P a (x) The deviation degree D is obtained by the deviation degree calculation method p : Degree of deviation D p Indicates the degree of abnormal state of the monitored object; σ a is the standard deviation of physiological parameters in abnormal conditions in historical monitoring data; according to the degree of deviation D p Dynamically adjust the security monitoring time threshold T of IoT devices s : Among them, T s0 is the initial safety monitoring time threshold; λ is the correction factor used to control the sensitivity of the deviation degree to the adjustment of the safety monitoring time threshold; It depends on the degree of deviation D p An increasing and decreasing exponential correction term.

Images