Abnormality monitoring method of internet of things device, storage medium and computer program product

Abstract

The application discloses an abnormality monitoring method of an Internet of Things device, a storage medium and a computer program product, relates to the technical field of Internet of Things, and comprises the following steps: acquiring current device state data of the Internet of Things device at a current time; determining a data value range based on the current device state data, and detecting whether the current device state data is within the data value range; if the current device state data is not within the data value range, determining that the Internet of Things device is in an abnormal running state at the current time; determining a current abnormality degree of the Internet of Things device at the current time, determining a target strategy corresponding to the current abnormality degree from preset response strategies corresponding to different abnormality degrees, and controlling the Internet of Things device to run according to the target strategy. The application improves the abnormality detection accuracy of the Internet of Things device and the reliability and efficiency of the whole system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Internet of Things, and particularly relates to an abnormality monitoring method of Internet of Things equipment, a storage medium and a computer program product. BACKGROUND

[0002] With the development of Internet of Things technology, more and more intelligent devices are integrated into smart home, smart city, industrial automation and various application scenarios, for example, in Internet of Things and intelligent systems, in order to better integrate and utilize resources, super devices appear, that is, a plurality of physical devices are abstracted into a virtual whole through software definition, so as to facilitate unified management and scheduling, the state and performance of Internet of Things equipment directly affect the reliability and efficiency of the whole system. At present, the method for detecting device abnormalities of Internet of Things equipment is mostly based on fixed threshold setting, however, due to environmental changes and uncertainty of work load, the fixed threshold may cause a large number of false negatives and false positives, which affects the normal operation of Internet of Things equipment.

[0003] The above content is only used to assist in understanding the technical solutions of the present application, and does not represent the acknowledgement of the above content as prior art. SUMMARY

[0004] The main purpose of the present application is to provide an abnormality monitoring method of Internet of Things equipment, a storage medium and a computer program product, which aims to solve the technical problem of low accuracy of abnormality detection of Internet of Things equipment.

[0005] In order to achieve the above purpose, the present application provides an abnormality monitoring method of Internet of Things equipment, which comprises:

[0006] Obtaining current device state data of Internet of Things equipment at a current time, wherein the device state data is used to represent the running state of the Internet of Things equipment at the current time and the running environment of the Internet of Things equipment at the current time;

[0007] Determining a data value range based on the current device state data, and detecting whether the current device state data is within the data value range;

[0008] If the current device state data is not within the data value range, it is determined that the Internet of Things equipment is in an abnormal running state at the current time;

[0009] Determining a current abnormality degree of the Internet of Things equipment at the current time, and determining a target strategy corresponding to the current abnormality degree from different abnormality degrees corresponding to preset response strategies, and controlling the Internet of Things equipment to run according to the target strategy.

[0010] In an embodiment, before the step of determining the data value range based on the current device state data, the method further comprises:

[0011] obtaining historical device state data before the current device state data;

[0012] The step of determining the data value range based on the current device state data comprises:

[0013] moving a preset moving window by a preset step, and determining reference data of multiple time points from the current device state data and the historical device state data before the current time through the moving window;

[0014] calculating a reference mean and a reference standard deviation of the reference data of each time point;

[0015] obtaining a floating value by weighting the reference standard deviation through a preset adjustment coefficient, taking a difference value between the reference mean and the floating value as a lower limit value of the data value range, taking a sum value between the reference mean and the floating value as an upper limit value of the data value range, and determining the data value range based on the lower limit value and the upper limit value.

[0016] In an embodiment, the step of determining the data value range based on the current device state data comprises:

[0017] inputting the current device state data into a preset threshold output model to obtain the data value range, wherein the threshold output model is obtained by training initial device state data.

[0018] In an embodiment, before the step of inputting the current device state data into a preset threshold output model to obtain the data value range, the method further comprises:

[0019] obtaining historical device state data before the current device state data;

[0020] when a model update instruction is detected, updating the threshold output model through the historical device state data before the current time, and taking the updated threshold output model as a preset threshold output model.

[0021] In an embodiment, after the step of controlling the operation of the Internet of Things device according to the target strategy, the method further comprises:

[0022] detecting whether the Internet of Things device returns to normal;

[0023] if the Internet of Things device does not return to normal, determining a model update instruction;

[0024] The step of updating the threshold output model by the historical device state data before the current time comprises:

[0025] updating the threshold output model based on the current abnormality degree, the current device state data and the historical device state data before the current time.

[0026] In an embodiment, before the step of determining the data value range based on the current device state data, the method further comprises:

[0027] performing denoising processing on the current device state data to obtain denoised state data;

[0028] determining outliers in the denoised state data, and removing the outliers from the denoised state data to obtain abnormality-processed state data;

[0029] performing data smoothing processing on the abnormality-processed state data by a sliding window technique to obtain smoothed state data;

[0030] The step of determining the data value range based on the current device state data comprises:

[0031] determining the data value range based on the smoothed state data.

[0032] In an embodiment, the step of determining the current abnormality degree of the Internet of Things device at the current time comprises:

[0033] obtaining a next-time running state of the Internet of Things device at a next time;

[0034] if the next-time running state is an abnormal running state, updating the next time to the current time, and performing the step of obtaining the next-time running state of the Internet of Things device at a next time;

[0035] if the next-time running state is a normal running state, determining an abnormality duration of the Internet of Things device at the current time, and determining a data deviation degree between the current device state data and the data value range, taking the abnormality duration and the data deviation degree as the current abnormality degree, wherein the abnormality duration is a duration composed of the current time and continuous times before the current time at which the Internet of Things device is in an abnormal running state.

[0036] In an embodiment, the step of controlling the Internet of Things device to run according to the target strategy comprises:

[0037] if the current abnormality degree is a mild abnormality, outputting a warning information by the Internet of Things device;

[0038] if the current abnormality degree is moderate abnormality, reducing a device load of the IoT device;

[0039] if the current abnormality degree is severe abnormality, sending device traffic of the IoT device to a backup device and shutting down the IoT device.

[0040] In addition, to achieve the above object, the present application also provides a storage medium, which is a computer readable storage medium, and a computer program is stored on the storage medium. The computer program is executed by a processor to implement the steps of the abnormality monitoring method of the IoT device.

[0041] In addition, to achieve the above object, the present application also provides a computer program product, which includes a computer program. The computer program is executed by a processor to implement the steps of the abnormality monitoring method of the IoT device.

[0042] In the present application, current device state data of the IoT device at a current time is obtained, wherein the device state data is used to represent the running state of the IoT device at the current time and the running environment of the IoT device at the current time. Based on the current device state data, a data value range is determined, and it is detected whether the current device state data is within the data value range. If the current device state data is not within the data value range, it is determined that the IoT device is in a running abnormal state at the current time. The current abnormality degree of the IoT device at the current time is determined, and a target strategy corresponding to the current abnormality degree is determined from different abnormality degree corresponding preset response strategies. The IoT device is controlled to run according to the target strategy.

[0043] Compared with the traditional fixed threshold method, which is prone to false negatives and false positives when facing environmental changes and workload uncertainty, the present application dynamically determines the data value range based on the current device state data that can reflect the actual situation of the device, rather than relying on fixed thresholds, thereby improving the accuracy of abnormality detection and reducing false negatives and false positives. In addition, after determining that the IoT device is in a running abnormal state, the current abnormality degree is further determined, and a suitable target strategy is selected from the preset response strategies to control the device to run. This refined control process can more effectively cope with different types of abnormalities, ensure that the device resumes normal operation in the shortest possible time, and improve the overall reliability and efficiency of the system. Therefore, the present application improves the accuracy of abnormality detection of the IoT device and the overall reliability and efficiency of the system. BRIEF DESCRIPTION OF DRAWINGS

[0044] The accompanying drawings, which are incorporated herein and constitute part of the specification, illustrate embodiments consistent with the application and, together with the description, further serve to explain the principles of the application.

[0045] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the accompanying drawings required by the embodiments or prior art description will be briefly introduced as follows. Obviously, for those of ordinary skill in the art, the other drawings can also be obtained based on these drawings without any creative effort.

[0046] Figure 1 A flowchart provided for the abnormal monitoring method of the Internet of Things device in Embodiment One of the present application;

[0047] Figure 2 A flowchart provided for the abnormal monitoring method of the Internet of Things device in Embodiment Two of the present application;

[0048] Figure 3 A flowchart provided for the abnormal monitoring method of the Internet of Things device in Embodiment Three of the present application;

[0049] Figure 4 A brief flowchart of the abnormal monitoring method of the Internet of Things device provided for an embodiment of the present application;

[0050] Figure 5 A module structure diagram of the abnormal monitoring device of the Internet of Things device in an embodiment of the present application;

[0051] Figure 6 A device structure diagram of the hardware running environment involved in the abnormal monitoring method of the Internet of Things device in an embodiment of the present application.

[0052] The object implementation, functional features and advantages of the present application will be further explained with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0053] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application, and are not used to limit the present application.

[0054] In order to better understand the technical solutions of the present application, the following will be described in detail in combination with the drawings in the specification and the specific embodiments.

[0055] The main solution of the embodiment of the application is: obtaining current device state data of an Internet of Things device at a current time, wherein the device state data is used to represent the running state of the Internet of Things device at the current time and the running environment of the Internet of Things device at the current time; determining a data value range based on the current device state data, and detecting whether the current device state data is within the data value range; if the current device state data is not within the data value range, determining that the Internet of Things device is in an abnormal running state at the current time; determining a current abnormality degree of the Internet of Things device at the current time, and determining a target strategy corresponding to the current abnormality degree from preset response strategies corresponding to different abnormality degrees, and controlling the Internet of Things device to run according to the target strategy.

[0056] In the embodiment, for convenience of description, the following describes an abnormality monitoring device of an Internet of Things device as an execution subject.

[0057] With the development of Internet of Things technology, more and more intelligent devices are integrated into various application scenarios such as smart home, smart city, and industrial automation. For example, in the Internet of Things and intelligent systems, in order to better integrate and utilize resources, super devices appear, that is, a plurality of physical devices are abstracted into a virtual whole through software definition, so as to facilitate unified management and scheduling. The state and performance of the Internet of Things device directly affect the reliability and efficiency of the entire system. At present, the method for detecting device abnormalities of the Internet of Things device is mostly based on fixed threshold setting. However, due to environmental changes and uncertainty of work load, the fixed threshold may cause a large number of false negatives and false positives, thereby affecting the normal operation of the Internet of Things device.

[0058] The application provides a solution, which dynamically determines a data value range based on current device state data that can reflect the actual situation of the device, instead of relying on a fixed threshold, thereby improving the accuracy of abnormality detection and reducing false negatives and false positives. In addition, after determining that the Internet of Things device is in an abnormal running state, the current abnormality degree is further determined, and a suitable target strategy is selected from preset response strategies to control the device to run. This fine control process can more effectively cope with different types of abnormalities, ensure that the device recovers to normal operation in the shortest possible time, and thus improve the overall reliability and efficiency of the system. Therefore, the application improves the accuracy of abnormality detection of the Internet of Things device and the overall reliability and efficiency of the system.

[0059] It should be noted that the execution subject of the embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, or an abnormality monitoring device of an electronic device or an Internet of Things device capable of realizing the above functions. The following takes the abnormality monitoring device as an example to describe the embodiment and the following embodiments.

[0060] Based on this, the embodiment of the present application provides an abnormality monitoring method of an Internet of Things device, which is described with reference to Figure 1 , Figure 1 The flowchart of the first embodiment of the abnormality monitoring method of the Internet of Things device of the present application is shown in the figure.

[0061] In the embodiment, the abnormality monitoring method of the Internet of Things device includes steps S10-S40:

[0062] Step S10, obtaining current device state data of the Internet of Things device at the current time, wherein the device state data is used to represent the running state of the Internet of Things device at the current time and the running environment of the Internet of Things device at the current time;

[0063] It should be noted that the current device state data refers to the data about the device running state obtained by the sensor or other data acquisition means of the Internet of Things device at a certain specific time point or time period, and the device state data is used to represent the running state of the Internet of Things device at the current time and the running environment of the Internet of Things device at the current time. These data can include physical parameters such as temperature, humidity, pressure, speed, voltage and current of the device, and logical parameters such as running state, working mode and fault information of the device. These data are the basis for device abnormality monitoring and state evaluation.

[0064] It can be understood that by monitoring and analyzing the current device state data in real time, the abnormal condition of the device can be found in time, and decision support for the maintenance and management of the device can be provided.

[0065] Step S20, determining the data value range based on the current device state data, and detecting whether the current device state data is within the data value range;

[0066] It should be noted that the data value range refers to the interval between the maximum value and the minimum value allowed by the Internet of Things device state data under normal running state, which can be determined according to the design specification, running experience and historical data of the device, and is used to judge whether the device state data is within the normal range.

[0067] The system determines a reasonable value range for each state data according to historical data, equipment specifications, environmental requirements and other factors, compares the current equipment state data with the range, judges whether the data is within the normal range, and when the equipment state data exceeds the data value range, it is generally considered that the equipment is in an abnormal state and corresponding measures need to be taken for processing.

[0068] It can be understood that in the specific embodiments, the data value range can be determined based on the current equipment state data in a statistical manner, or a machine learning model can be used to determine the data value range, which is not limited here and can be set according to actual needs.

[0069] Exemplarily, before the step S20 of determining the data value range based on the current equipment state data, the method further includes:

[0070] Step S01, performing denoising processing on the current equipment state data to obtain denoised state data;

[0071] In the abnormal monitoring process of the Internet of Things device, the equipment state data can contain various noises such as sensor errors and data transmission errors. Through denoising processing techniques such as filtering algorithms and data smoothing, these noises can be reduced or eliminated, thereby obtaining more accurate and reliable equipment state data, i.e., denoised state data.

[0072] It can be understood that denoising processing can significantly improve the quality of equipment state data, providing a more accurate basis for subsequent data analysis and anomaly detection. By reducing noise interference, the system can more accurately identify the true state of the equipment, thereby improving the robustness of the entire abnormal monitoring system.

[0073] Step S02, determining an outlier in the denoised state data and removing the outlier from the denoised state data to obtain abnormality-processed state data;

[0074] After denoising processing, some extreme or abnormal data points may still exist in the denoised state data. By identifying and removing these outliers, the data for subsequent analysis is more stable and reliable. Specifically, statistical methods such as threshold judgment and box plot analysis can be used, which are not limited here. By removing outliers, false positives caused by data anomalies can be reduced, improving the accuracy of anomaly detection, and the abnormality-processed state data is more stable, which is beneficial to subsequent data analysis and anomaly detection.

[0075] Step S03, performing data smoothing processing on the abnormality-processed state data through a sliding window technique to obtain smoothed state data;

[0076] The sliding window technique is a commonly used data smoothing method. By performing weighted average or other statistical processing on the data in the window, the fluctuation and noise of the data are reduced. By applying the sliding window technique to smooth the state data after abnormality processing, more smooth and stable state data is obtained. The smoothed data is easier to identify small abnormal changes, thereby improving the sensitivity of abnormality detection.

[0077] In this embodiment, the step S20 of determining the data value range based on the current device state data comprises:

[0078] In step S201, the data value range is determined based on the smoothed state data.

[0079] After obtaining the smoothed state data, the data value range of the device state data is determined using these data for subsequent abnormality detection. It can be understood that the data value range determined based on the smoothed state data is more accurate and reliable, which can reduce the occurrence of false positives and false negatives, and help to develop a more reasonable abnormality monitoring strategy to improve the performance and stability of the entire system.

[0080] It should be noted that through the steps of denoising processing, abnormal value removal, data smoothing, and value range determination, the quality of the device state data and the accuracy of the abnormality detection are significantly improved, which provides a strong guarantee for the healthy operation of the Internet of Things device.

[0081] In step S30, if the current device state data is not within the data value range, it is determined that the Internet of Things device is in an abnormal running state at the current time.

[0082] It should be noted that the abnormal running state refers to a state in which the Internet of Things device cannot work normally or the performance is reduced due to various reasons during operation. The abnormal running state can be manifested as physical parameters of the device exceeding the normal range, logic parameter errors, fault alarms, etc. These abnormal states can cause damage to the device, downtime, or degradation of service quality.

[0083] It can be understood that if the current device state data exceeds the normal range, the system determines that the device is in an abnormal state. This provides a basis for rapid response and ensures the accuracy of abnormality detection.

[0084] In step S40, the current abnormality degree of the Internet of Things device at the current time is determined, and the target strategy corresponding to the current abnormality degree is determined from the preset response strategies corresponding to different abnormality degrees. The Internet of Things device is controlled to operate according to the target strategy.

[0085] It should be noted that the current abnormality degree refers to the degree to which the state data of the Internet of Things device deviates from the normal threshold range or the severity of the abnormal condition at a certain moment or time period, and the current abnormality degree is usually measured by quantitative indicators such as abnormal duration, data deviation degree, etc. These indicators reflect the severity and urgency of the device abnormal condition.

[0086] After confirming the device abnormality, the severity of the abnormality is further determined, and the specific determination of the current abnormality degree is not limited here, which can be set according to the corresponding quantitative indicators. For example, when the quantitative indicator of the current abnormality degree is the abnormal duration, the current abnormality degree can be determined by the range in which the abnormal duration is located.

[0087] According to the severity of the abnormality, a suitable strategy is selected from the preset response strategy to control the operation of the device to ensure stable operation and service quality of the device. These strategies can include alarm, automatic adjustment of device parameters, starting of standby devices, etc. The specific strategy can be set according to actual needs, which is not limited here.

[0088] It can be understood that different response strategies are taken according to different degrees of abnormality to ensure the accuracy and effectiveness of the processing, and automatic processing is realized through the preset response strategy to reduce manual intervention and improve processing efficiency.

[0089] The embodiment is based on the current device state data that can reflect the actual situation of the device to dynamically determine the data value range, rather than relying on fixed thresholds, which improves the accuracy of abnormality detection and reduces false positives and false negatives. In addition, after determining that the Internet of Things device is in an abnormal operating state, the current abnormality degree is further determined, and a suitable target strategy is selected from the preset response strategy to control the operation of the device. This fine control process can more effectively deal with different types of abnormalities and ensure that the device recovers to normal operation in the shortest possible time, thereby improving the overall reliability and efficiency of the system. Therefore, the embodiment improves the accuracy of abnormality detection of the Internet of Things device and the overall reliability and efficiency of the system.

[0090] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as the above-mentioned first embodiment can be referred to the above introduction, and the following will not be repeated. On this basis, please refer to Figure 2 , the step S20: before the step of determining the data value range based on the current device state data, further comprising:

[0091] Step A01, obtaining the historical device state data before the current device state data;

[0092] The historical device state data refers to data records about the running state of the Internet of Things device in the past period of time obtained by the sensor or other data acquisition means. The data records the state change of the device at different time points, including the data of the normal running state and the abnormal state. The historical device state data can be used to analyze the running law, fault mode, performance change trend, etc. of the device, and provide data support for the maintenance and management of the device.

[0093] In this embodiment, the step S20 of determining the data value range based on the current device state data comprises:

[0094] In step S202, a preset moving window is moved by a preset step size, and reference data at multiple time points is determined from the current device state data and the historical device state data before the current time through the moving window.

[0095] The moving window technology is used to slide on the time axis according to the preset step size. In each window, the data at the corresponding time point is extracted from the current device state data and the historical device state data as the reference data, which is used to calculate the mean value and the standard deviation of the subsequent data.

[0096] In step S203, the reference mean value and the reference standard deviation of the reference data at each time point are calculated.

[0097] For the reference data in each moving window, the reference data mean value and the reference standard deviation at each time point are calculated. The mean value reflects the average level of the data, and the standard deviation reflects the dispersion degree of the data.

[0098] In step S204, a floating value is obtained by weighting the reference standard deviation by a preset adjustment coefficient. The difference between the reference mean value and the floating value is taken as the lower limit value of the data value range, and the sum of the reference mean value and the floating value is taken as the upper limit value of the data value range. The data value range is determined based on the lower limit value and the upper limit value.

[0099] A preset adjustment coefficient is used to weight the reference standard deviation to obtain a floating value. The floating value reflects the volatility of the data. The lower limit value of the data value range is obtained by subtracting the floating value from the reference mean value, and the upper limit value of the data value range is obtained by adding the floating value to the reference mean value. Through the calculation of the adjustment coefficient and the floating value, the system can more flexibly adapt to the volatility of different device state data.

[0100] Specifically, the data value range = [μ_w-kσ_w, μ_w+kσ_w], where (μ_w) is the real-time mean value and (σ_w) is the standard deviation, and k is the adjustment coefficient.

[0101] It can be understood that through the introduction of historical data reference, moving window technology, mean and standard deviation calculation, and floating value determination, etc., comprehensive, dynamic and accurate analysis of the device state data is realized. These steps collectively improve the accuracy, flexibility and adaptability of anomaly detection, helping to timely discover and handle device anomalies, and ensuring the stable operation of the Internet of Things system.

[0102] In an implementable embodiment, the step 20 of determining the data value range based on the current device state data comprises:

[0103] In step S205, the current device state data is input into a preset threshold output model to obtain the data value range, wherein the threshold output model is trained based on initial device state data.

[0104] The threshold output model is trained based on the initial device state data of the Internet of Things device, and the model output is the corresponding data value range, i.e. the learning goal of the model is to identify the normal fluctuation range of the device state data. In this embodiment, the model structure and training process of the threshold output model are not limited, and can be set according to actual needs.

[0105] The data value range of the current device state data is determined by the pre-trained threshold output model. When new current device state data is input into the model, the model will output a data value range based on its learned knowledge, which represents the expected fluctuation of the device state data under normal circumstances. It can be understood that by introducing a machine learning model, the system can automatically determine the data value range based on the device state data, without the need for manual setting of fixed thresholds, improving the automation and intelligence level of anomaly monitoring. Since the model is trained based on the historical state data of the device, it can adapt to the state changes of the device under different workloads and environmental conditions, i.e. the model can dynamically adjust the data value range to reflect the current actual state of the device, thereby improving the accuracy and flexibility of anomaly detection.

[0106] In an implementable embodiment, before the step S205 of inputting the current device state data into a preset threshold output model to obtain the data value range, the method further comprises:

[0107] Step B01: obtaining historical device state data before the current device state data;

[0108] The historical state data of the Internet of Things device before the current time is retrieved and obtained from a storage device or data warehouse, which records the running state, performance parameters, environmental indicators and other key information of the device in the past period of time, and is the basis for subsequent model training and updating.

[0109] By continuously accumulating historical data, the system can better adapt to the state changes of the device under different environments and conditions, providing rich historical data support for subsequent model training, updating and anomaly detection, ensuring that the model can learn the normal operation mode and characteristics of the device, and improving the accuracy and adaptability of anomaly detection.

[0110] Step B02, when detecting the trigger model update instruction, updating the threshold output model by using the historical device state data before the current time, and using the updated threshold output model as the preset threshold output model.

[0111] It should be noted that the model update instruction can be triggered by monitoring specific trigger conditions (such as significant changes in device state data, model performance degradation, etc.). When the trigger model update instruction is detected, the threshold output model is updated using historical device state data before the current time. The update process may include retraining the model, adjusting the model parameters or updating the model structure, etc. to ensure that the model can continuously reflect the latest state and data distribution of the device. The updated model will replace the original preset model and become the new anomaly detection benchmark.

[0112] By regularly or on-demand updating the model, the model can continuously adapt to changes in device state, improving the accuracy and efficiency of anomaly detection, helping to improve the generalization ability and robustness of the model, reducing false positives and false negatives, and improving the performance of the entire anomaly monitoring system. The anomaly monitoring system of the Internet of Things device can run stably for a long time, providing strong support for the health management and maintenance of the device.

[0113] In a feasible implementation, the step S40 of controlling the operation of the Internet of Things device according to the target strategy further comprises:

[0114] Step C01, detecting whether the Internet of Things device has returned to normal;

[0115] By continuously monitoring the device state data, it can be determined whether the device has returned to normal from the abnormal state. Specifically, it can be determined whether the device has returned to normal by real-time monitoring and analyzing the device performance parameters, running state and environmental indicators. The specific process is not limited here. By real-time detection of device state, the system can timely find out whether the device has returned to normal, so as to take corresponding measures to avoid the device being in abnormal state for a long time.

[0116] Step C02, if the Internet of Things device has not returned to normal, determining to trigger a model update instruction;

[0117] If the IoT device does not recover, it means that the current threshold output model may not accurately reflect the latest state or data distribution of the device, and therefore the model needs to be updated to improve the accuracy of anomaly detection. By updating the model, the system can more accurately reflect the latest state of the device, improve the accuracy and adaptability of anomaly detection, and help reduce false positives and false negatives caused by outdated or inaccurate models, improving the performance of the entire anomaly monitoring system.

[0118] The step B02 of updating the threshold output model based on the historical device state data before the current time comprises:

[0119] Step B021, updating the threshold output model based on the current anomaly degree, the current device state data and the historical device state data before the current time.

[0120] After triggering the model update instruction, the system updates the threshold output model based on the current anomaly degree, the current device state data and the historical device state data, re-trains the model, adjusts the model parameters or updates the model structure, to ensure that the model can more accurately reflect the latest state and data distribution of the device.

[0121] It can be understood that by comprehensively considering the current anomaly degree, the current device state data and the historical data, the system can more accurately update the model, continuously learn and adapt to the changes in the device state, improve the accuracy and generalization ability of the model, and continuously optimize the performance of the anomaly monitoring system.

[0122] Exemplarily, the process of determining the data value range by the model and performing device anomaly detection can be: obtaining current device state data of the Internet of Things device at the current time; inputting the current device state data into the preset threshold output model to obtain the data value range; detecting whether the current device state data is within the data value range; if the current device state data is not within the data value range, determining that the Internet of Things device is in an abnormal running state at the current time; determining the current abnormality degree of the Internet of Things device, determining the target strategy corresponding to the current abnormality degree according to the preset response strategy corresponding to different abnormality degrees, and controlling the Internet of Things device to run according to the strategy; after executing the target strategy, detecting whether the Internet of Things device has recovered to the normal running state; if the Internet of Things device has not recovered to the normal state, determining a model update instruction, specifically, obtaining historical device state data before the current time, updating the threshold output model based on the current abnormality degree, the current device state data and the historical device state data, retraining or adjusting the model parameters using new data to improve the accuracy and adaptability of the model; the updated threshold output model will be used as the preset model for the next time of anomaly monitoring, and continue to participate in the determination of the value range of the device state data and the anomaly detection process. Exemplarily, taking the smart air conditioner in the smart home as an example, the process of determining the data value range by the model and performing device anomaly detection can be: the built-in sensors of the smart air conditioner monitor the indoor temperature, humidity, air conditioner running power and other parameters in real time, and transmit these data to the central control system through the Internet of Things technology; the central control system uses the preset threshold output model to determine the normal value range of the indoor temperature, humidity and other parameters according to the initial state data and historical running data of the smart air conditioner; if the indoor temperature suddenly rises or falls, which exceeds the preset normal range, the system will detect the anomaly; the system comprehensively evaluates the current abnormality degree of the smart air conditioner according to the degree of deviation of the indoor temperature from the normal range, the duration of the anomaly and other factors; according to the abnormality degree, the system can choose to send warning information to the user, automatically adjust the air conditioner power to try to restore the normal state or shut down the air conditioner to prevent further damage; after executing the target strategy, the system continues to monitor the state of the smart air conditioner to determine whether it has recovered to the normal running state; if the smart air conditioner has not recovered to the normal state, a model update instruction is triggered, and the threshold output model is retrained using the latest device state data and historical data to improve the detection ability of the smart air conditioner in the abnormal state.

[0123] It can be understood that, by constructing the closed-loop process of device state monitoring, model updating and optimization in the anomaly monitoring process of the Internet of Things device, through continuous monitoring and model updating, the healthy running of the Internet of Things device and the long-term stability of the system are ensured.

[0124] Based on the first and / or second embodiments of the present application, in the third embodiment of the present application, the same or similar contents as the above first and / or second embodiments can be referred to the above description, and the subsequent description will not be repeated. On this basis, please refer to Figure 3 The step S40 of determining the current abnormality degree of the Internet of Things device at the current time comprises:

[0125] Step S401, obtaining the next time running state of the Internet of Things device at the next time;

[0126] The next time running state refers to the expected running state of the Internet of Things device at a future time point or time period, which is usually predicted and estimated according to the current device state data, historical data, running law and other factors.

[0127] Obtaining the next time running state of the Internet of Things device in real time or periodically can monitor the running state of the device in real time, discover abnormalities in time, and dynamically track the changes of the device state, providing a basis for subsequent abnormality degree evaluation and strategy making.

[0128] Step S402, if the next time running state is an abnormal running state, updating the next time to the current time, and executing the step of obtaining the next time running state of the Internet of Things device at the next time;

[0129] When it is detected that the device is still in an abnormal running state at the next time, the next time is updated to the current time, and the state of the device at the next time point is continued to be monitored. Through recursive monitoring, the system can continuously track the abnormal state of the device, and can accumulate the duration of the abnormal running state, which is used for subsequent judgment of the abnormality degree.

[0130] Step S403, if the next time running state is a normal running state, determining the abnormal duration of the Internet of Things device at the current time, and determining the data deviation degree between the current device state data and the data value range, taking the abnormal duration and the data deviation degree as the current abnormality degree, wherein the abnormal duration is the duration of the current time and the continuous time before the Internet of Things device is in an abnormal running state.

[0131] It should be noted that the abnormal duration refers to the continuous time length from the beginning of the device entering the abnormal state to the recovery of the normal state, which reflects the persistence and influence range of the abnormal event. The data deviation degree refers to the deviation degree between the current device state data and its normal value range, which provides a direct measurement of the severity of the abnormality by quantifying the data difference.

[0132] When the system detects that the device resumes normal operation at the next time, it calculates the duration of the abnormality of the device in the current abnormal event (i.e. the length of time from the start of the abnormality of the device to the resumption of normal operation), and calculates the data deviation of the current device state data relative to the normal value range. The data deviation and the duration of the abnormality are used as quantitative representations of the current abnormality degree. Through the two indicators of the duration of the abnormality and the data deviation, the severity and the influence range of the abnormal event are quantitatively evaluated, and the abnormality processing strategy is more accurately formulated, such as issuing an alarm, triggering a repair process, etc., thereby improving the efficiency and accuracy of abnormality processing and providing data support for subsequent decision-making and strategy formulation.

[0133] For example, the process of determining the current abnormality degree and determining the operation strategy can be: obtaining the current device state data of the Internet of Things device at the current time; determining the data value range and detecting whether the current device state data is within the data value range; if the current device state data is within the data value range, the device is considered to be in a normal operation state and does not require further processing; if the current device state data is not within the data value range, it is determined that the device is in an abnormal operation state, the current abnormality degree of the Internet of Things device at the current time is determined, and the process of specifically determining the current abnormality degree can be: obtaining the next time operation state of the Internet of Things device at the next time, if the next time operation state is still abnormal, updating the current time to the next time, and repeating the step of obtaining the next time operation state, forming a recursion, when it is detected that the next time operation state is normal, calculating the duration of the abnormality from the current time to the end of the abnormality (i.e. the next time is normal), and calculating the data deviation between the current device state data and the data value range, the duration of the abnormality and the data deviation are used as indicators of the current abnormality degree; according to the current abnormality degree of the device, a suitable strategy is selected from the pre-set response strategy, for example, if the abnormality degree of the device is high, the device operation needs to be stopped immediately and an alarm needs to be sent to the maintenance personnel.

[0134] It can be understood that through the steps of real-time acquisition of device state, recursive monitoring of abnormal events, and quantitative evaluation of abnormality degree, accurate monitoring and evaluation of abnormal events of Internet of Things devices are realized, which provides strong support for the healthy operation and abnormality processing of the devices.

[0135] In an embodiment, the step S40 of controlling the operation of the Internet of Things device according to the target strategy comprises:

[0136] If the current abnormality degree is mild, the step S404 of outputting a warning message through the Internet of Things device is performed.

[0137] In the abnormality monitoring process of the Internet of Things device, when it is detected that the current abnormality degree of the device is mild abnormality, pre-warning information is output through the Internet of Things device (such as a display screen, an indicator light, or a sound alarm, etc.). These pre-warning information usually includes the abnormality type, the abnormality level, the occurrence time, and other key information, so that the operating personnel can quickly understand the abnormality condition of the device and take corresponding measures.

[0138] Step S405, if the current abnormality degree is moderate abnormality, the device load of the Internet of Things device is reduced;

[0139] When the current abnormality degree of the Internet of Things device reaches moderate, the device load of the Internet of Things device is reduced to prevent the abnormality condition from further deteriorating. The specific way can be reducing the operation amount of the device, reducing the data transmission rate, turning off part of the non-core functions, etc., which is not limited here and can be set according to actual needs. It can be understood that by reducing the device load, the energy consumption and wear of the device can be reduced, timely reducing the burden of the device helps to prevent the abnormality condition from further deteriorating, reduces the risk of device failure, and prolongs the service life of the device.

[0140] Step S406, if the current abnormality degree is severe abnormality, the device service of the Internet of Things device is sent to a standby device, and the Internet of Things device is closed.

[0141] When the current abnormality degree of the Internet of Things device reaches severe, the service of the Internet of Things device is sent to a standby device, and the current device is closed to ensure the safe and stable operation of the whole Internet of Things and ensure the continuity and stability of the service. It can specifically include real-time migration of data, smooth switching of services, and other complex operations, which are not limited here and can be set according to actual needs. It can be understood that by sending the service to a standby device, the continuity and stability of the service can be ensured, and the user can avoid any interruption or impact. Timely closing of the abnormal device can prevent further damage to the device and reduce maintenance costs and downtime.

[0142] By taking different control measures according to the abnormality degree of the device, both timely warning and response of the device are ensured, and further deterioration of the abnormality condition is prevented through measures such as reducing the load and service migration, thereby improving the reliability and stability of the whole Internet of Things system.

[0143] Exemplarily, in order to help understand the implementation process of the abnormality monitoring method of the Internet of Things device obtained after combining the above-mentioned embodiment one, embodiment two, and embodiment three, please refer to Figure 4 , Figure 4 A brief flowchart of an abnormality monitoring method of an Internet of Things device is provided, specifically:

[0144] 1. Data acquisition and data preprocessing:

[0145] 1.1 Device state data collection, deploy sensors and monitoring systems to collect multi-dimensional state data of the device in real time. These data include but are not limited to hardware indicators (such as CPU utilization, memory usage, temperature, power consumption), network indicators (such as bandwidth utilization, delay, packet loss rate) and environmental indicators (such as temperature, humidity, vibration). Set the data collection frequency to ensure the real-time and accuracy of the data, for example, collect data every second (i.e. obtain the current device state data of the Internet of Things device at the current time).

[0146] 1.2 Data cleaning and denoising, use Kalman filter or moving average method to denoise the data, eliminate the influence of random fluctuations, identify and process abnormal values in the data, such as data points outside the reasonable range or abnormal spikes caused by collection errors (i.e. denoising the current device state data to obtain denoised state data; determining the abnormal values in the denoised state data, and removing the abnormal values from the denoised state data to obtain the state data after abnormality processing).

[0147] 1.3 Data smoothing processing, apply sliding window technology to smooth the data. Choose an appropriate window size (e.g. the last 10 data points), calculate the mean value in the window to eliminate the influence of short-term fluctuations on subsequent threshold calculation (i.e. smooth the abnormality processing state data by sliding window technology to obtain smoothed state data).

[0148] 2. Calculation and adjustment of dynamic threshold [Lower_t, Upper_t] (i.e. determine the data value range based on the current device state data):

[0149] 2.1 Determine the dynamic threshold range (i.e. data value range) based on statistical methods. Set a moving window, the window size (W) can be set to the last 100 data points, calculate the real-time mean (μ_w) and standard deviation (σ_w) of the data in the window, then determine the dynamic adjustment threshold range based on the real-time mean and standard deviation, dynamic threshold range = [μ_w-kσ_w, μ_w+kσ_w], where k is the adjustment coefficient (i.e.), Where W is the size of the moving window, μ_w and σ_w are the mean and standard deviation of the data in the window, i represents the i-th data point, and t represents the current time.

[0150] 2.2 Determine the dynamic threshold range based on the machine learning model. Train a machine learning model (such as linear regression, support vector machine SVM or neural network) with the real-time state features of the device (such as CPU utilization, temperature, bandwidth utilization, etc.) as input and the dynamic threshold range as output: [Lower_t, Upper_t] = f(x_t), where x_t is the state feature vector at the current time, and update the model parameters periodically or based on real-time data to adapt to changes in device state.

[0151] 3. Abnormality detection and response:

[0152] 3.1 Abnormality detection (i.e., if the current device state data is not within the data value range, it is determined that the IoT device is in an abnormal state at the current time). For each time point (t), compare the device state data (x_t) with the dynamic threshold range. If x_t is not within the range [Lower_t, Upper_t], it is determined to be abnormal, and the time, duration, deviation degree, etc. of the abnormal event are recorded for subsequent analysis and response. According to the severity, duration and impact of the abnormality, the abnormal event is classified as mild, moderate or severe, and each abnormal event is prioritized to determine the urgency of response and resource scheduling strategy.

[0153] 3.2 Response strategy execution (i.e., determine the current abnormality degree of the IoT device at the current time, and determine the target strategy corresponding to the current abnormality degree from the pre-defined response strategies corresponding to different abnormality degrees, and control the operation of the IoT device according to the target strategy). For mild abnormalities, the system records abnormal logs and reminds the operation and maintenance personnel to check through the warning mechanism. Continue to monitor the device state, if the abnormality continues or worsens, upgrade the response level. For moderate abnormalities, the system automatically executes pre-defined countermeasures, such as reducing the task load of the device, reducing the running frequency of the device or adjusting the cooling system of the device, sending a notification to the operation and maintenance personnel, and suggesting to check or maintain the device. For severe abnormalities, immediately execute emergency measures such as automatic shutdown, switching to backup device or starting redundant system to prevent device damage or system crash, trigger emergency plan, start fault recovery process to ensure system stability and business continuity.

[0154] 4. The system continuously monitors the effect of the response measures. If the device state returns to normal after the response, confirm the effectiveness of the current threshold setting, if the abnormal event occurs frequently, adjust the threshold range or optimize the detection model to reduce false positives or false negatives.

[0155] For example, the threshold adjustment strategy and response mechanism are optimized using reinforcement learning algorithms such as Q-learning, learning the optimal threshold setting and response measures through reward and punishment mechanisms, and the system automatically updates the model parameters as the device ages or the environment changes, ensuring detection accuracy and response efficiency. For example, long-term data analysis, retrospective analysis of historical abnormal data, identification of long-term trends and device performance changes, updating of device threshold output models, and adjustment of threshold setting strategies based on analysis results to ensure that the system always maintains optimal detection capabilities throughout the device's life cycle.

[0156] For example, the adjustment coefficient in the dynamic threshold range is adjusted according to the current abnormality degree.

[0157] In this embodiment, for devices that perform abnormality detection for the first time, initial thresholds (Lower, Upper) are set, where Lower = μ - kσ and Upper = μ + kσ, μ is the mean of historical data of the same type of device, σ is the standard deviation of historical data of the same type of device, and k is the adjustment coefficient.

[0158] This embodiment can dynamically adjust the threshold range in device state monitoring, accurately detect abnormal events, and ensure stable operation of the system through adaptive response mechanisms, not only improving the accuracy and response speed of abnormality detection, but also continuously optimizing detection strategies through adaptive learning to adapt to long-term changes in device state, suitable for a wide range of Internet of Things and intelligent system scenarios, significantly improving the efficiency of device management.

[0159] It should be noted that the above examples are only for understanding the present application and do not constitute a limitation on the abnormality monitoring method of the Internet of Things device based on this technical concept, and more forms of simple transformation are within the protection scope of the present application.

[0160] The present application also provides an abnormality monitoring device for an Internet of Things device, please refer to Figure 5 The abnormality monitoring device for the Internet of Things device comprises:

[0161] The acquisition module 10 is configured to acquire current device state data of the Internet of Things device at a current time, wherein the device state data is used to represent the running state of the Internet of Things device at the current time and the running environment of the Internet of Things device at the current time.

[0162] The detection module 20 is configured to determine a data value range based on the current device state data and detect whether the current device state data is within the data value range.

[0163] The determination module 30 is configured to determine that the Internet of Things device is in a running abnormal state at the current time if the current device state data is not within the data value range.

[0164] The execution module 40 is configured to determine a current abnormality degree of the IoT device at the current time, determine a target strategy corresponding to the current abnormality degree from preset response strategies corresponding to different abnormality degrees, and control the IoT device to operate according to the target strategy.

[0165] Optionally, the abnormality monitoring apparatus of the IoT device further comprises:

[0166] acquire historical device state data before the current device state data;

[0167] The detection module 20 is configured to:

[0168] move a preset moving window by a preset step, and determine reference data of multiple time points from the current device state data and the historical device state data before the current time by using the moving window;

[0169] calculate a reference mean value and a reference standard deviation of the reference data of each time point;

[0170] weight the reference standard deviation by a preset adjustment coefficient to obtain a floating value, use a difference between the reference mean value and the floating value as a lower limit value of a data value range, use a sum between the reference mean value and the floating value as an upper limit value of the data value range, and determine the data value range based on the lower limit value and the upper limit value.

[0171] Optionally, the detection module 20 is configured to:

[0172] input the current device state data into a preset threshold output model to obtain the data value range, wherein the threshold output model is trained based on initial device state data.

[0173] Optionally, the apparatus further comprises:

[0174] acquire historical device state data before the current device state data;

[0175] when a model update instruction is detected, update the threshold output model by using the historical device state data before the current time, and use the updated threshold output model as a preset threshold output model.

[0176] Optionally, the apparatus further comprises:

[0177] detect whether the IoT device returns to normal;

[0178] if the IoT device does not return to normal, determine a model update instruction;

[0179] The step of updating the threshold output model by the historical device state data before the current time point comprises:

[0180] updating the threshold output model based on the current abnormality degree, the current device state data and the historical device state data before the current time point.

[0181] Optionally, the apparatus is further configured to:

[0182] performing denoising processing on the current device state data to obtain denoised state data;

[0183] determining outliers in the denoised state data and removing the outliers from the denoised state data to obtain abnormality-processed state data;

[0184] performing data smoothing processing on the abnormality-processed state data by a sliding window technique to obtain smoothed state data;

[0185] The step of determining the data value range based on the current device state data comprises:

[0186] determining the data value range based on the smoothed state data.

[0187] Optionally, the execution module 40 is configured to:

[0188] obtain a next time point running state of the Internet of Things device at a next time point;

[0189] if the next time point running state is an abnormal running state, update the next time point to the current time point and perform the step of obtaining the next time point running state of the Internet of Things device at the next time point;

[0190] if the next time point running state is a normal running state, determine an abnormality duration of the Internet of Things device at the current time point and determine a data deviation degree between the current device state data and the data value range, and take the abnormality duration and the data deviation degree as a current abnormality degree, wherein the abnormality duration is a duration of the current time point and continuous time points before the current time point at which the Internet of Things device is in an abnormal running state.

[0191] Optionally, the execution module 40 is configured to:

[0192] if the current abnormality degree is a mild abnormality, output a warning information by the Internet of Things device;

[0193] if the current abnormality degree is a moderate abnormality, reduce a device load of the Internet of Things device;

[0194] If the current abnormality degree is a severe abnormality, device service of the IoT device is sent to a backup device, and the IoT device is closed.

[0195] The abnormality monitoring device of the IoT device provided in the present application adopts the abnormality monitoring method of the IoT device in the above embodiments, and can solve the technical problem of low accuracy of abnormality detection of the IoT device. Compared with the prior art, the abnormality monitoring device of the IoT device provided in the present application has the same beneficial effects as the abnormality monitoring method of the IoT device provided in the above embodiments, and other technical features in the abnormality monitoring device of the IoT device are the same as the features disclosed in the above embodiments, which will not be repeated here.

[0196] The present application provides an abnormality monitoring device of an IoT device, which comprises at least one processor and a memory in communication connection with the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the abnormality monitoring method of the IoT device in Embodiment I.

[0197] Reference will be made to the following description Figure 6 which shows a structural schematic diagram of an abnormality monitoring device of an IoT device suitable for use to implement the embodiments of the present application. The abnormality monitoring device of the IoT device in the embodiments of the present application can include but is not limited to mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (Personal Digital Assistant), PADs (Portable Application Description), PMPs (Portable Media Player), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), and the like, and fixed terminals such as digital TVs, desktop computers, and the like. Figure 6 The abnormality monitoring device of the IoT device shown is only an example, and should not bring any limitation to the functions and use range of the embodiments of the present application.

[0198] As Figure 6As shown, the abnormality monitoring device of the Internet of Things device can include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.) that can perform various appropriate actions and processes according to programs stored in a read-only memory (ROM) 1002 or programs loaded from a storage device 1003 into a random access memory (RAM) 1004. In the RAM 1004, various programs and data required for the operation of the abnormality monitoring device of the Internet of Things device are also stored. The processing device 1001, the ROM 1002, and the RAM 1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; the storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the abnormality monitoring device of the Internet of Things device to communicate wirelessly or wired with other devices to exchange data. Although the abnormality monitoring device of the Internet of Things device with various systems is shown in the figure, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be alternatively implemented or provided.

[0199] In particular, according to embodiments of the present disclosure, the processes described above with reference to the flowcharts can be implemented as a computer software program. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a computer readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network through the communication device, or installed from the storage device 1003, or installed from the ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the methods of the embodiments of the present disclosure are performed.

[0200] The abnormality monitoring device of the Internet of Things device provided by the present application adopts the abnormality monitoring method of the Internet of Things device in the above-mentioned embodiments, and can solve the technical problem of low accuracy of abnormality detection of the Internet of Things device. Compared with the prior art, the beneficial effects of the abnormality monitoring device of the Internet of Things device provided by the present application are the same as those of the abnormality monitoring method of the Internet of Things device provided by the above-mentioned embodiments, and other technical features in the abnormality monitoring device of the Internet of Things device are the same as those disclosed in the previous embodiment method, which will not be repeated here.

[0201] It should be understood that various parts of the present application can be realized by hardware, software, firmware or a combination thereof. In the above description of embodiments, specific features, structures, materials or characteristics can be combined in any appropriate manner in any one or more embodiments or examples.

[0202] The above description is merely illustrative of the application and is not intended to limit the scope of the application. Any variations and modifications that can be made by any person skilled in the art within the spirit and scope of the application are intended to be encompassed by the application. Therefore, the scope of the application should be determined by the scope of the claims.

[0203] The present application provides a computer readable storage medium having stored thereon computer readable program instructions (i.e., a computer program) for performing the abnormality monitoring method of the Internet of Things device in the above-described embodiments.

[0204] The computer readable storage medium provided by the present application may, for example, be a U disk, but is not limited to an electric, magnetic, optical, electromagnetic, infrared, or semiconductor system or device, or any combination thereof. More specific examples of the computer readable storage medium can include, but are not limited to, an electric connection having one or more conductive wires, a portable computer disk, a hard disk, a random access memory (RAM), a read only memory (ROM), an erasable programmable read only memory (EPROM or flash memory), an optical fiber, a portable compact disk read only memory (CD-ROM), an optical storage device, a magnetic storage device, or any appropriate combination thereof. In the present embodiment, the computer readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system or device. The program code contained on the computer readable storage medium can be transmitted by any appropriate medium, including but not limited to an electric wire, an optical cable, an RF (Radio Frequency), etc., or any appropriate combination thereof.

[0205] The above-described computer readable storage medium can be contained in the abnormality monitoring device of the Internet of Things device; or can exist separately and not be assembled into the abnormality monitoring device of the Internet of Things device.

[0206] The computer readable storage medium described above carries one or more programs, when the one or more programs are executed by the abnormality monitoring device of the Internet of Things device, the abnormality monitoring device of the Internet of Things device: obtains current device state data of the Internet of Things device at a current time, wherein the device state data is used to represent the running state of the Internet of Things device at the current time and the running environment of the Internet of Things device at the current time; determines a data value range based on the current device state data, and detects whether the current device state data is within the data value range; if the current device state data is not within the data value range, it is determined that the Internet of Things device is in an abnormal running state at the current time; determine the current abnormality degree of the Internet of Things device at the current time, and determine the target strategy corresponding to the current abnormality degree from the different abnormality degree corresponding preset response strategy, control the Internet of Things device to run according to the target strategy.

[0207] Computer program code for carrying out operations of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0208] The flowcharts and block diagrams in the drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowcharts or block diagrams can represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and / or flowchart illustrations, and combinations thereof, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or combinations of hardware and software.

[0209] The modules involved in the embodiments of the present application can be implemented in software or in hardware. In some cases, the names of the modules do not limit the modules themselves.

[0210] The readable storage medium provided by the present application is a computer readable storage medium, which stores computer readable program instructions (i.e., a computer program) for executing the above-mentioned abnormal monitoring method of the Internet of Things device, and can solve the technical problem of low accuracy of abnormal detection of the Internet of Things device. Compared with the prior art, the computer readable storage medium provided by the present application has the same beneficial effects as the abnormal monitoring method of the Internet of Things device provided by the above-mentioned embodiments, which will not be repeated here.

[0211] The present application also provides a computer program product comprising a computer program, which, when executed by a processor, implements the steps of the above-mentioned abnormal monitoring method of the Internet of Things device.

[0212] The computer program product provided by the present application can solve the technical problem of low accuracy of abnormal detection of the Internet of Things device. Compared with the prior art, the computer program product provided by the present application has the same beneficial effects as the abnormal monitoring method of the Internet of Things device provided by the above-mentioned embodiments, which will not be repeated here.

[0213] The above only describes some embodiments of the present application, and does not limit the patent scope of the present application, and any equivalent structural transformation made by using the content of the present application specification and drawings, or direct / indirect application in other related technical fields is included in the patent protection scope of the present application.

Claims

1. An anomaly monitoring method of an Internet of Things device, characterized by, The anomaly monitoring method of the Internet of Things device comprises: obtaining current device state data of the Internet of Things device at a current time, wherein the device state data is used to represent the running state of the Internet of Things device at the current time and the running environment of the Internet of Things device at the current time; obtaining historical device state data before the current device state data; inputting the current device state data into a preset threshold output model to obtain a data value range, and detecting whether the current device state data is within the data value range, wherein the threshold output model is obtained based on initial device state data training; if the current device state data is not within the data value range, determining that the Internet of Things device is in an abnormal running state at the current time; determining a current abnormality degree of the Internet of Things device at the current time, and determining a target strategy corresponding to the current abnormality degree from different abnormality degree corresponding preset response strategies, and controlling the Internet of Things device to run according to the target strategy, wherein the current abnormality degree is determined based on an abnormality duration and a data deviation degree, the abnormality duration is a duration composed of continuous time points at which the Internet of Things device is in an abnormal running state from the current time point and before, and the data deviation degree is a deviation between the current device state data and the data value range; detecting whether the Internet of Things device has returned to normal; if the Internet of Things device has not returned to normal, determining a model update instruction; when the model update instruction is detected, updating the threshold output model based on the current abnormality degree, the current device state data and the historical device state data before the current time, and taking the updated threshold output model as a preset threshold output model. 2.The IoT device anomaly monitoring method of claim 1, wherein, The step of determining the data value range based on the current device state data comprises: moving a preset moving window according to a preset step length, and determining reference data at multiple time points from the current device state data and the historical device state data before the current time through the moving window; calculating reference mean and reference standard deviation of the reference data at each time point; weighting the reference standard deviation through a preset adjustment coefficient to obtain a floating value, taking the difference between the reference mean and the floating value as a lower limit value of the value, taking the sum between the reference mean and the floating value as an upper limit value of the value, and determining a data value range based on the lower limit value and the upper limit value. 3.The IoT device anomaly monitoring method of claim 1, wherein, Before the step of determining the data value range based on the current device state data, the method further comprises: performing denoising processing on the current device state data to obtain denoised state data; determining outliers in the denoised state data, and removing the outliers from the denoised state data to obtain abnormality-processed state data; performing data smoothing processing on the abnormality-processed state data through a sliding window technology to obtain smoothed state data; The step of determining the data value range based on the current device state data comprises: determining the data value range based on the smoothed state data. 4.The IoT device anomaly monitoring method of claim 1, wherein, The step of determining the current abnormality degree of the IoT device at the current time comprises: obtaining a next-time running state of the IoT device at a next time; if the next-time running state is an abnormal running state, updating the next time to the current time, and executing the step of obtaining the next-time running state of the IoT device at the next time; if the next-time running state is a normal running state, determining an abnormality duration of the IoT device at the current time, and determining a data deviation degree between the current device state data and the data value range, taking the abnormality duration and the data deviation degree as the current abnormality degree, wherein the abnormality duration is a duration composed of the current time and continuous times before the current time at which the IoT device is in an abnormal running state.

5. The method of claim 1 to 4, wherein The step of controlling the IoT device to run according to the target strategy comprises: if the current abnormality degree is mild abnormality, outputting a warning information through the IoT device; if the current abnormality degree is moderate abnormality, reducing a device load of the IoT device; if the current abnormality degree is severe abnormality, sending a device service of the IoT device to a backup device, and closing the IoT device.

6. A storage medium, characterized by The storage medium is a computer readable storage medium, and the storage medium stores a computer program. The computer program is executed by a processor to implement the steps of the abnormality monitoring method of the IoT device according to any one of claims 1 to 5.

7. A computer program product, characterised in that, The computer program product comprises a computer program. The computer program is executed by a processor to implement the steps of the abnormality monitoring method of the IoT device according to any one of claims 1 to 5.

Images