Equipment monitoring method and system, equipment and medium

Abstract

The invention discloses an equipment monitoring method and system, computer equipment and a medium. The method comprises the following steps: acquiring real-time sensing data acquired by equipment; screening out abnormal data from the real-time sensing data; acquiring an equipment operation parameter and an environment parameter corresponding to each piece of abnormal data in a log of the equipment; determining an abnormal reason of the equipment according to the abnormal data and the corresponding equipment operation parameters and environment parameters; and carrying out classified alarm for the abnormal reasons of the equipment. According to the scheme provided by the invention, the perception data collected by the equipment is combined with the equipment operation parameters and the environment parameters, the equipment problem is predicted and diagnosed, and the equipment state can be monitored more accurately and comprehensively, so that the preventive maintenance of potential faults of the equipment is realized, and the equipment fault rate and the maintenance cost are reduced.

Description

Technical Field

[0001] The present invention relates to the field of vehicle-road collaboration, and in particular to a device monitoring method, system, device and medium. Background Art

[0002] As an advanced transportation system concept, vehicle-road collaboration aims to achieve efficient communication and collaboration between vehicles and road infrastructure to improve traffic efficiency, safety and sustainability. Roadside perception is the basis of vehicle-road collaboration. Through various sensor devices installed on both sides of the road, such as cameras, radars, and lidars, real-time information on vehicles, pedestrians, road conditions, etc. is collected to provide key data support for the vehicle-road collaboration system. The normal operation of roadside perception equipment is crucial to the reliability and effectiveness of the vehicle-road collaboration system.

[0003] At present, the traditional solution for monitoring equipment mainly detects equipment problems by monitoring the equipment status log. This method usually focuses on indicators such as the equipment's hardware status, network connection status, and running time. However, existing methods are difficult to effectively monitor the impact of environmental interference on equipment accuracy. For example, severe weather conditions (such as heavy rain, heavy snow, haze, etc.) may reduce the performance of sensors and cause inaccurate data; occlusion of objects may also affect the sensor's field of view and limit the perception range. In addition, traditional monitoring solutions will not be effective in monitoring problems such as system errors and data delays. Summary of the invention

[0004] In view of this, in order to overcome at least one aspect of the above problems, an embodiment of the present invention provides a device monitoring method, comprising the following steps: Obtain real-time perception data collected by the device; Filtering out abnormal data from the real-time sensing data; Obtaining device operating parameters and environmental parameters corresponding to each of the abnormal data in the log of the device; Determine the abnormal cause of the device according to the abnormal data and the corresponding device operating parameters and environmental parameters; Classify alarms based on the causes of device abnormalities.

[0005] In some embodiments, filtering out abnormal data from the real-time sensing data further includes: Use normal perception data to build a training set; Using the training set to construct a multi-feature regression analysis model to obtain a normal data joint threshold; Based on the normal data combined with the threshold, the historical perception data is mined to obtain a plurality of historical abnormal data; Calculating an abnormal data joint threshold based on the plurality of historical abnormal data; Abnormal data is filtered out from the real-time perception data based on the abnormal data joint threshold.

[0006] In some embodiments, it also includes: The abnormal data joint threshold is updated using the plurality of abnormal data screened within a preset time period, and the updated abnormal data joint threshold is used for subsequent screening of abnormal data.

[0007] In some embodiments, obtaining the device operating parameters and environmental parameters corresponding to each of the abnormal data in the log of the device further includes: Based on the timestamp corresponding to each of the abnormal data, corresponding device operating parameters and environmental parameters are obtained in the log of the device.

[0008] In some embodiments, determining the abnormal cause of the device according to the abnormal data and the corresponding device operating parameters and environmental parameters further includes: Get the rule base; Each abnormal data and the corresponding equipment operation parameters and environmental parameters are matched in the rule base, and the abnormal cause of the equipment is determined according to the matching result.

[0009] In some embodiments, it also includes: Inputting the abnormal data without corresponding matching results and the corresponding equipment operation parameters and environmental parameters into the prediction model; The cause of the abnormality is determined based on the prediction model.

[0010] In some embodiments, filtering out abnormal data from the real-time sensing data further includes: The real-time perception data is filtered through a stream processing framework.

[0011] Based on the same inventive concept, according to another aspect of the present invention, an embodiment of the present invention further provides a device monitoring system, including: A first acquisition module is configured to acquire real-time perception data collected by a device; A screening module, configured to screen out abnormal data from the real-time sensing data; A second acquisition module is configured to acquire device operating parameters and environmental parameters corresponding to each of the abnormal data in the log of the device; an attribution module, configured to determine the abnormal cause of the device according to the abnormal data and the corresponding device operating parameters and environmental parameters; The alarm module is configured to generate classified alarms based on abnormal causes of the equipment.

[0012] Based on the same inventive concept, according to another aspect of the present invention, an embodiment of the present invention further provides a computer device, including: at least one processor; and A memory, wherein the memory stores a computer program that can be run on the processor, and when the processor executes the program, the steps of any one of the device monitoring methods described above are performed.

[0013] Based on the same inventive concept, according to another aspect of the present invention, an embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of any one of the device monitoring methods described above are performed.

[0014] The present invention has one of the following beneficial technical effects: The solution proposed in the present invention combines the sensory data collected by the equipment with the equipment operating parameters and environmental parameters to predict and diagnose equipment problems, and can monitor the equipment status more accurately and comprehensively, thereby achieving preventive maintenance of potential equipment failures and reducing equipment failure rates and maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other embodiments can be obtained based on these drawings without paying creative work.

[0016] Figure 1 A schematic diagram of a flow chart of a device monitoring method provided by an embodiment of the present invention; Figure 2 A schematic diagram of the structure of a device monitoring system provided by an embodiment of the present invention; Figure 3 A schematic diagram of the structure of a computer device provided by an embodiment of the present invention; Figure 4 A schematic diagram of the structure of a computer-readable storage medium provided for an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the embodiments of the present invention are further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0018] It should be noted that all expressions using "first" and "second" in the embodiments of the present invention are for distinguishing two non-identical entities with the same name or non-identical parameters. It can be seen that "first" and "second" are only for the convenience of expression and should not be understood as limitations on the embodiments of the present invention. The subsequent embodiments will not explain this one by one.

[0019] According to one aspect of the present invention, an embodiment of the present invention provides a device monitoring method, such as Figure 1 As shown, it may include the steps of: S1, obtain real-time perception data collected by the device; S2, filtering out abnormal data from the real-time sensing data; S3, obtaining device operating parameters and environmental parameters corresponding to each of the abnormal data in the log of the device; S4, determining the abnormal cause of the device according to the abnormal data and the corresponding device operating parameters and environmental parameters; S5: Classify and issue alarms based on the abnormal causes of the equipment.

[0020] The solution proposed in the present invention combines the sensory data collected by the equipment with the equipment operating parameters and environmental parameters to predict and diagnose equipment problems, and can monitor the equipment status more accurately and comprehensively, thereby achieving preventive maintenance of potential equipment failures and reducing equipment failure rates and maintenance costs.

[0021] In some embodiments, raw perception data is collected in the road test perception device, which may include but is not limited to information such as location, distance, speed, direction, shape, type, etc. These data can come from a variety of sensors, such as radar, camera, laser radar (LiDAR), etc. In order to ensure the quality and integrity of the data, the raw perception data can be cleaned, such as removing noise. The data formats from different sources can also be unified for subsequent processing and analysis. This step involves operations such as timestamp alignment and coordinate system conversion to ensure that data from different data sources can be seamlessly integrated.

[0022] In some embodiments, filtering out abnormal data from the real-time sensing data further includes: Use normal perception data to build a training set; Using the training set to construct a multi-feature regression analysis model to obtain a normal data joint threshold; Based on the normal data combined with the threshold, the historical perception data is mined to obtain a plurality of historical abnormal data; Calculating an abnormal data joint threshold based on the plurality of historical abnormal data; Abnormal data is filtered out from the real-time perception data based on the abnormal data joint threshold.

[0023] Specifically, normal perception data refers to the roadside perception data of vehicles with normal trajectories. The roadside perception data of vehicles with normal trajectories means that there is no abnormality in these roadside perception data. Then, the normal perception data is marked manually. The manual marking method in this embodiment can be reflected in the process of obtaining multiple screening conditions based on big data statistics, and filtering the obtained roadside perception data through multiple screening conditions to obtain a large amount of normal data. It can be understood by those skilled in the art that the above method of filtering data by setting filtering conditions is only an example and not a limitation of the "manual marking" method.

[0024] Then, a multi-feature regression analysis model is constructed through the marked normal perception data to calculate the joint threshold of the normal perception data. The joint threshold in this embodiment can be understood as a range composed of multiple feature thresholds. The multiple features may include but are not limited to position, distance, speed, direction, shape, type, etc.

[0025] Then, based on the normal data joint threshold, abnormal data is mined from historical data. Abnormal data usually refers to data that does not conform to the normal data distribution or pattern, which may be target missed detection, false detection, position mutation, speed mutation, direction mutation, size change, clone, flicker, jump, etc. Abnormal data can be defined based on the performance indicators of the device and the needs of the actual application scenario.

[0026] Then, the abnormal data joint threshold is calculated based on the abnormal data. The abnormal data joint threshold can be calculated based on business rule methods, machine learning methods, probability density functions, etc. For example, when calculating the abnormal data joint threshold based on the probability density function, the covariance matrix of multiple features can be constructed first, and then the probability density function of the Gaussian distribution can be constructed based on the covariance matrix. Finally, the abnormal data joint threshold in the probability density function is calculated using the preset confidence. Finally, the abnormal data is filtered out in the real-time perception data according to the abnormal point joint threshold.

[0027] In some embodiments, a distributed computing framework can be used for data processing. For example, since Apache Kafka can handle a large amount of historical data writing and reading operations, ensuring efficient storage and access of data, thereby meeting the needs of large-scale data processing, Apache Kafka can be used to store historical data. Apache Spark is a powerful distributed computing framework that can handle large-scale data sets. It can perform computing tasks in parallel on multiple nodes, greatly improving computing efficiency. Therefore, Apache Spark can be used to read data from Apache Kafka and then calculate the joint threshold of abnormal data.

[0028] In some embodiments, it also includes: The abnormal data joint threshold is updated using the plurality of abnormal data screened within a preset time period, and the updated abnormal data joint threshold is used for subsequent screening of abnormal data.

[0029] Specifically, the threshold can be iterated using a preset time period as a cycle, using all the abnormal data mined in the previous preset time period, and the latest abnormal data joint threshold is continuously iterated based on the new abnormal data. By periodically and continuously iteratively updating the threshold, the accuracy and robustness of abnormal data detection can be gradually improved.

[0030] In some embodiments, obtaining the device operating parameters and environmental parameters corresponding to each of the abnormal data in the log of the device further includes: Based on the timestamp corresponding to each of the abnormal data, corresponding device operating parameters and environmental parameters are obtained in the log of the device.

[0031] Specifically, the log of the device at the corresponding time point is found according to the timestamp of the abnormal data. The device log usually contains the operating parameters and environmental parameters of the device, and the sensing data can be associated with the device log through timestamp matching. In some embodiments, the device operating parameters may include network delay, data packet loss rate, communication bandwidth, error rate, signal strength, processing delay, voltage, current and other parameters, and the environmental parameters may include weather, wind speed, temperature, humidity, visibility and the like.

[0032] In some embodiments, determining the abnormal cause of the device according to the abnormal data and the corresponding device operating parameters and environmental parameters further includes: Get the rule base; Each abnormal data and the corresponding equipment operation parameters and environmental parameters are matched in the rule base, and the abnormal cause of the equipment is determined according to the matching result.

[0033] Specifically, the rule base is the key basis for determining the cause of equipment abnormality. It contains a series of relationship rules between known abnormal causes and abnormal data, corresponding equipment parameters, and environmental parameters. The rule base can quickly and accurately identify the cause of abnormality by matching the actual abnormal data, corresponding equipment parameters, and environmental parameters with the rules, providing guidance for subsequent troubleshooting and equipment maintenance.

[0034] When building a rule base, we can summarize common abnormal situations and their corresponding parameter characteristics based on previous experience in handling equipment failures to form rules. For example, if the temperature of the equipment rises significantly before and after the abnormal data, it may be a cooling system failure. We can also analyze a large amount of equipment operation data to find the correlation pattern between abnormal data and equipment parameters and environmental parameters, and extract rules. We can also convert it into rules based on the knowledge and experience of experts in the equipment field and store it in the library.

[0035] When the abnormal data is obtained, it and the corresponding device parameters and environmental parameters are matched one by one with the rules in the rule base. Matching can be based on a variety of methods, such as numerical range matching, logical condition judgment, etc. If the abnormal data meets the conditions of a rule, it is considered that a possible cause of the abnormality has been found. Based on the matching results, one or more possible causes of the abnormality may be obtained. These results can be further analyzed and judged to determine the most accurate cause of the abnormality. For example, the degree of match between the abnormal data and each possible cause of the abnormality can be evaluated based on the matching degree. The higher the matching degree, the greater the possibility of the cause. You can also view the causes of similar abnormal situations in the past for the device as a reference. Finally, combined with the actual operation of the equipment, maintenance records and other information, a comprehensive analysis is conducted to determine the final cause of the abnormality.

[0036] In some embodiments, it also includes: Inputting the abnormal data without corresponding matching results and the corresponding equipment operation parameters and environmental parameters into the prediction model; The cause of the abnormality is determined based on the prediction model.

[0037] Specifically, when the abnormal data and the corresponding equipment operating parameters and environmental parameters do not find a match in the existing rule base, other methods can be used to determine the cause of the abnormality. This may be because a new abnormal situation has occurred, or the existing rule base is not perfect enough to cover all possible abnormal situations. At this time, machine learning methods can be used to determine the cause of the abnormality, and the abnormal data and the corresponding equipment operating parameters and environmental parameters are input into the trained prediction model. The model will calculate and analyze the input data and output the predicted cause of the abnormality.

[0038] Furthermore, the results output by the prediction model can be interpreted and evaluated. If the result is a clear abnormal cause, it can be directly used as the determined abnormal cause. If the result is a probability distribution or multiple possible abnormal causes, it can be further analyzed and judged in combination with other information. At the same time, feedback and improvement of the prediction model can be provided based on the accuracy and practicality of the prediction results. If the prediction results are inaccurate, the model parameters can be readjusted, training data can be added, or the model structure can be improved. In this way, as more abnormal data is processed and analyzed, the prediction model can continue to learn and optimize, improving its ability to determine the abnormal cause.

[0039] In some embodiments, the prediction model may be a decision tree model. Before actual reasoning, the decision tree model may be trained. During training, the training set and the test set are constructed using the annotated abnormal data, equipment operating parameters, and weather parameters. The training set is used to train the decision tree model, and then the test set is used to predict the trained model. The performance of the model is evaluated through the accuracy and classification report.

[0040] In some embodiments, in step S5, classified alarms are issued based on the causes of equipment abnormalities. Classification can be performed based on the importance of the causes of the abnormalities, and different alarm levels can be set, such as severe alarms, important alarms, general alarms, etc. After the cause of the abnormality is determined based on the abnormal data, corresponding equipment operating parameters, and environmental parameters, the corresponding alarm level is determined, and an alarm is issued through the corresponding alarm method. For example, when the abnormal cause is determined to be a power failure or hardware failure according to the abnormal data, the corresponding equipment operating parameters and the environmental parameters, if the above abnormal cause is set to require a serious alarm, an alarm can be issued by SMS alarm / email alarm, or a red pop-up window with a large area can be popped up on the management system interface of the monitoring center for alarm; when the abnormal cause is determined to be a cooling system failure, severe environmental interference, etc. according to the abnormal data, the corresponding equipment operating parameters and the environmental parameters, if the above abnormal cause is set to require an important alarm, an alarm can be issued by SMS alarm / email alarm, or an orange pop-up window can be popped up on the management system interface of the monitoring center for alarm; when the abnormal cause is determined to be a network bandwidth fluctuation, slight environmental interference, etc. according to the abnormal data, the corresponding equipment operating parameters and the environmental parameters, if the above abnormal cause is set to require a general alarm, a small pop-up window can be popped up on the management system interface of the monitoring center for alarm. In this way, different types of alarms are classified and graded, and different alarm processing methods are adopted, which can more effectively manage and handle abnormal conditions of equipment and achieve fast and accurate equipment monitoring effects.

[0041] In some embodiments, filtering out abnormal data from the real-time sensing data further includes: The real-time perception data is filtered through a stream processing framework.

[0042] Specifically, stream processing frameworks, such as Apache Flink, can capture and analyze anomalies in streaming data in real time, which is very useful for scenarios that require real-time response. In addition, Flink's stream processing framework has efficient processing capabilities, can quickly process large amounts of data, and detect anomalies in real time. This helps improve data processing efficiency, reduce latency, and quickly respond to abnormal situations.

[0043] In some embodiments, the perception data and analysis results can also be displayed using a variety of visualization methods such as heat maps, 3D point cloud maps, and time series maps. Data visualization can help intuitively understand the distribution, change trends, and anomalies of the data, and improve the efficiency of problem location and diagnosis.

[0044] The solution proposed by the present invention combines the sensory data collected by the device with the device operating parameters and environmental parameters, rather than relying solely on the device status log, to more accurately determine whether the device is working properly and whether it is affected by environmental interference, system errors or data delays, thereby providing a more comprehensive understanding of the device's operating status and improving the coverage and effectiveness of monitoring. Moreover, the monitoring method based on the sensory result data can more accurately detect device problems. For example, if the data output by the device does not match the actual situation, or the data change trend is abnormal, it can be determined that the device may be faulty or interfered with, thereby promptly discovering potential problems and improving the reliability and accuracy of the device.

[0045] Based on the same inventive concept, according to another aspect of the present invention, an embodiment of the present invention further provides a device monitoring system 400, such as Figure 2 As shown, including: A first acquisition module 401 is configured to acquire real-time perception data collected by a device; A screening module 402 is configured to screen out abnormal data from the real-time sensing data; The second acquisition module 403 is configured to acquire the device operation parameters and environmental parameters corresponding to each of the abnormal data in the log of the device; Attribution module 404, configured to determine the abnormal cause of the device according to the abnormal data and the corresponding device operating parameters and environmental parameters; The alarm module 405 is configured to generate classified alarms according to the abnormal causes of the equipment.

[0046] In some embodiments, the screening module 402 is further configured to: Use normal perception data to build a training set; Using the training set to construct a multi-feature regression analysis model to obtain a normal data joint threshold; Based on the normal data combined with the threshold, the historical perception data is mined to obtain a plurality of historical abnormal data; Calculating an abnormal data joint threshold based on the plurality of historical abnormal data; Abnormal data is filtered out from the real-time perception data based on the abnormal data joint threshold.

[0047] In some embodiments, the screening module 402 is further configured to: The abnormal data joint threshold is updated using the plurality of abnormal data screened within a preset time period, and the updated abnormal data joint threshold is used for subsequent screening of abnormal data.

[0048] In some embodiments, the screening module 402 is further configured to: Based on the timestamp corresponding to each of the abnormal data, corresponding device operating parameters and environmental parameters are obtained in the log of the device.

[0049] In some embodiments, the attribution module 404 is further configured to: Get the rule base; Each abnormal data and the corresponding equipment operation parameters and environmental parameters are matched in the rule base, and the abnormal cause of the equipment is determined according to the matching result.

[0050] In some embodiments, the attribution module 404 is further configured to: Inputting the abnormal data without corresponding matching results and the corresponding equipment operation parameters and environmental parameters into the prediction model; The cause of the abnormality is determined based on the prediction model.

[0051] In some embodiments, the screening module 402 is further configured to: The real-time perception data is filtered through a stream processing framework.

[0052] The solution proposed in the present invention combines the sensory data collected by the equipment with the equipment operating parameters and environmental parameters to predict and diagnose equipment problems, and can monitor the equipment status more accurately and comprehensively, thereby achieving preventive maintenance of potential equipment failures and reducing equipment failure rates and maintenance costs.

[0053] Based on the same inventive concept, according to another aspect of the present invention, Figure 3 As shown, an embodiment of the present invention further provides a computer device 501, including: at least one processor 520; and The memory 510 stores a computer program 511 that can be run on a processor. When the processor 520 executes the program, the steps of any of the above-mentioned device monitoring methods are performed.

[0054] Based on the same inventive concept, according to another aspect of the present invention, Figure 4 As shown, an embodiment of the present invention further provides a computer-readable storage medium 601, which stores a computer program 610. When the computer program 610 is executed by a processor, the steps of any of the above device monitoring methods are performed.

[0055] Finally, it should be noted that a person skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods.

[0056] Furthermore, it should be appreciated that the computer-readable storage medium (eg, memory) herein may be either volatile memory or nonvolatile memory, or may include both volatile memory and nonvolatile memory.

[0057] It will also be appreciated by those skilled in the art that various exemplary logic blocks, modules, circuits and algorithm steps described in conjunction with the disclosure herein can be implemented as electronic hardware, computer software or a combination of the two. In order to clearly illustrate this interchangeability of hardware and software, a general description has been given to the functions of various schematic components, blocks, modules, circuits and steps. Whether this function is implemented as software or hardware depends on specific applications and the design constraints imposed on the entire system. Those skilled in the art can implement the function in various ways for each specific application, but this implementation decision should not be interpreted as causing a departure from the disclosed scope of the embodiments of the present invention.

[0058] The above are exemplary embodiments disclosed in the present invention, but it should be noted that various changes and modifications may be made without departing from the scope disclosed in the embodiments of the present invention as defined in the claims. The functions, steps and / or actions of the method claims according to the disclosed embodiments described herein do not need to be performed in any particular order. In addition, although the elements disclosed in the embodiments of the present invention may be described or required in individual form, they may also be understood as multiple unless explicitly limited to the singular.

[0059] It should be understood that, as used herein, the singular forms "a", "an" are intended to include the plural forms as well, unless the context clearly supports an exception. It should also be understood that, as used herein, "and / or" refers to any and all possible combinations including one or more of the associated listed items.

[0060] The serial numbers of the embodiments disclosed in the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0061] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0062] A person skilled in the art should understand that the discussion of any of the above embodiments is only exemplary and is not intended to imply that the scope of the disclosure of the embodiments of the present invention (including the claims) is limited to these examples; under the concept of the embodiments of the present invention, the technical features in the above embodiments or different embodiments can also be combined, and there are many other changes in different aspects of the above embodiments of the present invention, which are not provided in detail for the sake of simplicity. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present invention should be included in the protection scope of the embodiments of the present invention.

Claims

1. A device monitoring method, characterized in that: The following steps are involved: Obtain real-time perception data collected by the device; Filtering out abnormal data from the real-time sensing data; Obtaining device operating parameters and environmental parameters corresponding to each of the abnormal data in the log of the device; Determine the abnormal cause of the device according to the abnormal data and the corresponding device operating parameters and environmental parameters; Classify alarms based on the causes of device abnormalities.

2. The method according to claim 1, characterized in that Screening out abnormal data from the real-time sensing data further includes: Use normal perception data to build a training set; Using the training set to construct a multi-feature regression analysis model to obtain a normal data joint threshold; Based on the normal data combined with the threshold, the historical perception data is mined to obtain a plurality of historical abnormal data; Calculating an abnormal data joint threshold based on the plurality of historical abnormal data; Abnormal data is filtered out from the real-time perception data based on the abnormal data joint threshold.

3. The method according to claim 2, characterized in that Also includes: The abnormal data joint threshold is updated using the plurality of abnormal data screened within a preset time period, and the updated abnormal data joint threshold is used for subsequent screening of abnormal data.

4. The method according to claim 1, characterized in that Obtaining device operating parameters and environmental parameters corresponding to each of the abnormal data in the log of the device further includes: Based on the timestamp corresponding to each of the abnormal data, corresponding device operating parameters and environmental parameters are obtained in the log of the device.

5. The method according to claim 1, characterized in that Determining the abnormal cause of the device according to the abnormal data and the corresponding device operating parameters and environmental parameters further includes: Get the rule base; Each abnormal data and the corresponding equipment operation parameters and environmental parameters are matched in the rule base, and the abnormal cause of the equipment is determined according to the matching result.

6. The method according to claim 5, characterized in that Also includes: Inputting the abnormal data without corresponding matching results and the corresponding equipment operation parameters and environmental parameters into the prediction model; The cause of the abnormality is determined based on the prediction model.

7. The method according to claim 1, characterized in that Screening out abnormal data from the real-time sensing data further includes: The real-time perception data is filtered through a stream processing framework.

8. A device monitoring system, characterized in that: include: A first acquisition module is configured to acquire real-time perception data collected by a device; A screening module, configured to screen out abnormal data from the real-time sensing data; A second acquisition module is configured to acquire device operating parameters and environmental parameters corresponding to each of the abnormal data in the log of the device; an attribution module, configured to determine the abnormal cause of the device according to the abnormal data and the corresponding device operating parameters and environmental parameters; The alarm module is configured to generate classified alarms based on abnormal causes of the equipment.

9. A computer device comprising: at least one processor; as well as A memory storing a computer program executable on the processor, wherein the processor executes the steps of the method according to any one of claims 1 to 7 when executing the program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are performed.

Images