Equipment operation and maintenance method and device, electronic equipment and storage medium

Abstract

One or more embodiments of the invention provide an equipment operation and maintenance method and device, electronic equipment and a storage medium. The method comprises the following steps: responding to an abnormal alarm sent by any device in a device network, regarding the device as a target device, and obtaining historical operation and maintenance data of the target device and other devices except the target device in the device network; according to the historical operation and maintenance data of the target device, obtaining an abnormal index of the target device, the abnormal index of the target device indicating an abnormal degree of the target device in a historical time period and a change trend of the abnormal degree; according to the historical operation and maintenance data of the target equipment and the historical operation and maintenance data of the other equipment, obtaining a normal index of the target equipment, the normal index of the target equipment indicating the similarity between the target equipment and the abnormal condition of the other equipment in the historical time period; according to the abnormal index of the target equipment and the normal index of the target equipment, the abnormal index of the target equipment is obtained, and the abnormal index is used for formulating an operation maintenance strategy for the target equipment.

Description

Technical Field

[0001] One or more embodiments of the present application relate to the technical field of equipment operation and maintenance, and in particular, to a method, device, electronic device, and storage medium for equipment operation and maintenance. Background Art

[0002] The information infrastructure refers to the digital infrastructure system that supports modern social and economic activities, public services, and technological development. It consists of hardware, software, networks, data resources, and related technical standards, aiming to achieve efficient transmission, storage, processing, and sharing of information. Currently, the most common hardware facility of the information infrastructure is the data center, such as a server system, and the server system usually includes multiple server devices.

[0003] The traditional operation and maintenance method of the information infrastructure usually requires operation and maintenance personnel to go to the site for inspection, maintenance, and troubleshooting. This not only consumes time and labor costs but also has low efficiency. Intelligent remote operation and maintenance can, through technologies such as the Internet of Things, big data, and artificial intelligence, monitor the running status of devices in real time, automatically collect and analyze data, timely discover potential problems and faults, and give early warnings and handle them. By adopting the intelligent remote operation and maintenance mode, there is no need to dispatch a large number of operation and maintenance personnel to each site for maintenance, reducing labor costs and travel expenses. At the same time, through the analysis and prediction of device operation data, potential device faults can be discovered in advance, and preventive maintenance measures can be taken to avoid downtime losses and increased maintenance costs caused by sudden device failures. Therefore, it is crucial for the intelligent remote operation and maintenance of the infrastructure.

[0004] When there are multiple infrastructure devices in the information infrastructure, the existing remote operation and maintenance method of the information infrastructure is to perform remote operation and maintenance on each infrastructure device in the information infrastructure separately, and monitor the anomalies of each infrastructure device according to the running data of each infrastructure device. Summary of the Invention

[0005] In view of this, the purpose of one or more embodiments of the present application is to propose a method, device, electronic device, and storage medium for equipment operation and maintenance to solve the problems in the background art.

[0006] Based on the above purpose, one or more embodiments of the present application provide a method for equipment operation and maintenance, including:

[0007] In response to an abnormal alarm issued by any device in the device network, regard the device as a target device and obtain the historical operation and maintenance data of the target device and other devices in the device network except the target device;

[0008] Obtain the anomaly index of the target device based on the historical operation and maintenance data of the target device, where the anomaly index of the target device indicates the anomaly degree and the change trend of the anomaly degree of the target device in the historical period;

[0009] Obtain the normal index of the target device based on the historical operation and maintenance data of the target device and the historical operation and maintenance data of other devices, where the normal index of the target device indicates the similarity of the anomaly conditions of the target device with those of other devices in the historical period;

[0010] Obtain the anomaly metric of the target device based on the anomaly index and the normal index of the target device, where the anomaly metric is used to formulate an operation and maintenance strategy for the target device.

[0011] Optionally, the anomaly index is obtained by performing the following operations:

[0012] Obtain the anomaly degree of the device at each historical moment in the historical period and the change trend of the anomaly degree over time based on the historical operation and maintenance data of the device;

[0013] Obtain the anomaly index based on the anomaly degree of the device at each historical moment and the change trend of the anomaly degree.

[0014] Optionally, the normal index is obtained by performing the following operations:

[0015] Obtain the anomaly conditions of the first device and the second device at each historical moment in the historical period based on the historical operation and maintenance data of the first device and the second device, where the first device and the second device are any two devices in the device network;

[0016] Obtain the anomaly similarity of the first device and the second device at each historical moment in the historical period and the average anomaly degree of the first device and the second device at each historical moment in the historical period based on the anomaly conditions of the first device at each historical moment in the historical period and the anomaly conditions of the second device at each historical moment in the historical period;

[0017] Obtain the normal index of the first device based on the anomaly similarity of the first device and the second device at each historical moment in the historical period and the average anomaly degree of the first device and the second device at each historical moment in the historical period.

[0018] Optionally, the anomaly metric is obtained by performing the following operations:

[0019] Determine the operating state of the target device at each historical moment in the historical period, where the operating state includes a normal state and an abnormal state;

[0020] Based on the status values of the target device at each historical moment in the historical period, obtain the normal state average value and the normal state extreme difference value of the target device in the historical period;

[0021] Based on the normal state average value and the normal state extreme difference value of the target device in the historical period and the abnormal state value of the target device, obtain the abnormal index of the target device.

[0022] Optionally, it further includes:

[0023] Obtain the normal device set and the abnormal device set of the device network, and obtain the first environmental characteristics of all normal devices in the normal device set and the second environmental characteristics of all abnormal devices in the abnormal device set;

[0024] Obtain the observed operation and maintenance data of the target device, where the observed operation and maintenance data represents the operation and maintenance data in a preset time period after the target device issues an abnormal alarm;

[0025] Based on the observed operation and maintenance data of the target device, obtain the observed abnormal index of the target device, and the observed abnormal index is obtained through the same calculation steps as the abnormal index;

[0026] Based on the abnormal index, the observed abnormal index, the first environmental characteristics, and the second environmental characteristics of the target device, determine whether the target device is an abnormal device.

[0027] Optionally, the determining whether the target device is an abnormal device based on the abnormal index, the observed abnormal index, the first environmental characteristics, and the second environmental characteristics of the target device includes:

[0028] In response to determining that the distance between the first environmental characteristics and the second environmental characteristics is less than or equal to a preset threshold, and the observed abnormal index of the target device is less than or equal to the abnormal index of the target device, determine that the target device is a normal device;

[0029] In response to determining that the distance between the first environmental characteristics and the second environmental characteristics is greater than the preset threshold, or the observed abnormal index of the target device is greater than the abnormal index of the target device, determine that the target device is an abnormal device.

[0030] Optionally, it further includes:

[0031] In response to determining that the difference between the first environmental characteristics and the second environmental characteristics is greater than the pre-examination threshold, obtain the observed operation and maintenance data of all devices in the device network;

[0032] Based on the observed operation and maintenance data of all devices in the device network, obtain the observed abnormal index of all devices in the device network;

[0033] Perform a linear fit on the first environmental characteristics and observed anomaly indices of all normal devices in the set of normal devices to obtain a fitted line model and the root mean square error corresponding to the fitted line model;

[0034] Based on the fitted line model and the second environmental characteristics of all abnormal devices in the set of abnormal devices, obtain the fitted anomaly indices of all abnormal devices in the set of abnormal devices;

[0035] Based on the observed anomaly indices, fitted anomaly indices of the abnormal devices in the set of abnormal devices, and the root mean square error corresponding to the fitted line model, obtain the temperature mismatch degree of the abnormal devices in the set of abnormal devices;

[0036] Based on the temperature mismatch degree and the fitted anomaly indices of the abnormal devices in the set of abnormal devices, obtain the failure probabilities of the abnormal devices in the set of abnormal devices;

[0037] Based on the failure probabilities of the abnormal devices in the set of abnormal devices, update the set of abnormal devices.

[0038] As can be seen from the above, the device operation and maintenance method provided by one or more embodiments of the present application not only considers the anomaly indices and normal indices of each infrastructure device itself in each historical time period, but also considers the similarity of the anomaly occurrence situations among infrastructure devices in each historical time period, realizes accurate fault judgment of each infrastructure device, and improves the accuracy of fault monitoring of infrastructure devices.

[0039] Based on the same inventive concept, one or more embodiments of the present application also provide a device operation and maintenance apparatus, including:

[0040] An acquisition module, configured to respond to an anomaly alarm issued by any device in the device network, regard the device as a target device, and acquire the historical operation and maintenance data of the target device and other devices in the device network except the target device;

[0041] A first calculation module, configured to obtain the anomaly index of the target device according to the historical operation and maintenance data of the target device, and the anomaly index of the target device indicates the anomaly degree and the change trend of the anomaly degree of the target device in the historical period;

[0042] A second calculation module, configured to obtain the normal index of the target device according to the historical operation and maintenance data of the target device and the historical operation and maintenance data of the other devices, and the normal index of the target device indicates the similarity of the anomaly situation of the target device with the other devices in the historical period;

[0043] A third computing module, configured to obtain an anomaly metric of the target device according to the anomaly index and the normal index of the target device, where the anomaly metric is used to formulate an operation and maintenance strategy for the target device.

[0044] Based on the same inventive concept, one or more embodiments of the present application further provide an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the device operation and maintenance method as described in any one of the above.

[0045] Based on the same inventive concept, one or more embodiments of the present application further provide a non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the device operation and maintenance method described in any one of the above.

[0046] A device operation and maintenance device, an electronic device, and a computer-readable storage medium provided by the present application can all implement the steps of the above device operation and maintenance method, and thus also have the beneficial effects of the above device operation and maintenance method. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] To more clearly illustrate the technical solutions in one or more embodiments of the present application or in the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only one or more embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0048] Figure 1 It is a flowchart of the device operation and maintenance method for one or more embodiments of the present application;

[0049] Figure 2 It is a structural diagram of the device operation and maintenance device for one or more embodiments of the present application;

[0050] Figure 3 It is a hardware structure diagram of the electronic device for one or more embodiments of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0051] To make the objectives, technical solutions, and advantages of the present disclosure clearer and more understandable, the following further details the present disclosure in conjunction with specific embodiments and with reference to the accompanying drawings.

[0052] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in one or more embodiments of this application should have the ordinary meanings understood by those of ordinary skill in the art to which this disclosure pertains. The terms "first", "second" and similar words used in one or more embodiments of this application do not denote any order, quantity or importance, but are only used to distinguish different components. Words such as "including" or "comprising" mean that the elements or objects appearing before this word cover the elements or objects listed after this word and their equivalents, without excluding other elements or objects. Words such as "connected" or "linked" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Upper", "lower", "left", "right", etc. are only used to represent relative position relationships. When the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0053] As described in the background art section, in the related art, when there are multiple infrastructure devices in an information infrastructure, the existing remote operation and maintenance method for the information infrastructure is to perform remote operation and maintenance on each infrastructure device in the information infrastructure separately, and abnormal monitoring of each infrastructure device is realized based on the operation data of each infrastructure device. However, this method has the problem of poor monitoring effect.

[0054] During the implementation of this disclosure, the applicant found that multiple devices in the information infrastructure affect each other and are interrelated. It is easy to have deviations in fault judgment based only on the monitoring data of the target device, thus affecting the accuracy of fault monitoring of infrastructure devices.

[0055] Reference Figure 1 , the device operation and maintenance method of one or more embodiments of this application includes the following steps:

[0056] Step S101: In response to an abnormal alarm issued by any device in the device network, regard this device as the target device and obtain the historical operation and maintenance data of the above target device and other devices in the above device network except the above target device;

[0057] Step S102: Obtain the abnormal index of the above target device according to the historical operation and maintenance data of the above target device. The abnormal index of the above target device indicates the abnormal degree and the change trend of the abnormal degree of the above target device in the historical period;

[0058] Step S103: Obtain the normal index of the above target device according to the historical operation and maintenance data of the above target device and the historical operation and maintenance data of the above other devices. The normal index of the above target device indicates the similarity of the abnormal conditions of the above target device and the above other devices in the historical period;

[0059] Step S104: Obtain an anomaly index of the target device according to the anomaly index and the normal index of the target device, where the anomaly index is used to formulate an operation and maintenance strategy for the target device.

[0060] In the embodiments of the present disclosure, operation and maintenance data of a device can be obtained through monitoring sensors deployed on the device.

[0061] Judging whether a device issues an anomaly alarm according to the operation and maintenance data can be: inputting the operation and maintenance data of the device into a preset operation and maintenance model (such as a model constructed based on a support vector machine), and judging whether the device is in an abnormal state through the output result.

[0062] The above monitoring data can be the CPU occupancy rate, etc. Taking the CPU occupancy rate as an example, a normal range of the monitoring data can be set. The monitoring data within this range is normal data, and the monitoring data not within this range is abnormal data. For example, too high a CPU occupancy rate indicates abnormal monitoring data.

[0063] In the embodiments of the present disclosure, normal and abnormal can be represented by the numbers 0 and 1 respectively, 0 represents normal, 1 represents abnormal, and the number 0 or 1 is used as the label of the monitoring data at each moment. After determining the label, the result can also be stored in the log file corresponding to the infrastructure device. The sampling frequency of the monitoring data is set according to the actual situation, such as once every 0.5 seconds.

[0064] In the embodiments of the present disclosure, taking the moment when the device issues an anomaly alarm as a reference, multiple historical time periods before the moment when the anomaly occurs are determined. In an exemplary embodiment, the moment when the anomaly occurs is the cut-off moment of the last historical time period in terms of time sequence. It should be understood that the number of historical time periods and the time length of each historical time period are set according to the actual situation, and this embodiment does not make any limitations. In an exemplary embodiment, the number of historical time periods is 30, the length of each historical time period is 2 minutes, and the historical time periods are continuous in terms of time sequence.

[0065] According to the above monitoring data, the anomaly degree of the device can be obtained. In the embodiments of the present disclosure, the above anomaly degree can be obtained from the anomaly occurrence frequency of the device within the historical time period. In an exemplary embodiment, the total number of monitoring times of the monitoring data of the device within the historical time period is obtained, and then the number of anomaly occurrences of the monitoring data of the device within the historical time period is obtained. By calculating the ratio of the number of anomaly occurrences to the total number of monitoring times, the anomaly occurrence frequency (i.e., the proportion of the number of anomalies) is obtained, and the anomaly occurrence frequency is used as the anomaly severity degree of the historical time period of the device. The greater the anomaly occurrence frequency, the more serious the abnormal state of the device within the historical time period, and the higher the anomaly severity degree of the device within the historical time period.

[0066] The anomaly index of the device can be determined based on the severity of the anomaly of the device and the change trend of the severity of the anomaly within the historical time period.

[0067] In the process of implementing the present disclosure, the applicant found that even when the device is operating normally, an anomaly alarm may be received. For example, the CPU occupancy causes the device to issue an anomaly alarm because the operating degree of the device increases within the historical time period, the load increases, resulting in the generation of anomaly monitoring data. However, at this time, by combining the monitoring data analysis of other devices, it can be determined that the device may be in the peak period of the information infrastructure service, or in a scenario where there is a significant increase in the user access volume, etc.

[0068] Therefore, based on the severity of the anomaly and the increasing trend of the severity of the anomaly at each historical moment within the historical time period, the anomaly situation of the device can be better reflected.

[0069] In an exemplary embodiment, taking the time series of each historical time period as the abscissa and the severity of the anomaly at each historical moment as the ordinate for linear fitting, the slope of the fitted line is obtained. The slope characterizes the change trend of the anomaly degree of the device at each historical moment. The larger the slope, the more obvious the increasing trend of the anomaly degree of the device, and the more serious the anomaly state of the device. It should be understood that the slope here has positive and negative values. When it is positive, the larger the slope value, the more obvious the increasing trend of the anomaly degree; when it is negative, the anomaly degree shows a decreasing trend.

[0070] In the embodiments of the present disclosure, the overall level of the anomaly degree of the device can also be obtained according to the anomaly degree of each historical time period of the device. The overall level can be the average value of the anomaly degree of the device for all historical moments. The higher the overall level of the anomaly degree, the more serious the anomaly situation of the device at each historical moment.

[0071] In the embodiments of the present disclosure, normalization processing can also be performed on one or more of the anomaly degree, the increasing trend of the anomaly degree, and the overall level.

[0072] In an exemplary embodiment, the slope of the fitted line corresponding to the device is normalized. The normalization method here is: obtaining the slopes of the fitted lines corresponding to all devices, determining the maximum slope and the minimum slope from them, and normalizing the slope of the fitted line corresponding to the device according to the maximum-minimum normalization method.

[0073] In an exemplary embodiment, the method for normalizing the average value of the anomaly degree of the device is: obtaining the average values of the anomaly degrees corresponding to all devices, determining the maximum average value and the minimum average value from them, and normalizing the average value of the severity of the anomaly corresponding to the device according to the maximum-minimum normalization method.

[0074] In an exemplary embodiment, a weighted sum is performed on the average degree of abnormality of the normalized device and the slope corresponding to the normalized device. For example, the weights of both parts are 0.5, that is, essentially the average degree of abnormality of the normalized device and the average value of the slope corresponding to the normalized device are calculated, and the obtained result is used as the abnormality index of the device.

[0075] The normal index of a device represents the synchronism in time series of the change in the degree of abnormality of the device with other devices in the device network. If they are synchronous, it indicates that the change is due to the normal working state of the device.

[0076] In an exemplary embodiment, the synchronism in time series of the change in the degree of abnormality between the target device and other devices in the device network can be judged by similarity calculation methods such as Pearson correlation coefficient, cosine similarity, DTW distance, etc. It should be noted that if the dynamic time warping distance is used to obtain the similarity, the dynamic time warping distance needs to be negatively correlated to obtain the similarity.

[0077] The calculation steps of the normal index of the target device can be as follows:

[0078] Obtain the average degree of abnormality of the target device and other devices in the historical time period as the overall level of the degree of abnormality of the target device and other devices.

[0079] According to the similarity of the occurrence of abnormalities (such as the change trend of the degree of abnormality) between the target device and other devices in the historical time period, and the average degree of abnormality of the target device and other devices in the historical time period, obtain the normal index of the target device relative to other devices. Among them, the more similar the occurrence of abnormalities between the target device and other devices in the historical time period, the more it indicates that the target device belongs to the change caused by the normal working state of the device. That is to say, the normal index of the target device is proportional to the above similarity.

[0080] Moreover, if both the target device and other devices show overloaded operation, that is, these two devices show significant abnormalities within the historical time period. At this time, although the abnormal states of the monitoring data between these two devices show a high degree of similarity, but these two devices are under overloaded operation, with a high degree of similarity, the possibility of device abnormality is instead higher.

[0081] Therefore, if the similarity of the occurrence of abnormalities between the target device and other devices in the historical time period is high, and the overall level of the degree of abnormality of the target device and other devices is low, it indicates that the target device is in a normal working state within the historical time period.

[0082] In an exemplary embodiment, the following gives a specific quantification method for the normal index of the target device relative to other devices:

[0083]

[0084] where c i represents the normal index of the target device relative to the i-th other device in the historical time period, and s i represents the similarity of the abnormal occurrence situation between the target device and the i-th other device in the historical time period, a represents the degree of abnormality of the target device in the historical time period, and a i represents the degree of abnormality of the i-th other device in the historical time period, and norm(s i ) represents the normalization of s i . The normalization method can be: obtain the maximum value and the minimum value among the similarities of the target device with each other device in the historical time period, and then use the maximum-minimum normalization method to normalize s i .

[0085] Calculate the average value of the normal indexes of the target device relative to each other device in the historical time period as the normal state of the historical time period of the target device. The larger this value is, the more the target device is in a normal working state in this historical time period, reflecting the working state of the target device through comparison with each other device.

[0086] In the embodiments of the present disclosure, the abnormal index can be calculated according to the abnormality of the target device, the overall normal state level of each historical moment of the device network, and the overall fluctuation of the device network. Among them, the higher the abnormal state of the target device, the higher the abnormal index of the target device; the higher the overall level of the normal state of each historical moment of the target device, the lower the device abnormal index of the target device; the weaker the overall fluctuation of the normal state of each historical moment of the target device, that is, the more stable the normal state, the lower the abnormal index of the target device. Therefore, according to the abnormal state of the target device, the overall level of the normal state of each historical moment of the target device, and the overall fluctuation situation, the abnormal index of the target device can be obtained.

[0087] In an exemplary embodiment, the overall level of the normal state of each historical moment of the target device is the average value of the normal states of each historical moment of the target device, defined as the average normal state. The process of obtaining the fluctuation situation of the normal state of each historical moment of the target device is: obtain the maximum value and the minimum value among the normal states of all historical moments of the target device, and then obtain the range of the normal state of the target device according to the maximum value and the minimum value. The fluctuation situation of the normal state is specifically the range of the normal state. The larger the range of the normal state is, the greater the fluctuation of the normal state of the target device is, and the higher the probability of equipment failure is, and remote operation and maintenance are required. Therefore, the device abnormal index of the target device is higher.

[0088] Accordingly, the process of obtaining the abnormal index of the target device may include: obtaining the abnormal index of the target device based on the average value of the normal index, the range of the normal state, and the abnormal state of the target device. Among them, the abnormal index is directly proportional to the abnormal state and the range of the normal state, and inversely proportional to the average value of the normal state.

[0089] In an exemplary embodiment, the range of the normal state of the target device is normalized. The normalization method may be: obtaining the maximum and minimum values of the ranges of the normal states of all target devices, and then normalizing the range of the normal state of the target device by using the maximum-minimum normalization method.

[0090] The calculation formula of the abnormal index is as follows:

[0091] yc = norm[(1 - C)×b×R];

[0092] Where yc is the abnormal index of the target device, C is the average value of the normal state of the target device, b is the abnormal state of the target device, and R is the range of the normal state of the target device after normalization.

[0093] norm is the normalization of [(1 - C)×b×R]. The normalization method here may be: obtaining the maximum and minimum values of (1 - C)×b×R corresponding to all other devices, and then normalizing (1 - C)×b×R of the target device by using the maximum-minimum normalization method to improve the reliability of the operation state monitoring of the target device.

[0094] By using the above process, the abnormal indexes of each device are obtained, and the abnormal indexes of the devices are obtained in the edge server. The higher the abnormal index, the more abnormal the operation state of the corresponding device.

[0095] In an exemplary embodiment, an abnormal threshold may be preset. The numerical range of the preset abnormal threshold is 0 - 1, and the specific value of the preset abnormal threshold is set according to the actual situation. If a more secure monitoring logic is required to avoid missing some infrastructure devices with abnormal operation states, the preset abnormal threshold can be set slightly smaller. In this embodiment, the preset abnormal threshold is taken as an example of 0.7.

[0096] Compare the abnormal indexes of each target device with the preset abnormal threshold. Determine the infrastructure devices greater than or equal to the preset abnormal threshold as abnormal devices; determine the infrastructure devices less than the preset abnormal threshold as normal devices to improve the accuracy of the fault monitoring of the infrastructure devices.

[0097] To further improve the accuracy of the judgment, it may further include

[0098] Obtain the normal device set and the abnormal device set of the above network, and obtain the first environmental characteristics of all normal devices in the above normal set and the second environmental characteristics of all abnormal devices in the above abnormal device set.

[0099] The above environmental characteristics may include temperature, etc. Since the temperature of infrastructure devices is an important factor affecting the operation of infrastructure devices, for example, overheating during operation may cause abnormal conditions in infrastructure devices. Therefore, temperature is used as an environmental factor affecting the monitoring of infrastructure devices.

[0100] For any device in the normal device set and the abnormal device set, a temperature sensor is set in the device (the temperature sensor can be set on the core component, such as the CPU) to detect the temperature of the device according to a preset sampling frequency, so as to obtain the temperature time series data of the device during the observation period, and then calculate the average value of the temperature time series data as the device temperature of the device during the observation period.

[0101] In this embodiment, a standard temperature is preset, such as 25 °C. The smaller the difference between the device temperature of the device during the observation period and the standard temperature, the more normal the device temperature; the larger the temperature difference, the more abnormal the device temperature. Therefore, calculate the absolute value of the difference between the device temperature of the device during the observation period and the standard temperature, and use the obtained absolute value of the temperature difference as the device temperature characteristic of the device during the observation period. In this way, the device temperature characteristics of each device during the observation period are obtained.

[0102] Then calculate the average value of the device temperature characteristics of each normal device in the normal device set during the observation period as the device temperature characteristic of the normal device set during the observation period. Calculate the average value of the device temperature characteristics of each abnormal device in the abnormal device set during the observation period as the device temperature characteristic of the abnormal device set during the observation period.

[0103] Obtain the observed operation and maintenance data of the above target device, and the above observed operation and maintenance data represents the operation and maintenance data of the target device during a preset period after an abnormal alarm is issued.

[0104] Obtain the difference in device temperature characteristics between the normal device set and the abnormal device set during the observation period. In an exemplary embodiment, the difference in device temperature characteristics between the normal device set and the abnormal device set during the observation period is specifically the absolute value of the difference between the device temperature characteristics of the normal device set and the abnormal device set during the observation period.

[0105] Since the differences in the temperature characteristics of the devices and the correlations between the abnormal indicators obtained for the devices twice can affect the more accurate determination of whether the devices are normal, the normal device set and the abnormal device set are updated according to the magnitudes of the differences in the temperature characteristics of the devices and the correlations between the abnormal indicators obtained for the devices twice.

[0106] Preset a temperature characteristic difference threshold, which is used to compare with the temperature characteristic difference of the device to determine whether the temperature characteristic difference of the device is too large. The specific value of the preset temperature characteristic difference threshold is set according to the actual situation. If more stringent monitoring of the temperature is required, the preset temperature characteristic difference threshold can be set slightly smaller, such as 5°C.

[0107] If the temperature characteristic difference of the device is greater than the preset temperature characteristic difference threshold, it indicates that the temperature characteristic differences of the devices in the normal device set and the abnormal device set are relatively large during the observation period. At this time, the abnormal operation of the device may be caused by temperature factors, and the normal performance differences between different devices result in different temperatures.

[0108] Sort the temperature characteristics of each infrastructure device in the normal device set from small to large. Use the temperature characteristics of the device as the abscissa and the device abnormal indicator obtained for the second time for each infrastructure device in the normal device set as the ordinate, and perform a linear fit on the device abnormal indicators obtained for the second time for each infrastructure device in the normal device set after sorting the temperature characteristics of the device to obtain a fitting line. This fitting line reflects the variation relationship between the temperature characteristics of each infrastructure device in the normal device set and the device abnormal indicator. The fitting line represents the degree of variation through the parameter of the slope. At the same time, obtain the root mean square error RMSE of the fitting line. The root mean square error reflects the variation amplitude during the fitting process of the device abnormal indicator.

[0109] Substitute the temperature characteristics of each infrastructure device in the abnormal device set into the fitting line to obtain the fitting abnormal indicators of each infrastructure device in the abnormal device set.

[0110] If there are significant differences between the fitting abnormal indicators of each infrastructure device in the abnormal device set and the device abnormal indicators obtained for the second time for the corresponding infrastructure device, it indicates that the device abnormal indicators obtained for the second time for the infrastructure device do not conform to the normal temperature variation of the device operation. For example, an increase in load leads to a higher CPU occupancy rate. At this time, the corresponding device temperature of the infrastructure device should also be higher, which is a normal phenomenon of device operation. However, if there is a mismatch between the CPU occupancy rate and the temperature, it indicates that the device is abnormal.

[0111] Therefore, according to the differences between the fitting anomaly indicators of the infrastructure devices in the abnormal device set and their device anomaly indicators obtained the second time, as well as the root mean square error, the temperature mismatch degree of each infrastructure device in the abnormal device set is obtained.

[0112] In an exemplary embodiment, the absolute value of the difference between the fitting anomaly indicator of the infrastructure device in the abnormal device set and its device anomaly indicator obtained the second time is calculated, and the difference between the absolute value of the difference and the obtained root mean square error is calculated to obtain the temperature mismatch degree. Taking any one infrastructure device in the abnormal device set as an example, the absolute value of the difference between the fitting anomaly indicator of the infrastructure device and its device anomaly indicator obtained the second time is calculated, and the difference between the absolute value of the difference and the root mean square error obtained in step S9-1 is calculated to obtain the temperature mismatch degree of the infrastructure device. The temperature mismatch degree characterizes the degree to which the anomaly indicator difference of the infrastructure device exceeds the anomaly indicator difference of the normal device. The greater the temperature mismatch degree, the more mismatched the device anomaly indicator of the infrastructure device is with the temperature, and the greater the probability of device failure.

[0113] In the embodiments of the present disclosure, the calculation process of the failure probability may include:

[0114] An adjustment coefficient is obtained according to the temperature mismatch degree, and the adjustment coefficient is obtained by performing a ReLu activation function operation on the temperature mismatch degree;

[0115] According to the adjustment coefficient, the device anomaly indicators obtained the second time for each infrastructure device in the abnormal device set are adjusted to obtain the failure probability of each infrastructure device.

[0116] For any one infrastructure device in the abnormal device set, the calculation formula for the failure probability of the infrastructure device is:

[0117] gz = yc next ×norm[ReLU(TC)];

[0118] where, gz represents the failure probability of the infrastructure device, yc next represents the device anomaly indicator obtained the second time for the infrastructure device, and TC represents the temperature mismatch degree of the infrastructure device.

[0119] ReLU represents the activation function. Then, if TC is less than or equal to 0, ReLU(TC) is equal to 0, and if TC is greater than 0, ReLU(TC) is equal to the value of TC.

[0120] norm[ReLU(TC)] represents the normalization of ReLU(TC), and the normalization method here can be: obtaining the maximum and minimum values of the activation function results corresponding to each infrastructure device in the abnormal device set, and then normalizing the ReLU(TC) of the infrastructure device by using the maximum-minimum normalization method.

[0121] Thus, the failure probabilities of each infrastructure device in the abnormal device set are obtained. The higher the failure probability, the more abnormal the operating state of the infrastructure device.

[0122] In an exemplary embodiment, a failure threshold can be preset. The numerical range of the preset failure threshold is 0 - 1, and the specific value of the preset failure threshold is set according to the actual situation. If a more secure monitoring logic is required, the preset failure threshold can be set slightly smaller. In this embodiment, the preset failure threshold is taken as 0.6 for example.

[0123] Compare the failure probabilities of each infrastructure device in the abnormal device set with the preset failure threshold. For the infrastructure devices in the abnormal device set whose failure probabilities are less than the preset failure threshold, the failure probabilities are smaller, and the corresponding infrastructure devices belong to normal devices. Then, remove the infrastructure devices in the abnormal device set whose failure probabilities are less than the preset failure threshold from the abnormal device set.

[0124] In addition, if the difference in device temperature characteristics is less than or equal to the preset temperature characteristic difference threshold, it means that the difference in device temperature characteristics between the normal device set and the abnormal device set during the observation period is small, and the environments are similar. In this case, theoretically, each infrastructure device should be in the same working state, and the temperature does not affect whether the infrastructure device is abnormal. However, during the peak period of the service, a certain infrastructure device operates at a high load, with a high CPU occupancy rate, and the other infrastructure devices share the pressure of this infrastructure device to reduce the CPU occupancy rate. Therefore, the device abnormality index of the abnormal device set should decrease, and the device abnormality index of the normal device set should increase to a certain extent.

[0125] For any infrastructure device in the normal device set and the abnormal device set, calculate the difference between the device abnormality index obtained for the second time and the device abnormality index obtained for the first time of this infrastructure device as the change in the abnormality index before and after.

[0126] For any abnormal device in the set of abnormal devices, if there is no upward trend in the device anomaly index of the abnormal device, that is, the difference between the device anomaly index obtained the second time and the device anomaly index obtained the first time is less than or equal to 0, which means that the device anomaly index obtained the second time is not greater than the device anomaly index obtained the first time, then the abnormal device is actually operating normally and belongs to the normal device. Therefore, the abnormal device is moved from the set of abnormal devices to the set of normal devices. On the contrary, if there is an upward trend in the device anomaly index of the abnormal device, that is, the abnormal device continues to operate at a high load and there is a trend of more serious anomalies, that is, the difference between the device anomaly index obtained the second time and the device anomaly index obtained the first time is greater than 0, then the abnormal device truly belongs to the abnormal device.

[0127] For any normal device in the set of normal devices, if there is a downward trend in the device anomaly index of the normal device, that is, the difference between the device anomaly index obtained the second time and the device anomaly index obtained the first time is less than 0, which means that the device anomaly index obtained the second time is smaller than the device anomaly index obtained the first time, then the normal device belongs to the abnormal device. Therefore, the normal device is moved from the set of normal devices to the set of abnormal devices. On the contrary, if there is no downward trend in the device anomaly index of the normal device, that is, the difference between the device anomaly index obtained the second time and the device anomaly index obtained the first time is greater than or equal to 0, which means that the device anomaly index obtained the second time is not smaller than the device anomaly index obtained the first time, then the normal device is operating normally.

[0128] It can be understood that this method can be executed by any device, equipment, platform, or device cluster with computing and processing capabilities.

[0129] It should be noted that the method of one or more embodiments of the present application can be executed by a single device, such as a computer or a server. The method of this embodiment can also be applied to a distributed scenario and completed by multiple devices cooperating with each other. In this case of a distributed scenario, one of the multiple devices can only execute one or more steps of the method of one or more embodiments of the present application, and these multiple devices will interact with each other to complete the described method.

[0130] It should be noted that the above describes specific embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be executed in a different order than in the embodiments and still achieve the desired result. Additionally, the processes depicted in the drawings do not necessarily require the specific order or sequential order shown to achieve the desired result. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0131] Based on the same inventive concept, corresponding to the method of any of the above embodiments, the present application further provides a device operation and maintenance apparatus. As Figure 2 shown, the above apparatus includes:

[0132] An acquisition module 11, configured to respond to an abnormal alarm issued by any device in the device network, regard the device as a target device, and acquire the historical operation and maintenance data of the target device and other devices in the device network except the target device;

[0133] A first calculation module 12, configured to obtain an abnormal index of the target device according to the historical operation and maintenance data of the target device, where the abnormal index of the target device indicates the abnormal degree and the change trend of the abnormal degree of the target device in the historical period;

[0134] A second calculation module 13, configured to obtain a normal index of the target device according to the historical operation and maintenance data of the target device and the historical operation and maintenance data of the other devices, where the normal index of the target device indicates the similarity of the abnormal conditions of the target device and the other devices in the historical period;

[0135] A third calculation module 14, configured to obtain an abnormal index of the target device according to the abnormal index and the normal index of the target device, where the abnormal index is used to formulate an operation and maintenance strategy for the target device.

[0136] For the convenience of description, when describing the above apparatus, it is divided into various modules according to functions and described separately. Of course, when implementing one or more embodiments of the present application, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0137] The apparatus of the above embodiment is used to implement the corresponding method in the foregoing embodiment, and has the beneficial effects of the corresponding method embodiment, which will not be elaborated herein.

[0138] Figure 3 FIG. shows a more specific schematic diagram of the hardware structure of an electronic device provided in this embodiment. The device may include: a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. Wherein, the processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040 are communicatively connected to each other inside the device through the bus 1050.

[0139] The processor 1010 can be implemented in the form of a general-purpose CPU (Central Processing Unit), a microprocessor, an Application Specific Integrated Circuit (ASIC), or one or more integrated circuits, etc., and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application.

[0140] The memory 1020 can be implemented in the form of a ROM (Read Only Memory), a RAM (Random Access Memory), a static storage device, a dynamic storage device, etc. The memory 1020 can store an operating system and other application programs. When implementing the technical solutions provided in the embodiments of the present application through software or firmware, the relevant program codes are stored in the memory 1020 and called and executed by the processor 1010.

[0141] The input / output interface 1030 is used to connect to the input / output module to achieve information input and output. The input / output module can be configured as a component in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. The input device can include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device can include a display, a speaker, a vibrator, an indicator light, etc.

[0142] The communication interface 1040 is used to connect to a communication module (not shown in the figure) to achieve communication interaction between this device and other devices. The communication module can achieve communication in a wired manner (such as USB, network cable, etc.) or in a wireless manner (such as mobile network, WIFI, Bluetooth, etc.).

[0143] The bus 1050 includes a path for transmitting information between various components of the device (such as the processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040).

[0144] It should be noted that although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040, and the bus 1050, in the specific implementation process, the device may also include other components necessary for normal operation. In addition, those skilled in the art can understand that the above device may also only include the components necessary to implement the solutions of the embodiments of the present application, and do not have to include all the components shown in the figure.

[0145] The electronic device in the above embodiment is used to implement the corresponding method in the foregoing embodiment and has the beneficial effects of the corresponding method embodiment, which will not be elaborated herein.

[0146] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tapes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device.

[0147] Those of ordinary skill in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of the present disclosure (including the claims) is limited to these examples; within the spirit of the present disclosure, the technical features in the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations in different aspects of one or more embodiments of the present application as described above, and for the sake of brevity, they are not provided in detail.

[0148] In addition, for the sake of simplicity of explanation and discussion, and in order not to make one or more embodiments of the present application difficult to understand, the well-known power / ground connections to integrated circuit (IC) chips and other components may or may not be shown in the provided drawings. Further, the devices may be shown in block diagram form in order to avoid making one or more embodiments of the present application difficult to understand, and this also takes into account the fact that the details of the implementation of these block diagram devices are highly dependent on the platform on which one or more embodiments of the present application are to be implemented (i.e., these details should be fully within the understanding of those skilled in the art). In cases where specific details (such as circuits) are set forth to describe exemplary embodiments of the present disclosure, it will be apparent to those skilled in the art that one or more embodiments of the present application can be implemented without these specific details or with variations of these specific details. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0149] Although the present disclosure has been described in connection with specific embodiments of the present disclosure, many alternatives, modifications, and variations of these embodiments will be apparent to those of ordinary skill in the art based on the foregoing description. For example, other memory architectures (such as dynamic RAM (DRAM)) can be used with the embodiments discussed.

[0150] One or more embodiments of the present application are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of one or more embodiments of the present application shall be included within the protection scope of the present disclosure.

Claims

1. A device operation and maintenance method, characterized in that, Including: In response to an abnormal alarm issued by any device in the device network, regarding the device as a target device and obtaining the historical operation and maintenance data of the target device and other devices in the device network except the target device; Based on the historical operation and maintenance data of the target device, obtaining the abnormal index of the target device, where the abnormal index of the target device indicates the abnormal degree and the change trend of the abnormal degree of the target device in the historical period; Based on the historical operation and maintenance data of the target device and the historical operation and maintenance data of other devices, obtaining the normal index of the target device, where the normal index of the target device indicates the similarity of the abnormal conditions of the target device and other devices in the historical period; Based on the abnormal index and the normal index of the target device, obtaining the abnormal indicator of the target device, where the abnormal indicator is used to formulate an operation and maintenance strategy for the target device.

2. The method according to claim 1, wherein The abnormal index is obtained by performing the following operations: Based on the historical operation and maintenance data of the device, obtaining the abnormal degree of the device at each historical moment in the historical period, and the change trend of the abnormal degree over time; Based on the abnormal degree of the device at each historical moment and the change trend of the abnormal degree, obtaining the abnormal index.

3. The method according to claim 2, wherein The normal index is obtained by performing the following operations: Based on the historical operation and maintenance data of the first device and the second device, obtaining the abnormal conditions of the first device and the second device at each historical moment in the historical period, where the first device and the second device are any two devices in the device network; Based on the abnormal conditions of the first device at each historical moment in the historical period and the abnormal conditions of the second device at each historical moment in the historical period, obtaining the abnormal similarity of the first device and the second device at each historical moment in the historical period, and the average value of the abnormal degrees of the first device and the second device at each historical moment in the historical period; Based on the abnormal similarity of the first device and the second device at each historical moment in the historical period, and the average value of the abnormal degrees of the first device and the second device at each historical moment in the historical period, obtaining the normal index of the first device.

4. The method according to claim 3, wherein The abnormal indicator is obtained by performing the following operations: Determining the operating state of the target device at each historical moment in the historical period, where the operating state includes a normal state and an abnormal state; Based on the state values of the target device at each historical moment in the historical period, obtaining the average value of the normal state and the extreme difference of the normal state of the target device in the historical period; Based on the average value of the normal state, the extreme difference of the normal state of the target device in the historical period, and the abnormal state value of the target device, obtaining the abnormal indicator of the target device.

5. The method according to claim 2, wherein Also including: Obtaining the normal device set and the abnormal device set of the device network, and obtaining the first environmental characteristics of all normal devices in the normal device set and the second environmental characteristics of all abnormal devices in the abnormal device set; Obtaining the observed operation and maintenance data of the target device, where the observed operation and maintenance data represents the operation and maintenance data in a preset time period after the target device issues an abnormal alarm; Based on the observed operation and maintenance data of the target device, an observed anomaly index of the target device is obtained, and the observed anomaly index is obtained through the same calculation steps as the anomaly index; Based on the anomaly index, observed anomaly index, first environmental feature, and second environmental feature of the target device, it is determined whether the target device is an abnormal device.

6. The method according to claim 5, wherein The determining whether the target device is an abnormal device based on the anomaly index, observed anomaly index, first environmental feature, and second environmental feature of the target device includes: In response to determining that the distance between the first environmental feature and the second environmental feature is less than or equal to a preset threshold, and the observed anomaly index of the target device is less than or equal to the anomaly index of the target device, it is determined that the target device is a normal device; In response to determining that the distance between the first environmental feature and the second environmental feature is greater than the preset threshold, or the observed anomaly index of the target device is greater than the anomaly index of the target device, it is determined that the target device is an abnormal device.

7. The method according to claim 5, characterized in that It further includes: In response to determining that the difference between the first environmental feature and the second environmental feature is greater than a pre-examination threshold, the observed operation and maintenance data of all devices in the device network are obtained; Based on the observed operation and maintenance data of all devices in the device network, the observed anomaly indexes of all devices in the device network are obtained; Performing a linear fit on the first environmental features and observed anomaly indexes of all normal devices in the normal device set to obtain a fitting straight-line model and the root mean square error corresponding to the fitting straight-line model; Based on the fitting straight-line model and the second environmental features of all abnormal devices in the abnormal device set, the fitting anomaly indexes of all abnormal devices in the abnormal device set are obtained; Based on the observed anomaly indexes, fitting anomaly indexes of the abnormal devices in the abnormal device set, and the root mean square error corresponding to the fitting straight-line model, the temperature mismatch degree of the abnormal devices in the abnormal device set is obtained; Based on the temperature mismatch degree and fitting anomaly index of the abnormal devices in the abnormal device set, the failure probabilities of the abnormal devices in the abnormal device set are obtained; Based on the failure probabilities of the abnormal devices in the abnormal device set, the abnormal device set is updated.

8. An equipment operation and maintenance device, characterized in that, It includes: An acquisition module, configured to, in response to an abnormal alarm sent by any device in the device network, regard the device as a target device and acquire the historical operation and maintenance data of the target device and other devices in the device network except the target device; A first calculation module, configured to obtain an anomaly index of the target device based on the historical operation and maintenance data of the target device, and the anomaly index of the target device indicates the anomaly degree and the change trend of the anomaly degree of the target device in the historical period; A second calculation module, configured to obtain a normal index of the target device based on the historical operation and maintenance data of the target device and the historical operation and maintenance data of the other devices, and the normal index of the target device indicates the similarity of the anomaly conditions of the target device and the other devices in the historical period; A third computing module, configured to obtain an anomaly metric of the target device according to the anomaly index and the normal index of the target device, where the anomaly metric is used to formulate an operation and maintenance strategy for the target device.

9. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the program, the method described in any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium stores computer instructions for causing the computer to execute the method described in any one of claims 1 to 7.

Images