Equipment monitoring method and device, electronic equipment and medium
By collecting equipment operation data multiple times and using a multi-head attention mechanism and attenuation weights to correct risk values, the problem of false alarms caused by single-point-of-time monitoring is solved, and the accuracy and timeliness of equipment fault monitoring are achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-26
- Publication Date
- 2026-03-31
AI Technical Summary
Existing technologies for equipment monitoring based on a single point in time are prone to false alarms due to data noise, leading to over-detection of occasional anomalies and low accuracy in fault monitoring.
By collecting equipment operation data multiple times, the risk value is corrected. The risk value is dynamically corrected using a multi-head attention mechanism and attenuation weight to ensure the accuracy of fault monitoring.
It improves the accuracy of fault monitoring, avoids false alarms caused by data noise, and ensures the timeliness and accuracy of equipment status assessment.
Smart Images
Figure CN121770975A_ABST
Abstract
Description
Technical Field
[0001] This application belongs to the field of equipment monitoring technology, specifically relating to an equipment monitoring method, device, electronic equipment, and medium. Background Technology
[0002] In the field of industrial IoT equipment status monitoring and risk management, existing technical solutions typically rely on equipment operation data reported by equipment sensors. They use deep learning models such as LSTM (Long Short-Term Memory) to analyze the data reported in a single instance, outputting real-time risks. Once the risk is determined to exceed a preset threshold, an alarm is immediately triggered and statically associated response measures (such as fixed access restrictions) are executed.
[0003] However, judgments based on a single point in time are prone to false alarms due to data noise, leading to over-detection of occasional anomalies. Summary of the Invention
[0004] Based on the technical problem of low fault monitoring accuracy in related equipment monitoring schemes, this application proposes an equipment monitoring method, device, electronic equipment, and medium.
[0005] A first aspect of this application provides a device monitoring method, the method comprising: If the risk value determined by the currently collected equipment operation data is higher than the first threshold, a preset number of equipment operation data will be collected, starting from the currently collected equipment operation data. The risk value is corrected based on multiple collections of equipment operation data; If the corrected risk value is higher than the second threshold, the monitored equipment is determined to be faulty.
[0006] In some embodiments of this application, the step of correcting the risk value based on multiple collected device operation data includes: Determine the correction factor for the risk value based on each collected device operation data; The risk value is corrected using the correction factor corresponding to each collected device operation data.
[0007] In some embodiments of this application, determining the correction factor for the risk value based on each collected device operation data includes: Perform linear transformations on the device operation data collected each time to generate the query vector, key vector, and value vector corresponding to the device operation data collected each time. Using the query vector and key vector corresponding to each collected device operation data, calculate the attention weight corresponding to each collected device operation data; The attention weights corresponding to each collected device operation data are combined into a weight sequence; Using the weight sequence and the value vector corresponding to the equipment operation data collected each time, the correction factor of the risk value for each collected equipment operation data is calculated.
[0008] In some embodiments of this application, the step of correcting the risk value using a correction factor corresponding to each collected device operation data includes: The attenuation weight corresponding to each piece of equipment operation data is determined based on the time difference between each piece of equipment operation data and the starting point. The risk value is corrected using the correction factor and attenuation weight corresponding to each collected device operation data.
[0009] In some embodiments of this application, the attenuation weight corresponding to each collected device operation data decreases exponentially as the time difference increases.
[0010] In some embodiments of this application, the method further includes: If the corrected risk value is higher than the second threshold, determine the target risk level to which the corrected risk value belongs; The access permission level and access duration are determined based on the target risk level. The access permission level is used to indicate the degree of access to equipment data that maintenance personnel can access, and the access duration is used to indicate the duration for which equipment data is accessed to maintenance personnel. The output includes a fault warning that includes the open permission level and the duration of the open permission.
[0011] In some embodiments of this application, the method further includes: Based on the daily usage time of the monitored device, the peak usage period and the off-peak usage period of the monitored device are determined; the peak usage period is used to indicate the time period when the device usage frequency is higher than a first frequency threshold, and the off-peak usage period is used to indicate the time period when the device usage frequency is lower than a second frequency threshold; The output includes fault warnings for the open permission level, the permission open duration, the peak usage period, and the off-peak usage period.
[0012] A second aspect of this application provides a device monitoring apparatus, the apparatus comprising: The data acquisition module is used to collect a preset number of equipment operation data points, starting from the currently collected equipment operation data, if the risk value determined by the currently collected equipment operation data is higher than the first threshold. The risk correction module is used to correct the risk value based on multiple collections of equipment operation data; The fault diagnosis module is used to determine that the monitored equipment is faulty if the corrected risk value is higher than the second threshold.
[0013] An embodiment of the third aspect of this application provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor running the computer program to implement the method as described in the first aspect.
[0014] An embodiment of the fourth aspect of this application provides a computer-readable storage medium storing computer-executable instructions for causing a computer to perform the method as described in the first aspect.
[0015] The device monitoring method, apparatus, electronic device, and medium provided in the embodiments of this application have at least the following beneficial effects: If the risk value is deemed high based on a single reported data, an early warning is not triggered immediately. Instead, the risk value is corrected by obtaining multiple reported data. If the corrected risk value is still high, a fault is confirmed in the monitored equipment, thus triggering an early warning. This ensures the accuracy of fault monitoring and avoids false alarms caused by data noise.
[0016] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description
[0017] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 A flowchart illustrating a device monitoring method provided in some embodiments of this application; Figure 2 A flowchart illustrating a specific implementation of risk assessment provided for some embodiments of this application; Figure 3 This is a schematic diagram of the structure of a device monitoring apparatus provided in some embodiments of this application; Figure 4 This is a schematic diagram of the structure of an electronic device provided in some embodiments of this application. Detailed Implementation
[0018] Exemplary embodiments of this application will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of this application are shown in the drawings, it should be understood that this application may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of this application and to fully convey the scope of this application to those skilled in the art.
[0019] It should be noted that, unless otherwise stated, the technical or scientific terms used in this application shall have the ordinary meaning as understood by one of ordinary skill in the art to which this application pertains.
[0020] As mentioned earlier, the model makes risk assessments based on data reported at a single point in time, which can easily lead to false alarms due to data noise. The lack of observation and secondary verification of risk evolution can easily result in overreacting to occasional anomalies.
[0021] Based on this, this application provides a device monitoring method. In this method, if the risk value is determined to be high based on a single reported device operation data, an early warning is not triggered immediately. Instead, the risk value is corrected by further obtaining multiple reported device operation data. If the corrected risk value is still high, it is determined that the monitored device has a fault, that is, an early warning is triggered. This can ensure the accuracy of fault monitoring and avoid false alarms caused by data noise.
[0022] In this embodiment, the monitored device can be any type of IoT device. The reported device operation data can include data from different types of sensors on the monitored device.
[0023] In an exemplary scenario, assuming the monitored device is an air conditioning unit, the reported device operation data can include various types of sensor data, such as the device's physical parameters (e.g., temperature, pressure, vibration amplitude, rotation speed, etc.), electrical parameters (e.g., current, voltage, power, etc.), operating status indicators (e.g., device load rate, running time, etc.), and some environmental parameters (e.g., the temperature and humidity of the environment in which the device is located).
[0024] Some embodiments of the device monitoring method proposed in this application are described in [reference]. Figure 1 The flowchart shown is a device monitoring method. This method can be executed by a device monitoring device, which can be implemented in hardware and / or software. The device monitoring device can be configured in any electronic device with network communication and computing capabilities. The electronic device can be a mobile control terminal of the monitored device. The method specifically includes the following steps 101-103.
[0025] Step 101: If the risk value determined based on the currently collected equipment operation data is higher than the first threshold, collect a preset number of equipment operation data starting from the currently collected equipment operation data.
[0026] Step 102: Correct the risk value based on the equipment operation data collected multiple times.
[0027] Step 103: If the corrected risk value is higher than the second threshold, it is determined that the monitored equipment is faulty.
[0028] The currently collected equipment operation data consists of various information reflecting the operating status of the monitored equipment, reported in real time. This data can cover multiple aspects, such as physical parameters like temperature, pressure, vibration frequency, rotational speed, current, and voltage, as well as log information and operation records generated during equipment operation.
[0029] For example, each time the monitored device reports its operating data, a risk value is determined. When the determined risk value is higher than a first threshold, the process of steps 101-103 is triggered.
[0030] The risk value is a numerical indicator calculated using a specific algorithm or model based on currently collected equipment operating data. The risk value is used to quantify the degree of risk that may exist in the current operating state of the equipment.
[0031] In an exemplary scenario, a multi-layer LSTM neural network is constructed. Its input layer receives device runtime sequence data, and the hidden layer of the network uses two layers of LSTM units, each containing 128 neurons. Long-term dependencies are captured through a gating mechanism. The output layer is connected to an activation function, which maps the final hidden state h of the LSTM to a k-dimensional risk probability distribution vector P, where k is the number of risk levels, for example, k=3-dimensional. These 3-dimensional vectors correspond to the risk probabilities of low-risk, medium-risk, and high-risk levels, respectively. When the risk probability of medium-risk or high-risk exceeds the first threshold, the process of steps 101-103 is triggered.
[0032] The preset amount of equipment operation data can be understood as data within a fixed-duration observation window, starting from the currently collected equipment operation data. In other words, once the risk value of the equipment operation data at a certain point in time triggers a threshold condition, the system will continue to observe whether there is indeed a fault at that point in time.
[0033] In this embodiment of the application, the risk value determined based on the currently collected equipment operation data can be regarded as an initial risk value. This initial risk value may be based on only limited data or contain some uncertain factors. By introducing data collected multiple times in the future, this initial risk value can be corrected with more comprehensive information, which can more accurately reflect the overall operational risk of the equipment over a period of time. When the corrected risk value is higher than the second threshold, it indicates that after a more comprehensive and accurate analysis, the operating status of the equipment does indeed have a risk that exceeds the normal range, and thus it can be determined that the monitored equipment has a fault.
[0034] It is understood that this application does not specifically limit the relationship between the first threshold and the second threshold mentioned above. The two thresholds may be the same or different, and both are used as risk assessment.
[0035] In some embodiments of this application, the risk value correction process in step 102 above may specifically include: Determine the correction factor for this risk value for each collected equipment operation data; The risk value is corrected using the correction factor corresponding to each collected device operation data.
[0036] The correction factor can be understood as a numerical indicator used to quantify the impact of subsequently collected equipment operation data on the initial risk value. Each collection of equipment operation data has a correction factor for that risk value, which reflects the correlation between the collected data and the initial risk value, as well as the adjustment effect of the data on risk assessment.
[0037] For example, suppose five data points are collected subsequently: data 1, data 2, data 3, data 4, and data 5. Then the risk correction factor for data 1 is a, the risk correction factor for data 2 is b, the risk correction factor for data 3 is c, the risk correction factor for data 4 is d, and the risk correction factor for data 5 is e.
[0038] In this embodiment, the correction factor for the risk value is different for the equipment operation data collected at different time points. By using multiple different correction factors to dynamically correct the initial risk value, the corrected risk value can reflect the overall operational risk of the equipment over a period of time.
[0039] In some embodiments of this application, the process of determining the correction factor for the risk value based on each collected device operation data can be implemented using a multi-head attention mechanism, specifically including: Perform linear transformations on the device operation data collected each time to generate the query vector, key vector, and value vector corresponding to the device operation data collected each time. Using the query vector and key vector corresponding to each collected device operation data, calculate the attention weight corresponding to each collected device operation data; The attention weights corresponding to each collected device operation data are combined into a weight sequence; Using the weight sequence and the value vector corresponding to the equipment operation data collected each time, the correction factor for the risk value is calculated for each collected equipment operation data.
[0040] Specifically, by generating query vectors, key vectors, and value vectors through linear transformations, attention features of device operation data can be obtained from different perspectives. For each collection of device operation data, a set of query vectors, key vectors, and value vectors can be obtained. The linear transformation can be implemented by multiplying the input data by a weight matrix and adding a bias vector.
[0041] The query vector represents the information or feature that needs to be focused on. The key vector is matched against the query vector to determine the relevance of each collected device operation data point to the current query requirement. The value vector contains information from the device operation data actually used to correct risk values.
[0042] Attention weights represent the relevance of device operation data to the current query requirement. The calculation method for attention weights typically employs a dot product attention mechanism.
[0043] Specifically, for each set of device operation data collected, the dot product of the query vector Q and the key vector K is calculated, and then the dot product result is normalized using a softmax function to obtain the attention weights. α。
[0044] Right now α =softmax(Q T K) For example, if there is data collected from three different devices, the dot product of their query vector and key vector is calculated for each, and then the three attention weights are obtained through the softmax function. α 1. α 2 and α 3. These weights reflect the relative importance of each data collection in the current risk assessment.
[0045] A weighted sequence is a sequence formed by arranging the attention weights corresponding to each collected device operation data in a certain order. This order is usually consistent with the data collection order, but can also be adjusted according to actual needs.
[0046] For example, weight sequence [ α 1, α 2, α[3] represents the attention weights corresponding to the three data collections. When calculating the correction factor, weighted summation and other operations can be performed based on this sequence.
[0047] Furthermore, the correction factor is calculated using the multi-head merging function sum_over_heads, which utilizes the weight sequence and value vector, i.e., f(j) = sum_over_heads(W * Vj), where W represents the weight sequence, Vj represents the value vector corresponding to the equipment operation data collected in the jth time, and f(j) represents the correction factor of the equipment operation data collected in the jth time for this risk value.
[0048] In some embodiments of this application, the process of correcting the risk value using a correction factor corresponding to each collected device operation data may specifically include: The attenuation weight corresponding to each piece of equipment operation data is determined based on the time difference between each piece of equipment operation data and the starting point. The risk value is corrected using the correction factor and attenuation weight corresponding to each collected device operation data.
[0049] The time difference refers to the time interval between the point in time corresponding to each piece of equipment operation data collected and the starting point. This time difference reflects the order of data collection and the distribution of data over time.
[0050] The attenuation weight is a numerical value used to measure how the impact of each collected device operation data point on risk value correction changes over time. It is determined based on the time difference, and generally follows a pattern where the attenuation weight gradually decreases as the time difference increases. This is because, over time, the further the collected data is from the starting point, the less important it may be for the current device risk assessment; conversely, data collected closer to the starting point better reflects the current operating status and risk situation of the device.
[0051] During the correction, firstly, obtain the risk value corresponding to each collected device operation data. The method for obtaining the risk value can be found in the calculation principle of the initial risk value mentioned above.
[0052] Then, the difference between the risk value and the initial risk value corresponding to each data collection, the correction factor and the attenuation weight can be multiplied together, and all the product results can be summarized to obtain the correction amount for the initial risk value. Finally, the correction amount is added to the initial risk value to obtain the corrected risk value.
[0053] The specific calculation formula can be expressed as follows:
[0054] In the above formula, This represents the risk value of the currently collected equipment operation data, also known as the initial risk value. This represents the attenuation weight corresponding to the device operation data collected in the j-th iteration. This represents the correction factor corresponding to the device operation data collected in the j-th iteration, which is also the correction factor based on attention features. This represents the risk value of the device operation data collected in the j-th iteration.
[0055] In this embodiment, the corrected risk value can more accurately reflect the current operational risk status of the equipment. By introducing attenuation weights and considering the temporal validity of the data, data closer to the starting point has a greater impact on the correction of the risk value, while data farther from the starting point has a relatively smaller impact, thereby improving the timeliness and accuracy of risk assessment.
[0056] In one specific implementation, the relationship between attenuation weight and time difference is as follows: the magnitude of the attenuation weight corresponding to each collected device operation data decreases exponentially as the time difference increases.
[0057] For example, a decay weight function can be defined:
[0058] in, Indicates the attenuation coefficient. This represents the time difference between each collected device operation data point and the starting point.
[0059] As can be seen from the above decay weight function, the corresponding decay weight decays exponentially as the time difference increases.
[0060] In other embodiments of this application, the method may further include: If the corrected risk value is higher than the second threshold, determine the target risk level to which the corrected risk value belongs; The access permission level and access duration are determined based on the target risk level. The access permission level indicates the extent to which maintenance personnel can access the equipment data, and the access duration indicates the duration for which the equipment data is accessible to maintenance personnel. The output includes fault warnings that specify the level of access permission and the duration of permission access.
[0061] The risk levels can be divided into three categories: low risk, medium risk, and high risk, with each level corresponding to a risk value range. It's understandable that the risk value ranges for medium and high risk are both higher than the aforementioned second threshold, ensuring that fault warnings are triggered under medium or high risk conditions.
[0062] Remote maintenance and equipment status viewing permissions can be divided into two levels: Level 1 and Level 2. Level 1 permissions have a higher level of access than Level 2 permissions, and the duration of Level 1 access is also longer than that of Level 2 access.
[0063] For example, first-level access allows viewing equipment engineering parameters, while second-level access allows controlling the equipment's on / off state, temperature, and viewing historically reported data.
[0064] In the specific allocation, for medium-risk levels, the corresponding level 2 access permissions are allocated, and for high-risk levels, the corresponding level 1 access permissions are allocated.
[0065] Fault warnings can be understood as a type of risk report. Based on the output fault warnings, users can grant recommended access levels and access durations, or they can contact maintenance personnel to adjust the access levels.
[0066] In other embodiments of this application, the method may further include: Based on the daily usage time of the monitored equipment, the peak and off-peak usage periods of the monitored equipment are determined; the peak usage period is used to indicate the time when the frequency of equipment use is higher than the first frequency threshold, and the off-peak usage period is used to indicate the time when the frequency of equipment use is lower than the second frequency threshold. The output includes fault warnings such as the level of access permission, the duration of permission access, peak usage periods, and off-peak usage periods.
[0067] Among them, the frequency of device use at each time point can be counted based on the daily usage time of the monitored device. Then, based on the device usage frequency at different time points, the first frequency threshold, and the second frequency threshold, the peak and low usage periods of the device can be obtained.
[0068] In other words, by analyzing the daily usage time of the equipment, the system can identify peak and off-peak usage periods and recommend them to users. This allows maintenance personnel to choose to perform remote debugging during off-peak periods or to conduct on-site debugging during peak periods, thereby reducing the disruption to users' normal use of the equipment.
[0069] To facilitate understanding of the device monitoring process provided in the embodiments of this application, the following is in conjunction with the appendix. Figure 2 A brief explanation is provided. Figure 2 A flowchart illustrating a specific implementation of risk assessment provided in some embodiments of this application is shown. The implementation process specifically includes: First, the monitored device reports raw data, i.e., device operation data, which is parsed and stored in a database to obtain a time series sequence. The device operation data at each time point in the time series is input into a pre-constructed LSTM neural network. The LSTM neural network outputs a k-dimensional risk probability distribution vector P, where k represents the number of risk levels (e.g., k=3 dimensions), corresponding to the risk probabilities of low, medium, and high risk levels, respectively. When the probability of a medium or high risk level exceeds a first threshold, the current time point t0 is marked as a high-risk trigger point. This high-risk trigger point can be viewed by the user, but a warning notification is not immediately issued; instead, the system continues to monitor whether the state at that point is indeed a high-risk point.
[0070] Then, starting from trigger point t0, a fixed-duration observation window is opened. Within this window, subsequent device operation data (e.g., from time t1 to time tn) is continuously collected. For this data, a multi-head attention mechanism is used for analysis: the device operation data at each time point is transformed linearly to generate a query vector Q, a key vector K, and a value vector V; the attention weight corresponding to the device operation data at each time point is calculated; and the correction factor for the risk value at time t0 is calculated through multi-head merging. Finally, the attenuation weight corresponding to the equipment operation data at each time point is calculated by a predefined attenuation weight function. Using the attenuation weight and correction factor, the difference between the risk value of the equipment operation data at time t1-tn and the risk value at time t0 is weighted and corrected. The sum of the weighted correction result and the risk value at time t0 is taken as the corrected risk value.
[0071] The final determination is based on the range of the corrected risk value: if the corrected risk value is greater than or equal to the second threshold high, then time t0 is confirmed as the fault point; otherwise, time t0 is marked as low risk.
[0072] Understandably, the final judgment result can be fed back into the LSTM neural network as a training sample to achieve continuous optimization of the model.
[0073] Some embodiments of this application also provide a device for monitoring equipment, which is used to execute the device monitoring method provided in the foregoing embodiments. See [link to relevant documentation]. Figure 3 The device specifically includes: The data acquisition module 310 is used to collect a preset number of equipment operation data starting from the current equipment operation data, based on the risk value determined by the currently collected equipment operation data being higher than a first threshold. The risk correction module 320 is used to correct the risk value based on multiple collections of equipment operation data; The fault determination module 330 is used to determine that the monitored equipment has a fault when the corrected risk value is higher than the second threshold.
[0074] In some embodiments of this application, the risk correction module 320 is specifically used to determine the correction factor of the risk value for each collected device operation data; and to correct the risk value using the correction factor corresponding to each collected device operation data.
[0075] In some embodiments of this application, the risk correction module 320 is specifically used to perform linear transformations on each collected device operation data to generate query vectors, key vectors, and value vectors corresponding to each collected device operation data during the process of determining the correction factor of each collected device operation data on the risk value; calculate the attention weights corresponding to each collected device operation data using the query vectors and key vectors corresponding to each collected device operation data; form a weight sequence using the attention weights corresponding to each collected device operation data; and calculate the correction factor of each collected device operation data on the risk value using the weight sequence and the value vectors corresponding to each collected device operation data.
[0076] In some embodiments of this application, the risk correction module 320 is specifically used to determine the attenuation weight corresponding to each piece of equipment operation data based on the time difference between each piece of equipment operation data and the starting point during the process of correcting the risk value using the correction factor corresponding to each piece of equipment operation data collected; and to correct the risk value using the correction factor and attenuation weight corresponding to each piece of equipment operation data collected.
[0077] In some embodiments of this application, the attenuation weight corresponding to each collected device operation data decreases exponentially as the time difference increases.
[0078] In some embodiments of this application, the apparatus further includes ( Figure 3 (not shown in the image) The access control module is used to determine the target risk level of the corrected risk value when the corrected risk value is higher than a second threshold; determine the access level and access duration based on the target risk level, wherein the access level indicates the degree of access to equipment data that maintenance personnel can access, and the access duration indicates the duration for which equipment data is accessible to maintenance personnel; and output a fault warning containing the access level and access duration.
[0079] In some embodiments of this application, the permission management module is further configured to determine the peak usage period and off-peak usage period of the monitored device based on the daily usage time of the monitored device; the peak usage period is used to indicate the time period when the device usage frequency is higher than a first frequency threshold, and the off-peak usage period is used to indicate the time period when the device usage frequency is lower than a second frequency threshold; and outputs a fault warning including the open permission level, the permission open duration, the peak usage period, and the off-peak usage period.
[0080] The device monitoring apparatus and the device monitoring method provided in the above embodiments of this application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the applications stored therein.
[0081] This application also provides an electronic device for performing the above-described device monitoring method.
[0082] Please refer to Figure 4 This illustrates a schematic diagram of an electronic device provided by some embodiments of this application. For example... Figure 4 As shown, the electronic device 40 includes: a processor 400, a memory 401, a bus 402, and a communication interface 403. The processor 400, the communication interface 403, and the memory 401 are connected via the bus 402. The memory 401 stores a computer program that can run on the processor 400. When the processor 400 runs the computer program, it executes the device monitoring method provided in any of the foregoing embodiments of this application.
[0083] The memory 401 may include high-speed random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Communication between this device network element and at least one other network element is achieved through at least one communication interface 403 (which can be wired or wireless), such as the Internet, wide area network, local area network, metropolitan area network, etc.
[0084] Bus 402 can be an ISA bus, PCI bus, or EISA bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. The memory 401 is used to store programs. After receiving an execution instruction, the processor 400 executes the program. The device monitoring method disclosed in any of the foregoing embodiments of this application can be applied to the processor 400, or implemented by the processor 400.
[0085] The processor 400 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of the processor 400 or by instructions in software form. The processor 400 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules may reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in memory 401. The processor 400 reads the information in memory 401 and, in conjunction with its hardware, completes the steps of the above method.
[0086] The electronic device and the device monitoring method provided in this application are based on the same inventive concept and have the same beneficial effects as the methods they employ, operate, or implement.
[0087] This application also provides a computer-readable storage medium corresponding to the device monitoring method provided in the foregoing embodiments. The computer-readable storage medium shown is an optical disc, on which a computer program (i.e., a program product) is stored. When the computer program is run by a processor, it executes the device monitoring method provided in any of the foregoing embodiments.
[0088] It should be noted that examples of the computer-readable storage medium may also 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 optical and magnetic storage media, which will not be elaborated here.
[0089] The computer-readable storage medium provided in the above embodiments of this application and the device monitoring method provided in the embodiments of this application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the applications stored therein.
[0090] It should be noted that: Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of this application may be practiced without these specific details. In some instances, well-known structures and techniques have not been shown in detail so as not to obscure the understanding of this specification.
[0091] Similarly, it should be understood that, for the sake of brevity and to aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of this application, various features of this application are sometimes grouped together in a single embodiment, figure, or description thereof. However, this disclosure should not be interpreted as reflecting the following schematic diagram: that is, the claimed application requires more features than are expressly recited in each claim. Rather, as reflected in the following claims, inventive aspects lie in fewer than all features of a single foregoing disclosed embodiment. Therefore, the claims following the detailed description are hereby expressly incorporated into this detailed description, wherein each claim itself is a separate embodiment of this application.
[0092] Furthermore, those skilled in the art will understand that although some embodiments described herein include certain features but not others included in other embodiments, combinations of features from different embodiments are intended to be within the scope of this application and form different embodiments. For example, in the following claims, any of the claimed embodiments can be used in any combination.
[0093] The above description is merely a preferred embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.
Claims
1. A device monitoring method characterized by, The method comprises: determining, according to a risk value determined based on currently collected equipment operation data, that the risk value is higher than a first threshold value, collecting a preset number of equipment operation data from the currently collected equipment operation data as a starting point; correcting the risk value based on the equipment operation data collected multiple times; determining that the monitored equipment has a fault when the corrected risk value is higher than a second threshold value.
2. The method of claim 1, wherein, The correction of the risk value based on the equipment operation data collected multiple times comprises: determining a correction factor of the risk value for each time of collecting the equipment operation data; correcting the risk value by using the correction factor corresponding to each time of collecting the equipment operation data.
3. The method of claim 2, wherein, The determination of the correction factor of the risk value for each time of collecting the equipment operation data comprises: performing linear transformation on each time of collecting the equipment operation data to generate a query vector, a key vector and a value vector corresponding to each time of collecting the equipment operation data; calculating an attention weight corresponding to each time of collecting the equipment operation data by using the query vector and the key vector corresponding to each time of collecting the equipment operation data; combining the attention weights corresponding to each time of collecting the equipment operation data into a weight sequence; calculating the correction factor of the risk value for each time of collecting the equipment operation data by using the weight sequence and the value vector corresponding to each time of collecting the equipment operation data.
4. The method of claim 2, wherein, The correction of the risk value by using the correction factor corresponding to each time of collecting the equipment operation data comprises: determining a decay weight corresponding to each time of collecting the equipment operation data according to a time difference of each time of collecting the equipment operation data relative to the starting point; correcting the risk value by using the correction factor and the decay weight corresponding to each time of collecting the equipment operation data.
5. The method of claim 4, wherein, The decay weight corresponding to each time of collecting the equipment operation data decreases exponentially with an increase in the time difference.
6. The method according to any one of claims 1 to 5, characterized in that, The method further comprises: determining a target risk level to which the corrected risk value belongs when the corrected risk value is higher than the second threshold value; determining an open permission level and a permission open time length according to the target risk level, the open permission level being used to indicate an open degree of equipment data that a maintenance personnel can access, and the permission open time length being used to indicate a time length for which the equipment data is open to the maintenance personnel; outputting a fault warning containing the open permission level and the permission open time length.
7. The method of claim 6, wherein, The method further comprises: determining a use peak period and a use trough period of the monitored equipment according to a daily use time of the monitored equipment, the use peak period being used to indicate a time period in which a use frequency of the equipment is higher than a first frequency threshold value, and the use trough period being used to indicate a time period in which the use frequency of the equipment is lower than a second frequency threshold value; outputting a fault warning containing the open permission level, the permission open time length, the use peak period and the use trough period.
8. An apparatus monitoring device, characterized by The device comprises: a data acquisition module configured to determine, according to a risk value determined based on currently collected equipment operation data, that the risk value is higher than a first threshold value, collect a preset number of equipment operation data from the currently collected equipment operation data as a starting point; a risk correction module configured to correct the risk value based on the equipment operation data collected multiple times; and a fault determination module configured to determine that the monitored equipment has a fault when the corrected risk value is higher than a second threshold value. The fault judging module is configured to determine that the monitored device has a fault if the corrected risk value is higher than a second threshold value.
9. An electronic device, comprising: A computer program product comprising a memory, a processor and a computer program stored on the memory and loadable on the processor, the processor being configured to run the computer program to implement the method according to any one of claims 1-7.
10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer executable instructions for causing a computer to perform the method according to any one of claims 1-7.