Intelligent monitoring operation and maintenance method and system based on big data, medium and equipment

Abstract

The invention provides an intelligent monitoring operation and maintenance method and system based on big data, a medium and equipment, and relates to the technical field of equipment management. According to the technical scheme provided by the invention, the feature tag of the target equipment is extracted, the similarity is calculated based on the feature tag, and the historical operation data of the similar equipment is analyzed after the similar equipment is found, so that the short-term abnormal threshold is determined, and the short-term operation data of the target equipment is monitored in the short-term acquisition time window. When the short-term operation data exceeds the short-term abnormal threshold value, the operation and maintenance scheme can be directly generated according to the abnormal reason corresponding to the threshold value. According to the technical scheme, the anomaly threshold is established by using the historical experience of similar equipment, so that the anomaly judgment accuracy is improved; the timeliness of operation and maintenance is improved by monitoring data in a short-term acquisition time window and quickly responding; by establishing the corresponding relation between the short-term exception threshold and the exception reason, accurate matching of the operation and maintenance scheme is realized, so that accurate operation and maintenance of the data center equipment are realized.

Description

Technical Field

[0001] This application relates to the field of equipment management technology, specifically to an intelligent monitoring and maintenance method, system, medium, and equipment based on big data. Background Technology

[0002] With the rapid development of information technology, data centers, as a key infrastructure supporting modern information society, are becoming increasingly important. Data centers are typically characterized by a wide variety of equipment, complex system architecture, and harsh operating environments, while also requiring stable 24 / 7 operation. In this demanding operating environment, the reliability and stability of the equipment directly affect the normal operation of the data center, thus placing extremely high demands on equipment operation and maintenance management.

[0003] Currently, data center equipment operation and maintenance primarily employs a combination of periodic inspections and preventative maintenance. However, in practical applications, this approach often struggles to accurately grasp the actual operating status of equipment due to the dynamic nature of its operation and the unpredictability of fault occurrences. Especially when equipment malfunctions, the lack of in-depth analysis of operational data and accurate identification of the cause of the anomaly easily leads to misjudgments or delayed assessments. Data centers demand extremely high timeliness and accuracy in equipment operation and maintenance; this traditional approach fails to quickly and accurately locate faults and develop targeted solutions, resulting in imprecise maintenance outcomes that cannot meet the stringent stability requirements of data centers. Summary of the Invention

[0004] This application provides a big data-based intelligent monitoring and maintenance method, system, medium, and equipment that can achieve accurate monitoring and maintenance of data center equipment.

[0005] Firstly, this application provides an intelligent monitoring and maintenance method based on big data, the method comprising: Obtain the device parameters of the target device in the data center, and extract the feature tags corresponding to the target device based on the device parameters; Based on the feature tags, a similarity calculation is performed in the device database, and similar devices with a similarity greater than a preset threshold are identified. Analyze the historical operating data of the similar devices to determine the short-term anomaly thresholds corresponding to different causes of anomalies. The short-term anomaly thresholds correspond to short-term collection time windows. Acquire short-term operating data of the target device within the short-term acquisition time window; When the short-term operating data exceeds the short-term anomaly threshold, an operation and maintenance plan corresponding to the target device is generated based on the anomaly cause corresponding to the short-term anomaly threshold.

[0006] By employing the above technical solution, feature tags of the target device are extracted, and similarity is calculated based on these tags. After identifying similar devices, their historical operational data is analyzed to determine short-term anomaly thresholds. Short-term operational data of the target device is then monitored within a short-term data collection window. When short-term operational data exceeds the short-term anomaly threshold, an operation and maintenance plan can be directly generated based on the corresponding anomaly cause. This technical solution improves the accuracy of anomaly judgment by utilizing historical experience from similar devices to establish anomaly thresholds; enhances the timeliness of operation and maintenance by monitoring data within a short-term data collection window and responding quickly; and achieves precise matching of operation and maintenance plans by establishing a correspondence between short-term anomaly thresholds and anomaly causes, thereby enabling accurate operation and maintenance of data center equipment.

[0007] Optionally, the step of analyzing the historical operating data of the similar devices to determine short-term anomaly thresholds corresponding to different causes of anomalies, wherein the short-term anomaly thresholds correspond to short-term data collection time windows, including: Analyze the historical operating data of the similar devices to generate data operating curves for the similar devices; Identify the fluctuation characteristics of the data running curve, and determine the long-term fluctuation level corresponding to the data running curve based on the fluctuation characteristics; Obtain the standard short-term anomaly thresholds and corresponding standard short-term data collection time windows for different anomaly causes; The standard short-term data collection time window and the standard short-term anomaly threshold are weighted according to the preset weight value corresponding to the long-term fluctuation level to obtain the short-term anomaly threshold and the corresponding short-term data collection time window corresponding to different anomaly causes.

[0008] By employing the aforementioned technical solution, historical operating data of similar equipment is analyzed to generate data operation curves, and their fluctuation characteristics are identified to determine long-term fluctuation levels. Then, based on preset weight values ​​for these fluctuation levels, standard short-term data acquisition time windows and standard short-term anomaly thresholds are weighted, ultimately yielding adaptive short-term anomaly thresholds and short-term data acquisition time windows. This technical solution adjusts standard parameters by introducing long-term fluctuation levels, enabling anomaly thresholds and acquisition time windows to adapt to the operating characteristics of different equipment. The weighting process using preset weight values ​​achieves dynamic optimization of threshold judgment, avoiding potential misjudgments caused by fixed thresholds. Furthermore, by combining long-term operating characteristics of the equipment with short-term anomaly judgment, the accuracy of anomaly identification is improved, making anomaly judgment based on this threshold more practical and reliable.

[0009] Optionally, the step of analyzing the historical operating data of the similar devices to generate data operating curves for the similar devices includes: Analyze the historical operating data of multiple similar devices to generate a first data operating curve corresponding to each of the similar devices; The first load operation mode of the target device and the second load operation mode of each of the similar devices are obtained respectively. Based on the degree of difference between the first load operation mode and each of the second load operation modes, the longitudinal displacement compensation of each of the first data operation curves is determined, and the second data operation curve is generated based on the longitudinal displacement compensation. The second data operation curves are integrated to generate the data operation curves of the similar devices.

[0010] By employing the aforementioned technical solution, historical operating data from multiple similar devices are analyzed to generate a first data operating curve. Then, by comparing the differences between the target device's first load operating mode and the similar devices' second load operating modes, a second data operating curve is generated after longitudinal displacement compensation. Finally, these curves are integrated to form the final data operating curve. This technical solution achieves data calibration under different operating conditions by introducing load operating mode difference analysis; it eliminates data deviations caused by load differences between devices through longitudinal displacement compensation; and by integrating multiple calibrated second data operating curves, a more representative data operating curve is obtained, thereby improving the accuracy of subsequent fluctuation characteristic analysis.

[0011] Optionally, identifying the fluctuation characteristics of the data running curve and determining the long-term fluctuation level corresponding to the data running curve based on the fluctuation characteristics includes: Within a preset time interval, calculate the fluctuation amplitude between two adjacent sampling points of the data running curve; The frequency of fluctuations is obtained by counting the number of times the fluctuation amplitude exceeds a preset fluctuation threshold. The fluctuation amplitudes are compared, and the maximum value of the fluctuation amplitude is determined as the maximum fluctuation value. The fluctuation frequency and the maximum fluctuation value are compared with a preset fluctuation characteristic classification standard to determine the target fluctuation frequency range where the fluctuation frequency is located and the target fluctuation value range where the maximum fluctuation value is located. By combining the target fluctuation frequency range and the target fluctuation value range, the long-term fluctuation level corresponding to the data operation curve is obtained.

[0012] By employing the aforementioned technical solution, two key characteristics of the data operation curve—maximum fluctuation value and fluctuation frequency—are identified and compared with a preset fluctuation characteristic grading standard to determine the long-term fluctuation level corresponding to the data operation curve. This technical solution, by simultaneously considering both fluctuation amplitude and fluctuation frequency, provides a more comprehensive assessment of the equipment's operating status; the preset fluctuation characteristic grading standard enables quantitative judgment of fluctuation characteristics, avoiding potential biases from subjective assessments; thus, the final determined long-term fluctuation level accurately reflects the equipment's operating characteristics.

[0013] Optionally, the fluctuation characteristic grading standard includes multiple fluctuation value intervals and multiple fluctuation frequency intervals. The step of comparing the fluctuation frequency and the maximum fluctuation value with the fluctuation characteristic grading standard to determine the target fluctuation frequency interval where the fluctuation frequency falls and the target fluctuation value interval where the maximum fluctuation value falls includes: The maximum fluctuation value is compared with the first interval threshold corresponding to the fluctuation value interval to determine the target fluctuation value interval in which the maximum fluctuation value is located. The fluctuation value interval includes a low fluctuation interval, a medium fluctuation interval, and a high fluctuation interval. The fluctuation frequency is compared with the second interval threshold corresponding to the fluctuation frequency to determine the target fluctuation frequency interval in which the fluctuation frequency is located. The fluctuation frequency interval includes a low frequency interval, a medium frequency interval, and a high frequency interval.

[0014] By adopting the above technical solution, a preset fluctuation characteristic grading standard is established, including fluctuation value range and fluctuation frequency range. The maximum fluctuation value and fluctuation frequency are compared with the first interval threshold and the second interval threshold, respectively, to determine the target fluctuation value range and the target fluctuation frequency range. Finally, the long-term fluctuation level is determined based on the combination of these two intervals. This technical solution improves the accuracy of fluctuation characteristic assessment by subdividing the judgment of fluctuation characteristics into three levels: low, medium, and high. The combination of two-dimensional intervals enables a three-dimensional assessment of equipment operating status. Furthermore, the use of preset interval thresholds for comparison makes the determination of the fluctuation level more objective and accurate.

[0015] Optionally, feature analysis is performed on the device data in the device database to determine the labeling rules for the type label, the operating environment label, and the operating status label; According to the label segmentation rules, the feature information in the device data is converted into corresponding standard feature labels; A structured dataset is constructed based on the correlation between the data of each device and the feature vectors corresponding to the standard feature labels; Calculate the similarity between the feature label and the structured dataset, and identify devices with a similarity greater than a preset threshold as similar devices.

[0016] By adopting the above technical solution, feature labels are subdivided into type labels, operating environment labels, and operating status labels. Big data technology is used to construct a feature index corresponding to device data in the device database, generating a structured dataset containing standard feature labels. Then, similar devices are identified by calculating the similarity between the feature labels and the structured dataset. This technical solution comprehensively describes the operating characteristics of devices through multi-dimensional feature label segmentation; the construction of a feature index using big data technology achieves structured data processing, improving the efficiency of similarity calculation.

[0017] Optionally, when the short-term operational data exceeds the short-term anomaly threshold, generating an operation and maintenance plan for the target device based on the anomaly cause corresponding to the short-term anomaly threshold includes: When the short-term operating data exceeds the short-term anomaly threshold, information on the upstream and downstream related devices of the target device is obtained, and the scope of the fault impact of the target device is determined based on the information on the upstream and downstream related devices. Based on the cause of the anomaly corresponding to the short-term anomaly threshold and the scope of the fault impact, the corresponding operation and maintenance instructions are obtained from the preset operation and maintenance solution library. The operation and maintenance plan corresponding to the target device is generated based on the operation and maintenance operation instructions.

[0018] By adopting the above technical solution, when short-term operational data is detected to exceed the short-term anomaly threshold, information on upstream and downstream related devices of the target device is obtained to determine the scope of the fault impact. Then, based on the anomaly cause and the scope of the fault impact corresponding to the short-term anomaly threshold, operation and maintenance instructions are retrieved from the operation and maintenance solution library to ultimately generate an operation and maintenance solution. This technical solution achieves accurate assessment of the scope of the fault impact by analyzing the correlation between devices; improves the pertinence of the operation and maintenance solution by using the anomaly cause and the scope of impact as dual criteria for matching operation and maintenance instructions; and ensures the standardization and reliability of the operation and maintenance solution by using a pre-set operation and maintenance solution library to provide standardized operation instructions. This achieves comprehensive optimization of the operation and maintenance solution from generation to execution, improving the accuracy and effectiveness of equipment operation and maintenance.

[0019] Secondly, this application provides an intelligent monitoring and maintenance system based on big data, the system comprising: The parameter acquisition module is used to acquire the device parameters of the target device in the data center and extract the feature tags corresponding to the target device based on the device parameters. The big data matching module is used to perform similarity calculation in the device database based on the feature tags, and to determine similar devices with similarity greater than a preset threshold; The analysis module is used to analyze the historical operating data of the similar devices and determine the short-term anomaly thresholds corresponding to different causes of anomalies. The short-term anomaly thresholds correspond to short-term collection time windows. The monitoring module is used to acquire short-term operating data of the target device within the short-term acquisition time window; The execution module is used to generate an operation and maintenance plan for the target device based on the cause of the short-term abnormality when the short-term operating data exceeds the short-term abnormality threshold.

[0020] Thirdly, this application provides a computer storage medium storing a plurality of instructions adapted for loading by a processor and executing any of the methods described above.

[0021] Fourthly, this application provides an electronic device including a processor, a memory, and a transceiver, wherein the memory is used to store instructions, the transceiver is used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to cause the electronic device to perform any of the methods described above.

[0022] In summary, the beneficial effects of the technical solution of this application include: By employing the above technical solution, feature tags of the target device are extracted, and similarity is calculated based on these tags. After identifying similar devices, their historical operational data is analyzed to determine short-term anomaly thresholds. Short-term operational data of the target device is then monitored within a short-term data collection window. When short-term operational data exceeds the short-term anomaly threshold, an operation and maintenance plan can be directly generated based on the corresponding anomaly cause. This technical solution improves the accuracy of anomaly judgment by utilizing historical experience from similar devices to establish anomaly thresholds; enhances the timeliness of operation and maintenance by monitoring data within a short-term data collection window and responding quickly; and achieves precise matching of operation and maintenance plans by establishing a correspondence between short-term anomaly thresholds and anomaly causes, thereby enabling accurate operation and maintenance of data center equipment. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating an intelligent monitoring and maintenance method based on big data, according to an embodiment of this application. Figure 2 This is a schematic diagram of the structure of an intelligent monitoring and maintenance system based on big data, according to an embodiment of this application. Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0024] Explanation of reference numerals in the attached drawings: 300, electronic device; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. Detailed Implementation

[0025] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0026] In the description of the embodiments of this application, words such as "illustrative," "for example," or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "illustrative," "for example," or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design solutions. Rather, the use of words such as "illustrative," "for example," or "for example" is intended to present the relevant concepts in a specific manner.

[0027] In the description of the embodiments of this application, the term "multiple" means two or more. For example, multiple systems means two or more systems, and multiple screen terminals means two or more screen terminals. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature. The terms "comprising," "including," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0028] Please see Figure 1 This document presents a flowchart illustrating a big data-based intelligent monitoring and maintenance method, as provided in an embodiment of this application. This method can be implemented using a computer program, a microcontroller, or run on a big data-based intelligent monitoring and maintenance system based on the von Neumann architecture. The computer program can be integrated into the application or run as a standalone tool application. The specific steps of the big data-based intelligent monitoring and maintenance method are described in detail below.

[0029] S101: Obtain the device parameters of the target device in the data center, and extract the feature tags corresponding to the target device based on the device parameters; In this context, a data center refers to a dedicated location where IT equipment such as computing, storage, and networking are centrally deployed. In the embodiments of this application, it can be understood as a computer room system composed of multiple devices, including servers, storage devices, network devices, and so on.

[0030] In this context, "target device" refers to a specific device within a data center that requires monitoring and maintenance. In this embodiment, it can be understood as a single device entity that currently requires status monitoring, fault warning, and maintenance management. This device can be any type of data center infrastructure equipment, such as a server, network switch, storage device, cooling equipment, or power supply equipment, used to perform specific business functions and collaborate with other devices to ensure the normal operation of the data center. "Device parameters" refers to various data indicators used to describe and characterize device features. In this embodiment, it can be understood as the basic attribute information and operating status indicators of the target device.

[0031] Among them, the feature label refers to the set of identification information obtained after classifying and standardizing the equipment parameters. In this embodiment of the application, it can be understood as a structured feature description including type label, operating environment label and operating status label. The type label reflects the basic classification attributes of the equipment, the operating environment label reflects the working environment characteristics of the equipment, and the operating status label reflects the performance index characteristics of the equipment.

[0032] In practice, the first step is to collect the device parameters of the target equipment through the data center's monitoring system. Since the parameter formats and numerical ranges differ among different devices, these raw parameters need to be standardized and converted into feature tags in a unified format. Specifically, based on the attribute categories of the device parameters, they are mapped into three types of feature tags: type tags, operating environment tags, and operating status tags. The type tag is a device classification identifier generated based on basic attribute information such as device model and manufacturer; the operating environment tag is a working environment feature identifier generated based on environmental parameters such as installation location and ambient temperature; and the operating status tag is a performance indicator feature identifier generated based on operating data such as load rate and energy consumption. This feature tag extraction method converts heterogeneous parameters from different devices into standardized feature descriptions, providing a unified comparison benchmark for subsequent similarity calculations in the device database, thereby improving the accuracy and efficiency of device matching.

[0033] S102: Calculate similarity in the device database based on feature tags and identify similar devices with similarity greater than a preset threshold; In this embodiment of the application, the device database can be understood as a structured data warehouse containing a large amount of historical operating data of devices, device parameter information and corresponding feature tags. The database adopts a distributed storage architecture and establishes a feature index for storing and retrieving device data.

[0034] In this context, "similar devices" refers to historical devices that have a high degree of matching in core feature tags. In this embodiment of the application, it can be understood as other devices in the device database that are significantly similar to the target device in terms of type tags, operating environment tags, and operating status tags.

[0035] Specifically, the feature tags of the target device are compared with the feature tags of all devices stored in the device database, and a weighted cosine similarity algorithm is used to calculate the similarity score. During the calculation, different weight coefficients are assigned to the type tag, operating environment tag, and operating status tag, with the type tag having the highest weight, followed by the operating environment tag, and the operating status tag having the lowest weight. The similarity score obtained through weighted calculation is between 0 and 1, with a score closer to 1 indicating a higher similarity. When the calculated similarity score is greater than a preset threshold, the device is considered a similar device.

[0036] Based on the above embodiments, as an optional implementation method, the method for determining similar devices in step S102 can be implemented in the following way.

[0037] Perform feature analysis on the device data in the device database to determine the labeling rules for the type label, the operating environment label, and the operating status label; According to the label segmentation rules, the feature information in the device data is converted into corresponding standard feature labels; A structured dataset is constructed based on the correlation between the data of each device and the feature vectors corresponding to the standard feature labels; Calculate the similarity between the feature label and the structured dataset, and identify devices with a similarity greater than a preset threshold as similar devices.

[0038] In practical implementation, due to the diverse types of equipment in data centers and the complex and ever-changing operating environments and states, a standardized feature labeling system needs to be established to accurately identify equipment similar to the target equipment. First, feature analysis is performed on equipment data in the equipment database to systematically analyze the feature distribution patterns of equipment across three dimensions: type, operating environment, and operating status. Based on the analysis results, the rules for classifying type labels are determined, such as categorizing equipment according to attributes like function and specifications; the rules for classifying operating environment labels are determined, such as classifying based on environmental parameters like temperature, humidity, and load; and the rules for classifying operating status labels are determined, such as classifying status based on equipment operating parameters and performance indicators.

[0039] After establishing the labeling rules, the feature information in the device data is converted into standard feature labels according to the rules. For example, for type labels, the specific model information of a device can be converted into a standardized device type label; for operating environment labels, specific environmental parameter values ​​can be mapped to corresponding environmental range labels; for operating status labels, the device's operating parameters can be converted into standardized status description labels. This conversion allows the feature information of different devices to be compared under a unified standard.

[0040] To facilitate subsequent similarity calculations, a structured dataset needs to be constructed. Specifically, a feature vector is built for each device data point and its corresponding standard feature label, and the relationships between these feature vectors are established. This structured processing enables the system to quickly retrieve and compare device features.

[0041] When searching for similar devices, the similarity between the target device's feature labels and the feature vectors in the structured dataset is calculated. By calculating the distance or correlation between feature vectors, the degree of similarity between different devices can be quantified. When the calculated similarity is greater than a preset threshold, the corresponding device is identified as a similar device.

[0042] When identifying similar devices, the feature labels of the target devices are also vectorized, converted into feature vectors of the same dimensions as those in the structured dataset. Similarity calculation uses a weighted cosine similarity algorithm, measuring the degree of similarity by calculating the cosine of the angle between feature vectors. During the calculation, the feature vectors are first normalized to eliminate the influence of dimensions. Then, the vector dot product is calculated and divided by the product of the vector magnitudes to obtain the cosine similarity value. Weighting coefficients are introduced into the calculation, assigning different importance weights to feature labels of different dimensions to increase the influence of key features in the similarity calculation. A preset threshold between 0 and 1 is set for the calculated similarity value, typically between 0.7 and 0.9, and can be adjusted according to actual application requirements. Devices with similarity values ​​greater than this threshold are selected and determined as the final set of similar devices.

[0043] S103: Analyze the historical operating data of similar equipment to determine the short-term anomaly thresholds corresponding to different causes of anomalies. The short-term anomaly thresholds correspond to short-term data collection time windows. The short-term anomaly threshold refers to the criterion value used in equipment operation status monitoring to determine whether a short-term anomaly has occurred in the equipment. In this embodiment, it can be understood as the limit on the degree to which the operating parameters of the equipment deviate from the normal operating range within a short period of time. This threshold is determined by analyzing the short-term fluctuation characteristics in the historical operating data of the equipment, and is used to monitor and identify instantaneous anomalies or short-term fluctuations that occur during equipment operation in real time, thereby promptly detecting early signs that may lead to a decline in equipment performance or failure.

[0044] The short-term data acquisition window refers to the continuous time period used to collect and analyze equipment operation data during equipment operation status monitoring. In this embodiment, it can be understood as the time interval for real-time acquisition and short-term trend analysis of equipment operation parameters. This time window captures short-term changes in equipment operation status by setting an appropriate time span, continuously collecting equipment operation parameter data and performing real-time analysis, thereby enabling timely detection and early warning of short-term abnormal equipment conditions.

[0045] In practice, by establishing a correspondence between the characteristics of changes in operating parameters and anomaly types, a quantitative indicator system based on data statistics is formed, making the anomaly identification process more objective and accurate. The specific analysis process begins with setting a short-term data acquisition window, considering factors such as the acquisition characteristics of equipment parameters, anomaly development patterns, and equipment response characteristics to determine an appropriate window length. Within the determined time window, statistical analysis is performed on the operating parameters to extract their time-domain and variation characteristics. Time-domain characteristics reflect the overall distribution of parameters within the time window, requiring the calculation of statistical characteristic values ​​reflecting the central trend and dispersion of the data. Variation characteristics reflect the dynamic changes in parameters, requiring the calculation of parameter change rate indicators, trend indicators, and volatility indicators.

[0046] This study analyzes the variation patterns of parameters within a short data collection window prior to anomalies recorded in historical data. Through analysis of numerous anomaly cases, it identifies unique parameter variation characteristics for different anomaly types. This requires studying the amplitude and temporal characteristics of parameter changes and establishing a mapping relationship between parameter characteristics and anomaly types. Based on statistical analysis methods, it calculates the distribution characteristics of characteristic parameters corresponding to each anomaly type, determining the probability distribution patterns of these characteristic parameters.

[0047] When determining short-term anomaly thresholds, it is necessary to comprehensively consider both the normal fluctuation range and the characteristics of abnormal states of the parameters. By analyzing the fluctuation patterns of the parameters during normal operation, a baseline range for parameter fluctuations is determined. Combined with the parameter characteristics under abnormal states, a threshold boundary that can effectively distinguish between normal and abnormal states is determined.

[0048] Based on the above embodiments, as an optional implementation method, step S103 further includes S201-S204.

[0049] S201: Analyze the historical operating data of similar equipment and generate data operating curves for similar equipment; In the specific implementation process, the historical operating data must first be arranged chronologically to ensure the continuity and integrity of the data over time. Then, statistical calculations are performed on the historical data to obtain statistical values ​​reflecting the characteristics of parameter changes. Based on these statistical values, a data operation curve is constructed, with the horizontal axis representing the time series and the vertical axis representing the operating parameter values. During the construction process, an appropriate time resolution needs to be determined so that the curve reflects both the overall trend of parameter changes and important local features. For long-term operating data, different time scales can be used for analysis to generate operating curves with different time spans.

[0050] Based on the above embodiments, as an optional implementation method, the determination of the long-term fluctuation level in step S201 can be achieved by steps S2011-S2014.

[0051] S2011: Analyze the historical operating data of multiple similar devices and generate the first data operating curve corresponding to each similar device; In this embodiment, the purpose of analyzing the historical operating data of multiple similar devices and generating a first data operating curve is to obtain a benchmark data expression that accurately reflects the operating patterns of each similar device. Since devices may be under different load states and operating environments during actual operation, the operating data of a single device cannot fully reflect its operating characteristics. Therefore, it is necessary to analyze the historical operating data of each similar device separately and generate an independent first data operating curve to provide a basis for subsequent load operating mode analysis and longitudinal displacement compensation.

[0052] In the specific implementation process, the key operating parameter data of each similar device during its historical operation is first obtained. These parameters may include the device's workload, performance indicators, environmental parameters, etc. By analyzing the changing patterns of these parameters in the historical operating data, the characteristics of the device at different operating stages can be identified. Based on these characteristics, a curve fitting algorithm is used to construct a first data operation curve reflecting the trend of parameter changes. This curve is plotted with time on the horizontal axis and operating parameter values ​​on the vertical axis.

[0053] For each similar device, a corresponding initial data operation curve needs to be generated independently. These curves can intuitively show the parameter change characteristics of each device during long-term operation, providing an important basis for subsequent analysis of the device's load operation mode. The initial data operation curves generated in this way retain the operating characteristics of each similar device, which helps to more accurately perform load mode difference analysis and curve compensation.

[0054] S2012: Obtain the first load operation mode of the target device and the second load operation mode of each similar device respectively; In the specific implementation process, it is first necessary to analyze the operating data of the target equipment. By calculating characteristic parameters such as the average load rate, load fluctuation range, and load duration of the equipment in different time periods, the typical load pattern of the target equipment can be identified. This load pattern reflects the business load characteristics of the equipment during actual operation and is denoted as the first load operation mode. The same method needs to be applied to various similar devices to calculate their load characteristic parameters and obtain the second load operation mode corresponding to each similar device.

[0055] S2013: Based on the degree of difference between the first load operation mode and each of the second load operation modes, determine the longitudinal displacement compensation of each first data operation curve, and generate the second data operation curve based on the longitudinal displacement compensation; In the specific implementation process, the difference between the first load operating mode of the target device and the second load operating modes of each similar device is first calculated. This difference can be quantified by comparing load characteristic parameters, such as calculating the difference in average load rate or the ratio of load fluctuation ranges. Based on the calculated difference, the required longitudinal displacement compensation amount for the first data operating curve of each similar device is determined. The magnitude of the compensation amount is proportional to the difference in load modes; the greater the difference, the greater the required compensation. By performing a corresponding longitudinal shift on the first data operating curve, the second data operating curve after load compensation can be obtained. This compensated curve eliminates the influence of load differences, making the operating data of different devices more comparable.

[0056] S2014: Integrate the second data operation curves to generate data operation curves for similar devices.

[0057] In the specific implementation process, it is first necessary to ensure that all second data running curves have the same time sampling point. For each time point, the parameter values ​​of all second data running curves at that time point are collected to form a parameter value set. For each parameter value set, its arithmetic mean is calculated as the baseline value of the integrated curve at that time point. At the same time, the standard deviation of the parameter value set is calculated to characterize the degree of parameter fluctuation at that time point.

[0058] To ensure the integrated curve better reflects the parameter variation trend, a piecewise fitting method is used to process the baseline value sequence. Specifically, the baseline value sequence is first segmented, with the time span of each segment determined based on the parameter variation characteristics. Then, a polynomial fitting method is applied to each segment. The order of the polynomial is chosen based on the data complexity, typically a 3-5 order polynomial to ensure a balance between fitting accuracy and smoothness. At the junctions of segments, the polynomial coefficients are adjusted to ensure the continuity and smoothness of the curve. The standard deviation sequence is also processed using a piecewise fitting method to obtain an envelope reflecting the degree of parameter fluctuation. Combining the baseline curve with the envelope yields a complete data flow curve that includes the parameter variation trend and fluctuation range.

[0059] S202: Identify the fluctuation characteristics of the data operation curve and determine the long-term fluctuation level corresponding to the data operation curve based on the fluctuation characteristics; Among them, the fluctuation characteristic is a characteristic quantity that reflects the dynamic change law of the equipment operating parameters deviating from their steady-state value. In the embodiments of this application, it can be understood as the degree of fluctuation of the equipment operating parameters relative to their mean or expected value within a certain time range.

[0060] The long-term fluctuation level is an evaluation index that quantifies and classifies the degree of fluctuation of equipment operating parameters over a long period of time. In this embodiment, it can be understood as classifying the degree of fluctuation into different levels based on the fluctuation amplitude, frequency, and duration of the parameter values ​​through statistical analysis of equipment operating parameters over a long time scale. The long-term fluctuation level is mainly used to assess the long-term stability of equipment operation, identify abnormal fluctuation states during equipment operation, and provide a basis for equipment maintenance decisions and operation optimization.

[0061] In practice, the system first uses data analysis algorithms to process the data operation curves. This algorithm can identify the curve's changing trends over different time periods, including its rising, falling, and oscillating characteristics. Simultaneously, the algorithm calculates the amplitude of the curve's fluctuations and analyzes the degree to which the curve deviates from a stable state. These identified fluctuation characteristics constitute the basic data for evaluating the equipment's operating status. After obtaining the fluctuation characteristics, the system performs a comprehensive analysis based on pre-set evaluation criteria. The evaluation process considers multiple dimensions such as the intensity and duration of fluctuations, transforming these characteristics into quantifiable indicators through specific calculation methods. Based on these quantifiable indicators, the system ultimately determines the long-term fluctuation level corresponding to the data operation curve.

[0062] Based on the above embodiments, as an optional implementation method, the determination of the long-term fluctuation level in step S202 can be achieved by steps S2021-S2022.

[0063] S2021: Within a preset time interval, calculate the fluctuation amplitude between two adjacent sampling points of the data running curve; Specifically, the system first determines the start and end times of the time interval, and then samples the data curve within this interval according to a preset sampling frequency. For the acquired sampled data, the system calculates the change in operating parameters between two adjacent sampling points, i.e., the fluctuation amplitude. This calculation method takes into account the continuous change characteristics of equipment operating parameters over time. By analyzing the parameter differences between adjacent time points, it can accurately capture various fluctuations during equipment operation.

[0064] S2022: Count the number of times the fluctuation amplitude is greater than the preset fluctuation judgment threshold to obtain the fluctuation frequency; After acquiring the fluctuation amplitude data, a quantitative analysis of the frequency of fluctuations is needed to accurately assess the instability of equipment operation. Since normal parameter fluctuations occur during equipment operation, setting a preset fluctuation threshold can filter out fluctuations within the normal range, focusing only on significant fluctuations that may affect stable equipment operation. The system compares each fluctuation value in the acquired fluctuation amplitude sequence with the preset fluctuation threshold; when the fluctuation amplitude exceeds the preset threshold, it is considered a significant fluctuation. By statistically analyzing the cumulative number of significant fluctuations occurring within a preset time interval, a fluctuation frequency index reflecting the stability of equipment operation is obtained.

[0065] S2023: Compare the fluctuation amplitudes and determine the maximum value of the fluctuation amplitudes as the maximum fluctuation value; When assessing equipment operational stability, in addition to understanding the frequency of fluctuations, the intensity of those fluctuations is also a crucial evaluation dimension. Fluctuation intensity reflects the degree to which equipment operating parameters deviate from normal values; larger fluctuations often indicate potential anomalies or malfunctions. Therefore, the system needs to systematically analyze and compare all fluctuation amplitudes within a preset time interval. By iterating and comparing the fluctuation amplitude sequence, the maximum fluctuation value is determined as the maximum fluctuation value. This value directly reflects the most severe fluctuation experienced by the equipment during the observation period. Obtaining the maximum fluctuation value provides another important quantitative indicator for assessing equipment operational stability. Combined with the statistical results of fluctuation frequency, it allows for a comprehensive characterization of the equipment's operational fluctuation features.

[0066] S2024: Compare the fluctuation frequency and the maximum fluctuation value with the preset fluctuation characteristic classification standard to determine the target fluctuation frequency range where the fluctuation frequency is located and the target fluctuation value range where the maximum fluctuation value is located. In practice, the system first performs a tiered assessment of fluctuation frequencies. By comparing the statistically obtained fluctuation frequencies with the fluctuation frequency thresholds in the fluctuation characteristic grading standard, it determines whether the fluctuation frequency belongs to the low-frequency, medium-frequency, or high-frequency range. This frequency grading reflects the intensity of significant fluctuations during equipment operation. Simultaneously, the system compares the acquired maximum fluctuation value with the fluctuation value thresholds in the fluctuation characteristic grading standard to determine whether the maximum fluctuation value belongs to the low-fluctuation, medium-fluctuation, or high-fluctuation range. This fluctuation value grading reflects the degree to which equipment operating parameters deviate from the normal range.

[0067] During the grading process, the system uses preset judgment rules to divide intervals. For example, when the fluctuation frequency is below the first frequency threshold, it is judged as a low-frequency interval; when the fluctuation frequency is above the second frequency threshold, it is judged as a high-frequency interval; and when it is in between, it is judged as a medium-frequency interval. The determination of fluctuation value intervals also uses a similar threshold comparison method. This judgment method based on multi-level thresholds enables the system to accurately divide the level range of fluctuation characteristics.

[0068] S2025: Combine the target fluctuation frequency range with the target fluctuation value range to obtain the long-term fluctuation level corresponding to the data operation curve.

[0069] In practice, the preset volatility characteristic grading standards typically include multiple levels, each corresponding to a specific range of maximum volatility values ​​and frequency ranges. For example, volatility characteristics can be divided into four levels: stable, slight volatility, moderate volatility, and severe volatility. When the maximum volatility value is less than a first threshold and the volatility frequency is less than a first frequency threshold, it is classified as a stable level; when the maximum volatility value is between the first and second thresholds and the volatility frequency is between the first and second frequency thresholds, it is classified as a slight volatility level; when the maximum volatility value is between the second and third thresholds and the volatility frequency is between the second and third frequency thresholds, it is classified as a moderate volatility level; and when the maximum volatility value is greater than the third threshold or the volatility frequency is greater than the third frequency threshold, it is classified as a severe volatility level. By comparing the identified maximum volatility value and volatility frequency with these preset ranges, the long-term volatility level corresponding to the data curve can be determined.

[0070] Based on the above embodiments, as an optional implementation, the fluctuation characteristic grading standard includes multiple fluctuation value ranges and multiple fluctuation frequency ranges, and step S2024 further includes: The maximum fluctuation value is compared with the first interval threshold corresponding to the fluctuation value interval to determine the target fluctuation value interval in which the maximum fluctuation value is located. The fluctuation value interval includes a low fluctuation interval, a medium fluctuation interval, and a high fluctuation interval. The fluctuation frequency is compared with the second interval threshold corresponding to the fluctuation frequency to determine the target fluctuation frequency interval in which the fluctuation frequency is located. The fluctuation frequency interval includes a low frequency interval, a medium frequency interval, and a high frequency interval.

[0071] In the specific implementation process, the first step is to obtain a preset fluctuation characteristic classification standard. This standard divides the fluctuation value range into low fluctuation range, medium fluctuation range, and high fluctuation range, and simultaneously divides the fluctuation frequency range into low frequency range, medium frequency range, and high frequency range. The division of fluctuation value ranges is achieved through a first interval threshold, for example, by setting first interval thresholds A1 and A2. When the maximum fluctuation value is less than A1, it belongs to the low fluctuation range; when the maximum fluctuation value is greater than or equal to A1 and less than A2, it belongs to the medium fluctuation range; and when the maximum fluctuation value is greater than or equal to A2, it belongs to the high fluctuation range. Similarly, the division of fluctuation frequency ranges is achieved through a second interval threshold, for example, by setting second interval thresholds B1 and B2. When the fluctuation frequency is less than B1, it belongs to the low frequency range; when the fluctuation frequency is greater than or equal to B1 and less than B2, it belongs to the medium frequency range; and when the fluctuation frequency is greater than or equal to B2, it belongs to the high frequency range.

[0072] Next, the identified maximum fluctuation value is compared with the first interval threshold to determine its target fluctuation value interval. Simultaneously, the fluctuation frequency is compared with the second interval threshold to determine its target fluctuation frequency interval. This process of determining these two target intervals essentially quantifies continuous fluctuation characteristics into discrete level attributes, facilitating subsequent combination and judgment.

[0073] Finally, based on the combination of the target fluctuation value range and the target fluctuation frequency range, the long-term fluctuation level corresponding to the data operation curve is determined. For example, when the target fluctuation value range is a low fluctuation range and the target fluctuation frequency range is a low frequency range, it can be judged as a stable operation level; when the target fluctuation value range is a medium fluctuation range or the target fluctuation frequency range is a medium frequency range, it can be judged as a medium fluctuation level; when the target fluctuation value range is a high fluctuation range or the target fluctuation frequency range is a high frequency range, it can be judged as a severe fluctuation level. This determination method based on range combinations fully considers the combined influence of fluctuation amplitude and frequency.

[0074] S203: Obtain the standard short-term anomaly thresholds and corresponding standard short-term acquisition time windows for different anomaly causes; In this embodiment, to accurately identify and diagnose various anomalies occurring during equipment operation, it is necessary to obtain standard short-term anomaly thresholds and standard short-term acquisition time windows corresponding to different anomaly causes. This is because different types of anomalies differ in their manifestation, development speed, and duration, requiring the use of matching anomaly thresholds and time windows for detection and judgment. By establishing a complete anomaly detection reference system, the accuracy and timeliness of anomaly detection can be improved.

[0075] In the specific implementation process, the first step is to establish an anomaly cause database, containing various reasons that may lead to equipment malfunctions. For each anomaly cause, based on historical operating data and professional experience, a corresponding standard short-term anomaly threshold is determined. These thresholds reflect the critical change level of operating parameters under specific anomaly conditions. Different types of anomalies may exhibit different parameter change characteristics; therefore, corresponding threshold standards need to be set for different types of anomalies. The setting of standard short-term anomaly thresholds needs to consider the normal fluctuation range of parameters, ensuring that it can effectively identify anomalies without being overly sensitive and causing false alarms. These thresholds can be fixed values ​​or dynamically adjusted values ​​based on operating conditions to adapt to the anomaly detection needs under different operating conditions.

[0076] Simultaneously, for each cause of anomaly, a corresponding standard short-term acquisition time window needs to be determined. The length of the time window directly affects the sensitivity and reliability of anomaly detection. The selection of the time window needs to consider the temporal characteristics of anomaly development to ensure accurate capture of parameter changes at the time of anomaly occurrence. Setting the standard short-term acquisition time window requires a balance between the requirements of detection timeliness and data reliability. Determining the window length also needs to consider the operating cycle characteristics of the equipment to ensure it can encompass the complete parameter change process. In practical applications, different time window lengths can be set according to the anomaly development speed and duration characteristics to achieve effective monitoring of different types of anomalies.

[0077] S204: The standard short-term data collection time window and the standard short-term anomaly threshold are weighted according to the preset weight value corresponding to the long-term fluctuation level to obtain the short-term anomaly threshold and the corresponding short-term data collection time window corresponding to different anomaly causes.

[0078] In the specific implementation process, the first step is to determine the preset weight values ​​corresponding to different long-term fluctuation levels. A higher long-term fluctuation level indicates greater volatility in equipment operation, requiring a larger weight value to appropriately relax the anomaly detection conditions. Conversely, when the long-term fluctuation level is low, the equipment operation is more stable, allowing for a smaller weight value to improve the sensitivity of anomaly detection. The setting of preset weight values ​​needs to consider the relationship between the long-term fluctuation level and the anomaly detection sensitivity, ensuring that the weight values ​​reasonably reflect the degree of influence of the fluctuation level on anomaly detection.

[0079] For the weighted processing of the standard short-term acquisition time window, the actual short-term acquisition time window is obtained by multiplying the standard short-term acquisition time window by a corresponding preset weight value. When the long-term fluctuation level is high, a larger weight value will appropriately extend the acquisition time window to obtain more data samples and improve the reliability of the judgment. When the long-term fluctuation level is low, a smaller weight value will shorten the acquisition time window to improve the timeliness of anomaly detection. This time window adjustment based on long-term fluctuation characteristics allows the time scale of anomaly detection to better adapt to the actual operating status of the equipment.

[0080] The weighted processing of the standard short-term anomaly threshold is also done by multiplying the standard short-term anomaly threshold by the corresponding preset weight value to obtain the short-term anomaly threshold for practical application. When the long-term fluctuation level is high, a larger weight value will increase the anomaly threshold, preventing normal fluctuations from being misjudged as anomalies. When the long-term fluctuation level is low, a smaller weight value will decrease the anomaly threshold, increasing the sensitivity of anomaly detection.

[0081] S104: Acquire short-term operational data of the target device within a short-term acquisition time window; Short-term operational data refers to equipment operating parameter data collected within a short-term acquisition window. This data can promptly reflect the current operating status and parameter change trends of the equipment, serving as the foundational data for short-term anomaly detection. The acquisition time span of short-term operational data is determined by the short-term acquisition window, and its length has been weighted and adjusted based on long-term fluctuation levels, thus better adapting to the actual operating conditions of the equipment.

[0082] In practice, the data acquisition time range is set according to the predetermined short-term acquisition time window. The acquisition system continuously collects operating parameter data of the target device according to the preset sampling frequency. The collected parameters include, but are not limited to, key indicators such as the device's operating parameters, performance parameters, and environmental parameters. This data is collected in real time by sensors on the device and, after data preprocessing, forms a standardized short-term operating dataset. Data preprocessing includes data cleaning, standardization, and format unification to ensure that the collected data has good quality and usability.

[0083] The process of acquiring short-term operational data is continuous; the system constantly updates the data cache to maintain data timeliness. A sliding time window approach ensures that the acquired short-term operational data always covers the most recent time period.

[0084] S105: When short-term operating data exceeds the short-term anomaly threshold, generate the corresponding operation and maintenance plan for the target device based on the anomaly cause corresponding to the short-term anomaly threshold.

[0085] In practice, short-term operational data is compared with corresponding short-term anomaly thresholds in real time. When short-term operational data exceeds the threshold, the anomaly handling mechanism is immediately triggered. The specific cause of the current anomaly is determined by querying the corresponding relationships in the anomaly cause database. Because the short-term anomaly thresholds have been weighted based on long-term fluctuation levels, this judgment has high reliability. After determining the cause of the anomaly, a corresponding operation and maintenance plan is automatically generated based on a preset operation and maintenance knowledge base and the specific characteristics of the current anomaly. The operation and maintenance plan includes complete handling guidance such as the priority of anomaly handling, specific handling steps, required tools and materials, and precautions. When generating the operation and maintenance plan, the system considers multiple factors such as the equipment's operating status, the availability of maintenance resources, and the urgency of the maintenance work to ensure the feasibility and practicality of the operation and maintenance plan.

[0086] Based on the above embodiments, as an optional implementation, step S105 further includes: when the short-term operating data exceeds the short-term abnormal threshold, obtaining the upstream and downstream related equipment information of the target equipment, and determining the fault impact range of the target equipment based on the upstream and downstream related equipment information; Based on the cause of the anomaly and the scope of the fault impact corresponding to the short-term anomaly threshold, the corresponding operation and maintenance instructions are obtained from the preset operation and maintenance solution library. Generate the corresponding operation and maintenance plan for the target device based on the operation and maintenance operation instructions.

[0087] In this embodiment, when short-term operational data is detected to exceed the short-term anomaly threshold, a comprehensive assessment of the impact of the target device's abnormal state on the entire system is required, and a suitable operation and maintenance plan should be formulated accordingly. This method not only considers the abnormal situation of the target device itself, but also focuses on the chain reactions that the anomaly may cause, thereby ensuring the comprehensiveness and effectiveness of the operation and maintenance plan.

[0088] In practice, upon discovering that short-term operational data exceeds the short-term anomaly threshold, information on upstream and downstream related equipment of the target device is immediately obtained. This information includes upstream supplying equipment and downstream receiving equipment directly connected to the target device, as well as their process relationships and material transfer relationships. By analyzing these relationships, the scope of the target device's failure can be determined. Determining the scope of the failure's impact requires considering multiple factors, such as the severity of the anomaly, its propagation path, and potential risks, to accurately assess the potential cascading effects of the anomaly.

[0089] While determining the scope of the fault's impact, based on the cause of the fault corresponding to the short-term anomaly threshold and the determined scope of the fault's impact, matching operation instructions are retrieved from a pre-set operation and maintenance solution library. This library stores standard handling procedures and operation instructions for various anomaly situations; these instructions have been validated in practice and possess strong operability and effectiveness. Based on the specific characteristics and scope of impact of the current anomaly, the most suitable combination of operation and maintenance instructions is selected.

[0090] After obtaining the relevant operation and maintenance instructions, these instructions will be integrated into a complete operation and maintenance plan based on the actual situation of the target equipment. The generation process of the operation and maintenance plan considers multiple factors, including the equipment's operating status, the configuration of maintenance resources, the priority of operations, and safety precautions. The generated operation and maintenance plan not only includes specific operating steps but also contingency plans to deal with possible chain reactions, as well as necessary adjustment suggestions for upstream and downstream equipment.

[0091] Please see Figure 2 This illustration shows a schematic diagram of the structure of a big data-based intelligent monitoring and maintenance system provided in an exemplary embodiment of this application. The system can be implemented through software, hardware, or a combination of both, forming all or part of a larger system. The big data-based intelligent monitoring and maintenance system includes: The parameter acquisition module is used to acquire the device parameters of the target device in the data center and extract the feature tags corresponding to the target device based on the device parameters. The big data matching module is used to perform similarity calculations in the device database based on feature tags and to identify similar devices with a similarity greater than a preset threshold. The analysis module is used to analyze the historical operating data of similar devices to determine the short-term anomaly thresholds corresponding to different causes of anomalies. The short-term anomaly thresholds correspond to short-term data collection time windows. The monitoring module is used to acquire short-term operating data of the target device within a short-term data collection window; The execution module is used to generate an operation and maintenance plan for the target device based on the cause of the short-term abnormality when the short-term operating data exceeds the short-term abnormality threshold.

[0092] Based on the above embodiments, as an optional embodiment, the big data matching module is also used for Feature analysis is performed on the device data in the device database to determine the labeling rules for the type label, the operating environment label, and the operating status label; according to the labeling rules, the feature information in the device data is converted into corresponding standard feature labels; a structured dataset is constructed based on the correlation between the feature vectors corresponding to each device data and the standard feature labels; the similarity between the feature labels and the structured dataset is calculated, and devices with a similarity greater than a preset threshold are identified as similar devices.

[0093] Based on the above embodiments, as an optional embodiment, the analysis module is further configured to analyze the historical operating data of similar devices to generate data operating curves for similar devices; identify the fluctuation characteristics of the data operating curves and determine the long-term fluctuation level corresponding to the data operating curves based on the fluctuation characteristics; obtain the standard short-term anomaly thresholds and corresponding standard short-term acquisition time windows corresponding to different anomaly causes; and perform weighted processing on the standard short-term acquisition time windows and standard short-term anomaly thresholds according to the preset weight values ​​corresponding to the long-term fluctuation levels to obtain the short-term anomaly thresholds and corresponding short-term acquisition time windows corresponding to different anomaly causes.

[0094] Based on the above embodiments, as an optional embodiment, the analysis module is further configured to analyze the historical operating data of multiple similar devices to generate a first data operating curve corresponding to each similar device; obtain the first load operating mode of the target device and the second load operating mode of each similar device; determine the longitudinal displacement compensation of each first data operating curve according to the degree of difference between the first load operating mode and each second load operating mode, and generate a second data operating curve according to the longitudinal displacement compensation; and integrate each second data operating curve to generate the data operating curve of each similar device.

[0095] Based on the above embodiments, as an optional embodiment, the analysis module is further configured to calculate the fluctuation amplitude between two adjacent sampling points of the data running curve within a preset time interval; count the number of times the fluctuation amplitude is greater than a preset fluctuation judgment threshold to obtain the fluctuation frequency; compare each fluctuation amplitude and determine the maximum value of the fluctuation amplitude as the maximum fluctuation value; compare the fluctuation frequency and the maximum fluctuation value with a preset fluctuation feature grading standard to determine the target fluctuation frequency interval where the fluctuation frequency is located and the target fluctuation value interval where the maximum fluctuation value is located; combine the target fluctuation frequency interval and the target fluctuation value interval to obtain the long-term fluctuation level corresponding to the data running curve.

[0096] Based on the above embodiments, as an optional embodiment, the analysis module is also used for The maximum fluctuation value is compared with the first interval threshold corresponding to the fluctuation value interval to determine the target fluctuation value interval in which the maximum fluctuation value is located. The fluctuation value interval includes a low fluctuation interval, a medium fluctuation interval, and a high fluctuation interval. The fluctuation frequency is compared with the second interval threshold corresponding to the fluctuation frequency to determine the target fluctuation frequency interval in which the fluctuation frequency is located. The fluctuation frequency interval includes a low frequency interval, a medium frequency interval, and a high frequency interval.

[0097] Based on the above embodiments, as an optional embodiment, the execution module is further configured to, when the short-term operating data exceeds the short-term anomaly threshold, obtain the upstream and downstream related equipment information of the target device, and determine the scope of the fault impact of the target device based on the upstream and downstream related equipment information; obtain the corresponding operation and maintenance operation instructions from the preset operation and maintenance solution library based on the anomaly cause and the scope of the fault impact corresponding to the short-term anomaly threshold; and generate the operation and maintenance solution corresponding to the target device based on the operation and maintenance operation instructions.

[0098] This application also provides a computer storage medium that can store multiple instructions. The instructions are adapted to be loaded and executed by a processor as described in the above embodiments for the intelligent monitoring and maintenance method based on big data. For the specific execution process, please refer to the detailed description of the embodiments, which will not be repeated here.

[0099] Please see Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 3 As shown, the electronic device 300 may include: at least one processor 301, at least one network interface 304, user interface 303, memory 305, and at least one communication bus 302.

[0100] The communication bus 302 is used to enable communication between these components.

[0101] The user interface 303 may include a display screen and a camera.

[0102] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0103] The processor 301 may include one or more processing cores. The processor 301 connects to various parts of the server using various interfaces and lines, and performs various server functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in memory 305, and by calling data stored in memory 305. Optionally, the processor 301 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 301 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content required for display; and the modem handles wireless communication. It is understood that the modem may also not be integrated into the processor 301 and may be implemented as a separate chip.

[0104] The memory 305 may include random access memory (RAM) or read-only memory. Optionally, the memory 305 may include a non-transitory computer-readable storage medium. The memory 305 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 305 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory 305 may also be at least one storage device located remotely from the aforementioned processor 301. Figure 3 As shown, the memory 305, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and an application program for an intelligent monitoring and maintenance method based on big data.

[0105] exist Figure 3In the electronic device 300 shown, the user interface 303 is mainly used to provide an input interface for the user and to obtain the user input data; while the processor 301 can be used to call an application program stored in the memory 305 that is a big data-based intelligent monitoring and maintenance method. When executed by one or more processors, the electronic device executes one or more methods as described in the above embodiments.

[0106] An electronic device readable storage medium stores instructions that, when executed by one or more processors, cause the electronic device to perform one or more methods as described in the above embodiments.

[0107] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0108] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0109] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the shown or discussed mutual couplings or direct couplings or communication connections may be through some service interfaces; indirect couplings or communication connections between apparatuses or units may be electrical or other forms.

[0110] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0111] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0112] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0113] The above are merely exemplary embodiments of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. Those skilled in the art will readily conceive of other embodiments of this disclosure upon considering the specification and the disclosure of practical truths. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described in this disclosure.

Claims

1. A big data-based intelligent monitoring and maintenance method, characterized in that, The method includes: Obtain the device parameters of the target device in the data center, and extract the feature tags corresponding to the target device based on the device parameters; Based on the feature tags, a similarity calculation is performed in the device database, and similar devices with a similarity greater than a preset threshold are identified. Analyze the historical operating data of the similar devices to determine the short-term anomaly thresholds corresponding to different causes of anomalies. The short-term anomaly thresholds correspond to short-term collection time windows. Acquire short-term operating data of the target device within the short-term acquisition time window; When the short-term operating data exceeds the short-term anomaly threshold, an operation and maintenance plan corresponding to the target device is generated based on the anomaly cause corresponding to the short-term anomaly threshold.

2. The method according to claim 1, characterized in that, The analysis of historical operating data of the similar devices determines short-term anomaly thresholds corresponding to different causes of anomalies. These short-term anomaly thresholds correspond to short-term data collection windows, including: Analyze the historical operating data of the similar devices to generate data operating curves for the similar devices; Identify the fluctuation characteristics of the data running curve, and determine the long-term fluctuation level corresponding to the data running curve based on the fluctuation characteristics; Obtain the standard short-term anomaly thresholds and corresponding standard short-term data collection time windows for different anomaly causes; The standard short-term data collection time window and the standard short-term anomaly threshold are weighted according to the preset weight value corresponding to the long-term fluctuation level to obtain the short-term anomaly threshold and the corresponding short-term data collection time window corresponding to different anomaly causes.

3. The method according to claim 2, characterized in that, The step of analyzing the historical operating data of the similar devices to generate data operating curves for the similar devices includes: Analyze the historical operating data of multiple similar devices to generate a first data operating curve corresponding to each of the similar devices; The first load operation mode of the target device and the second load operation mode of each of the similar devices are obtained respectively. Based on the degree of difference between the first load operation mode and each of the second load operation modes, the longitudinal displacement compensation of each of the first data operation curves is determined, and the second data operation curve is generated based on the longitudinal displacement compensation. The second data operation curves are integrated to generate the data operation curves of the similar devices.

4. The method according to claim 2, characterized in that, The step of identifying the fluctuation characteristics of the data running curve and determining the long-term fluctuation level corresponding to the data running curve based on the fluctuation characteristics includes: Within a preset time interval, calculate the fluctuation amplitude between two adjacent sampling points of the data running curve; The frequency of fluctuations is obtained by counting the number of times the fluctuation amplitude exceeds a preset fluctuation threshold. The fluctuation amplitudes are compared, and the maximum value of the fluctuation amplitude is determined as the maximum fluctuation value. The fluctuation frequency and the maximum fluctuation value are compared with a preset fluctuation characteristic classification standard to determine the target fluctuation frequency range where the fluctuation frequency is located and the target fluctuation value range where the maximum fluctuation value is located. By combining the target fluctuation frequency range and the target fluctuation value range, the long-term fluctuation level corresponding to the data operation curve is obtained.

5. The method according to claim 4, characterized in that, The fluctuation characteristic grading standard includes multiple fluctuation value intervals and multiple fluctuation frequency intervals. The step of comparing the fluctuation frequency and the maximum fluctuation value with the fluctuation characteristic grading standard to determine the target fluctuation frequency interval and the target fluctuation value interval where the maximum fluctuation value is located includes: The maximum fluctuation value is compared with the first interval threshold corresponding to the fluctuation value interval to determine the target fluctuation value interval in which the maximum fluctuation value is located. The fluctuation value interval includes a low fluctuation interval, a medium fluctuation interval, and a high fluctuation interval. The fluctuation frequency is compared with the second interval threshold corresponding to the fluctuation frequency to determine the target fluctuation frequency interval in which the fluctuation frequency is located. The fluctuation frequency interval includes a low frequency interval, a medium frequency interval, and a high frequency interval.

6. The method according to claim 1, characterized in that, The feature tags include type tags, operating environment tags, and operating status tags. The step of calculating similarity in the device database based on the feature tags and identifying similar devices with a similarity greater than a preset threshold includes: Perform feature analysis on the device data in the device database to determine the labeling rules for the type label, the operating environment label, and the operating status label; According to the label segmentation rules, the feature information in the device data is converted into corresponding standard feature labels; A structured dataset is constructed based on the correlation between the data of each device and the feature vectors corresponding to the standard feature labels; Calculate the similarity between the feature label and the structured dataset, and identify devices with a similarity greater than a preset threshold as similar devices.

7. The method according to claim 1, characterized in that, When the short-term operational data exceeds the short-term anomaly threshold, a maintenance plan for the target device is generated based on the anomaly cause corresponding to the short-term anomaly threshold, including: When the short-term operating data exceeds the short-term anomaly threshold, information on the upstream and downstream related devices of the target device is obtained, and the scope of the fault impact of the target device is determined based on the information on the upstream and downstream related devices. Based on the cause of the anomaly corresponding to the short-term anomaly threshold and the scope of the fault impact, the corresponding operation and maintenance instructions are obtained from the preset operation and maintenance solution library. The operation and maintenance plan corresponding to the target device is generated based on the operation and maintenance operation instructions.

8. A big data-based intelligent monitoring and maintenance system, characterized in that, The system includes: The parameter acquisition module is used to acquire the device parameters of the target device in the data center and extract the feature tags corresponding to the target device based on the device parameters. The big data matching module is used to perform similarity calculation in the device database based on the feature tags, and to determine similar devices with similarity greater than a preset threshold; The analysis module is used to analyze the historical operating data of the similar devices and determine the short-term anomaly thresholds corresponding to different causes of anomalies. The short-term anomaly thresholds correspond to short-term collection time windows. The monitoring module is used to acquire short-term operating data of the target device within the short-term acquisition time window; The execution module is used to generate an operation and maintenance plan for the target device based on the cause of the short-term abnormality when the short-term operating data exceeds the short-term abnormality threshold.

9. A computer storage medium, characterized in that, The computer storage medium stores a plurality of instructions, which are adapted to be loaded by a processor and executed as described in any one of claims 1 to 7.

10. An electronic device, characterized in that, The device includes a processor, a memory, and a transceiver, wherein the memory is used to store instructions, the transceiver is used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to cause the electronic device to perform the method as described in any one of claims 1 to 7.

Images